CT machine room environment intelligent adjusting method and system based on Internet of Things

Abstract

The invention relates to the field of environment adjustment, and discloses a CT machine room environment intelligent adjustment method and system based on the Internet of Things, and the method comprises the steps: positioning at least one to-be-adjusted environment node in a CT machine room based on an Internet of Things sensing network, and obtaining a to-be-adjusted point information set; wherein the to-be-adjusted node information comprises a to-be-adjusted node type, a to-be-adjusted node position and a local environment topological graph; the local environment topological graph represents a topological structure of a local CT machine room environment where the to-be-adjusted node is located; collecting real-time environment parameter information; wherein the real-time environment parameter information comprises at least one of temperature and humidity sensor node data and power load sensor node data. The CT machine room environment is monitored in real time through the Internet of Things sensing network, the system can accurately analyze and adjust the environment according to real-time environment parameter information, normal operation of equipment in the machine room is guaranteed, human intervention is reduced, and the automation level is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental regulation, and in particular to an Internet of Things-based intelligent regulation method and system for a CT machine room environment. Background Art

[0002] Traditional methods often rely on manual intervention or fixed adjustment rules, lacking real-time monitoring and adaptive adjustment capabilities, resulting in inaccurate environmental adjustment and an inability to effectively respond to complex environmental changes. Traditional methods are usually unable to identify abnormal nodes in the environment in real time, and are unable to automatically analyze and locate abnormal areas, resulting in a slow response to environmental problems and an inability to effectively merge related abnormal areas, resulting in poor adjustment effects. In traditional methods, environmental adjustment and energy storage management are often separated, and the lack of accurate analysis of energy storage needs may lead to energy waste. For example, energy storage equipment may be over-adjusted or fail to expand in time, wasting precious energy resources. Traditional methods are usually only adjusted based on a single parameter such as temperature, humidity or load, and lack the ability to comprehensively consider multiple environmental factors (such as temperature, humidity, load, energy storage demand, etc.). Therefore, they cannot flexibly respond to environmental changes, and the adaptability of the adjustment strategy is insufficient. Traditional methods often lack the ability to dynamically adjust environmental adjustments and lack the accurate collection and rapid processing of real-time data, resulting in a slow response to environmental changes and the inability to adjust to the optimal state in a timely manner. Summary of the Invention

[0003] To achieve the above objectives, the present invention provides the following technical solution: a method for intelligently adjusting the CT room environment based on the Internet of Things, comprising: Locating at least one node in the CT room's environment to be adjusted based on the Internet of Things sensor network to obtain a set of information about the node to be adjusted; wherein the information about the node to be adjusted includes the type of the node to be adjusted, the location of the node to be adjusted, and a local environment topology map; the local environment topology map represents the topological structure of the local CT room environment where the node to be adjusted is located; Collecting real-time environmental parameter information; wherein the real-time environmental parameter information includes at least one of temperature and humidity sensor node data and power load sensor node data; Performing environmental analysis on the node to be adjusted based on the real-time environmental parameter information to obtain an abnormal environment node to be adjusted; Based on the real-time environmental parameter information and the local environmental topology map, environmental adjustment is performed on the abnormal environment node to be adjusted to obtain environmental adjustment information; wherein the environmental adjustment information includes environmental adjustment type, environmental adjustment intensity and energy storage requirement.

[0004] Preferably, performing environmental analysis on the node to be adjusted based on the real-time environmental parameter information to obtain an abnormal environment node to be adjusted includes: Traversing each sensor node data in the real-time environmental parameter information, for the currently traversed sensor node data; determining whether the number of abnormal environment nodes to be adjusted obtained from the current environmental analysis is greater than a preset number threshold; If not, determining the abnormal environment node to be adjusted to which the sensor node belongs based on the adjacent sensor node data of the sensor node in the traversed sensor node data; If so, a merging operation is performed on the abnormal environment nodes to be adjusted that have an associated relationship among the abnormal environment nodes to be adjusted obtained by the current environmental analysis. At the same time, after performing the merging operation, the abnormal environment node to be adjusted to which the adjacent sensor node data belongs is determined, and the abnormal environment node to be adjusted to which the sensor node belongs is determined based on the abnormal environment node to which the adjacent sensor node data belongs.

[0005] Preferably, determining the abnormal environment node to be adjusted to which the sensor node belongs based on the adjacent sensor node data of the sensor node in the traversed sensor node data includes: In response to the adjacent sensor node data not indicating an environmental anomaly, re-creating an abnormal environment node to be adjusted, and using the re-created abnormal environment node to be adjusted as the abnormal environment node to be adjusted to which the sensor node belongs; In response to the presence of only one adjacent sensor node data indicating an abnormal environment among the adjacent sensor node data, taking the abnormal environment node to be adjusted to which the adjacent sensor node indicating the abnormal environment belongs as the abnormal environment node to be adjusted to which the sensor node belongs; In response to the presence of at least two adjacent sensor node data in the adjacent sensor node data indicating an abnormal environment, a target adjacent sensor node is selected from the adjacent sensor nodes indicating an abnormal environment, and the abnormal environment node to be adjusted to which the target adjacent sensor node belongs is used as the abnormal environment node to be adjusted to which the sensor node belongs.

[0006] Preferably, the association relationship between the abnormal environment nodes to be adjusted obtained through environmental analysis is recorded through an association relationship table; wherein, the association relationship table includes at least one data record corresponding to the abnormal environment node to be adjusted; wherein, the data record indicates that the abnormal environment node to be adjusted is a source abnormal area, or the data record indicates that the abnormal environment node to be adjusted is associated with an adjacent abnormal environment node to be adjusted among the abnormal environment nodes to be adjusted determined before the abnormal environment node to be adjusted. If the abnormal environment node to be adjusted is not associated with all abnormal environment nodes to be adjusted determined before the abnormal environment node to be adjusted, then the abnormal environment node to be adjusted is the source abnormal area.

[0007] Preferably, the operation of merging the abnormal environment nodes to be adjusted that have a correlation relationship among the abnormal environment nodes to be adjusted obtained through the current environment analysis includes: For any abnormal environment node to be adjusted that has been analyzed, recursively search the data records in the association table to determine the source abnormal area associated with the abnormal environment node to be adjusted, and update the data record in the association table. The updated data record indicates that the analyzed abnormal environment node to be adjusted is the source abnormal area, or that the analyzed abnormal environment node to be adjusted is associated with one of the source abnormal areas. Based on the updated association table, the abnormal area labels of each abnormal environment node to be adjusted that has been analyzed are updated. For a non-source abnormal area, the abnormal area label of the source abnormal area associated with the non-source abnormal area is used to update the abnormal area label of the non-source abnormal area. After the abnormal area labels of the abnormal environment nodes to be adjusted are updated, the abnormal environment nodes to be adjusted with the same abnormal area labels are merged into one abnormal environment node to be adjusted.

[0008] Preferably, the environmental regulation intensity represents the execution intensity of the environmental regulation type, and the energy storage requirement represents the degree of demand of the node to be regulated for the energy storage converter in the local CT machine room environment corresponding to the local environmental topology diagram.

[0009] Preferably, performing environmental adjustment on the abnormal environment node to be adjusted based on the real-time environmental parameter information and the local environment topology map to obtain environmental adjustment information includes: In response to the real-time environmental parameter information being temperature and humidity sensor node data, determining local environmental parameters based on a pre-trained environmental prediction model and the temperature and humidity sensor node data; obtaining the environmental adjustment information based on a pre-trained environmental adjustment model, the local environmental parameters, and a local environmental topology map of the abnormal node to be adjusted; In response to the real-time environmental parameter information being power load sensor node data, feature extraction is performed on the power load sensor node data to obtain power load features; and environmental adjustment information is obtained based on the power load features, a pre-trained power load prediction model, and the local environmental topology map; In response to the energy storage requirement being greater than a preset energy storage requirement and the environmental regulation intensity being less than a preset regulation intensity, generating an energy storage converter charge and discharge instruction matching the environmental regulation type; In response to the energy storage requirement being greater than the preset energy storage requirement and the environmental regulation intensity being greater than or equal to the preset regulation intensity, an energy storage expansion warning for the local CT machine room environment corresponding to the local environmental topology is generated.

[0010] Preferably, the environmental adjustment model includes a first convolution layer, a first feature fusion layer, a second convolution layer, a first dimensionality reduction layer, a second feature fusion layer, a third convolution layer, a second dimensionality reduction layer, a third feature fusion layer, a global pooling layer and a decision layer.

[0011] Preferably, obtaining the environmental adjustment information based on a pre-trained environmental adjustment model, the local environmental parameters, and the local environmental topology of the abnormal node to be adjusted includes: Performing convolution processing on the local environmental parameter through the first convolution layer to obtain a first environmental feature; Inputting the first environmental feature and the node feature of the local environmental topology map into the first feature fusion layer, the first feature fusion layer enhancing the correlation between the first environmental feature and the node feature through a gated fusion mechanism to obtain a second environmental feature; performing deep convolution processing on the second environmental feature through the second convolution layer to determine an interaction feature of a connection relationship between the second environmental feature and the local CT room environment in the local environmental topology map, thereby obtaining a first topological interaction feature; Performing dimensionality reduction on the first topological interaction feature through the first dimensionality reduction layer, retaining key topology-environment association information, and obtaining a reduced-dimensionality topological feature; Input the reduced-dimensionality topological features into the second feature fusion layer, and dynamically adjust the weights of the node features and the reduced-dimensionality topological features through the attention mechanism according to the feature information of the abnormal node to be adjusted, to obtain the first fused features; Performing a nonlinear transformation on the first fused features through the third convolutional layer to obtain a second fused feature; performing spatial dimension compression on the second fused features through the second dimensionality reduction layer to obtain a fused feature after dimensionality reduction; Inputting the fusion features after dimension reduction into the third feature fusion layer, obtaining the global environment topology features based on the interaction between the edge features and the node features of the local environment topology graph; performing global averaging on the global environment topology features through the global pooling layer to obtain a decision representation vector; The decision representation vector is input into the decision layer, and the environmental regulation type, environmental regulation intensity and energy storage requirement are output respectively.

[0012] An IoT-based intelligent CT room environment adjustment system, which is applicable to the aforementioned IoT-based intelligent CT room environment adjustment method, comprises: Locating at least one node in the CT room's environment to be adjusted based on the Internet of Things sensor network to obtain a set of information about the node to be adjusted; wherein the information about the node to be adjusted includes the type of the node to be adjusted, the location of the node to be adjusted, and a local environment topology map; the local environment topology map represents the topological structure of the local CT room environment where the node to be adjusted is located; Collecting real-time environmental parameter information; wherein the real-time environmental parameter information includes at least one of temperature and humidity sensor node data and power load sensor node data; Performing environmental analysis on the node to be adjusted based on the real-time environmental parameter information to obtain an abnormal environment node to be adjusted; Based on the real-time environmental parameter information and the local environmental topology map, environmental adjustment is performed on the abnormal environment node to be adjusted to obtain environmental adjustment information; wherein the environmental adjustment information includes environmental adjustment type, environmental adjustment intensity and energy storage requirement.

[0013] The present invention provides a technical solution: an intelligent adjustment system for a CT room environment based on the Internet of Things, which is applicable to the above-mentioned intelligent adjustment method for a CT room environment based on the Internet of Things, comprising: An environmental positioning unit is configured to locate at least one to-be-adjusted environmental node in a CT machine room based on an Internet of Things sensor network, and obtain a set of to-be-adjusted point information; wherein the to-be-adjusted node information includes the to-be-adjusted node type, the to-be-adjusted node location, and a local environmental topology map; the local environmental topology map represents the topological structure of the local CT machine room environment where the to-be-adjusted node is located; An environmental sensing unit, configured to collect real-time environmental parameter information; wherein the real-time environmental parameter information includes at least one of temperature and humidity sensor node data and power load sensor node data; An environment analysis unit, configured to perform an environment analysis on the node to be adjusted based on the real-time environment parameter information to obtain an abnormal environment node to be adjusted; An environmental adjustment unit is used to adjust the environment of the abnormal environment node to be adjusted based on the real-time environmental parameter information and the local environmental topology map to obtain environmental adjustment information; wherein the environmental adjustment information includes the environmental adjustment type, environmental adjustment intensity and energy storage requirement Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention monitors the CT room environment in real time through the Internet of Things sensor network. The system can accurately analyze and adjust the environment based on real-time environmental parameter information, ensure the normal operation of the equipment in the room, reduce human intervention, and improve the level of automation. Moreover, through environmental analysis and abnormal node identification, the system can accurately identify environmental abnormal points and intelligently adjust the environment of the adjustment node according to the local environmental topology map to optimize the adjustment effect. Through dynamic analysis of abnormal nodes, the system can gradually merge related abnormal areas to ensure the effectiveness of environmental adjustment. (2) The present invention not only considers the regulation of parameters such as temperature and humidity, but also includes the optimization of energy storage management. Based on the energy storage demand and the environmental regulation intensity, it can generate energy storage converter charging and discharging instructions that match the environmental regulation type, and issue energy storage expansion warnings when necessary, effectively reducing energy waste and improving energy efficiency. Moreover, through the multi-layer convolution and feature fusion mechanism of the environmental regulation model, it realizes the deep fusion of environmental parameters and local environmental topology, which can efficiently extract key environmental features and further improve the accuracy and response speed of environmental regulation. (3) The present invention can automatically adjust strategies according to changes in different environmental parameters. Whether it is temperature and humidity, load data or energy storage requirements, it can be dynamically adjusted according to real-time data, providing flexible adaptability to meet the ever-changing environmental requirements of the CT room; and by obtaining environmental data and energy storage requirements in real time, the system can rationally plan the use of energy storage devices, avoid over-adjustment and waste of energy storage resources, and ensure efficient use of resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A schematic flow chart of the steps of the overall method in one embodiment of the present invention; Figure 2 FIG. 1 is a schematic diagram of the system architecture of the overall system in one embodiment of the present invention.

[0015] In the figure: 1. Environmental positioning unit; 2. Environmental perception unit; 3. Environmental analysis unit; 4. Environmental adjustment unit. DETAILED DESCRIPTION

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0017] For example 1, please refer to Figure 1 The present invention provides a technical solution: a method for intelligently adjusting the CT room environment based on the Internet of Things, comprising: S1. Locating at least one node to be adjusted in a CT machine room based on an Internet of Things sensor network to obtain a set of information about the node to be adjusted; wherein the information about the node to be adjusted includes the type of the node to be adjusted, the location of the node to be adjusted, and a local environment topology map; the local environment topology map represents the topological structure of the local CT machine room environment where the node to be adjusted is located; S2. Collecting real-time environmental parameter information; wherein the real-time environmental parameter information includes at least one of temperature and humidity sensor node data and power load sensor node data; S3. Perform environmental analysis on the nodes to be adjusted based on real-time environmental parameter information to obtain abnormal environmental nodes to be adjusted; S4. Perform environmental adjustment on the abnormal environment node to be adjusted based on the real-time environmental parameter information and the local environmental topology map to obtain environmental adjustment information; wherein the environmental adjustment information includes environmental adjustment type, environmental adjustment intensity and energy storage requirement.

[0018] It should be noted that at least one node to be adjusted in the CT room is located through the IoT sensor network; each node will record its type, location and the local environmental topology of the area where it is located; the topology describes the physical layout of these nodes in the CT room, helping to understand how to adjust the environment of these nodes; node type: such as temperature and humidity sensors, power load sensors, etc.; node location: the specific location of the sensor in the room; local environmental topology: showing the relative positions of these nodes to understand the environmental structure; temperature, humidity, power load and other data are obtained through sensors; these data are used to understand the current environmental conditions of the room; temperature and humidity sensor node data measures the temperature and humidity of the room; power load sensor node data measures the power usage of the room, helping to monitor whether there is any power anomaly or excessive load; according to The real-time data collected is used to analyze the environment of each point to be adjusted; by comparing the normal range with abnormal data, nodes with abnormal environments are identified; for example, excessively high temperature and humidity or excessive power load may indicate that the environment of certain equipment or areas needs to be adjusted; the environment is adjusted according to the abnormal data and topology; the environment of abnormal nodes is corrected by adjusting the environmental control system; for example, areas with excessively high temperatures may need to increase the intensity of air conditioning, and areas with excessively high power load may need to adjust the power of equipment or enable backup power; the environmental adjustment type specifies the adjustment measures to be taken, such as cooling, humidification, adjusting power, etc.; the environmental adjustment intensity is the degree or intensity of adjustment, such as adjusting the air conditioner to a lower temperature or increasing the ventilation speed; the energy storage demand assessment determines whether additional energy storage support is required during the adjustment process, such as using backup batteries or energy storage systems.

[0019] In an optional embodiment, performing environmental analysis on the node to be adjusted based on real-time environmental parameter information to obtain abnormal environment nodes to be adjusted includes: Traverse the sensor node data in the real-time environmental parameter information, and for the currently traversed sensor node data, determine whether the number of abnormal environment nodes to be adjusted obtained from the current environmental analysis is greater than a preset number threshold; If not, determining the abnormal environment node to be adjusted to which the sensor node belongs based on the adjacent sensor node data of the sensor node in the traversed sensor node data; If so, a merge operation is performed on the abnormal environment nodes to be adjusted that have an associated relationship among the abnormal environment nodes to be adjusted obtained by the current environmental analysis. At the same time, after performing the merge operation, the abnormal environment node to be adjusted to which the adjacent sensor node data belongs is determined, and the abnormal environment node to be adjusted to which the sensor node belongs is determined based on the abnormal environment node to which the adjacent sensor node data belongs.

[0020] It should be noted that all sensor node data (such as temperature and humidity sensors, power load sensors, etc.) are traversed to conduct further environmental analysis; the sensor node data are analyzed one by one to determine whether there is an anomaly, and different processing strategies are adopted according to the situation; it is determined whether the number of abnormal environment nodes to be adjusted obtained by the current analysis exceeds the preset threshold; when traversing the data of each sensor node, the system will determine whether there are too many abnormal environment nodes based on the preset abnormal threshold; if the number of abnormal nodes is small, one processing method is adopted; if the number of abnormal nodes is too large, another strategy is adopted; the preset number threshold is used to determine the processing method of the abnormal nodes; for example, when the number of abnormal nodes exceeds a certain value, a more complex merging and adjustment method may be required; if the current number of abnormal nodes is small, a method based on adjacent node data is adopted for processing; if the current number of abnormal nodes is less than the preset threshold, the system will determine which sensor node belongs to which abnormal environment node to be adjusted based on the adjacent node data of the traversed sensor nodes; adjacent Sensor node data refers to the data of sensor nodes that are physically or logically adjacent to the current node; the abnormal environment area to which the current node belongs is inferred through the data of adjacent nodes, and corresponding environmental adjustments are made; if the number of abnormal nodes exceeds the threshold, the abnormal nodes are merged; when the number of abnormal nodes is large, the system will merge these nodes to more effectively manage and adjust the environment; the merging operation is to merge abnormal nodes with similar environmental characteristics or associations into a whole for processing; for example, multiple areas with excessively high temperatures can be regarded as a larger abnormal area for unified adjustment; there may be correlations between multiple abnormal nodes due to environmental factors; the merging operation helps identify these correlations and make unified adjustments; after the abnormal environment node to which the adjacent sensor node data belongs is merged, the system also needs to determine which merged abnormal environment node the adjacent sensor node data should belong to; for example, the temperature and humidity abnormalities in a certain area and the power load abnormalities in the adjacent area may form a new comprehensive abnormal area.

[0021] In an optional embodiment, determining the abnormal environment node to be adjusted to which the sensor node belongs based on the adjacent sensor node data of the sensor node in the traversed sensor node data includes: In response to the fact that the data of the adjacent sensor nodes do not indicate an environmental anomaly, a new abnormal environment node to be adjusted is created, and the new abnormal environment node to be adjusted is used as the abnormal environment node to be adjusted to which the sensor node belongs; In response to the fact that only one adjacent sensor node data among the adjacent sensor node data indicates an abnormal environment, the abnormal environment node to be adjusted to which the adjacent sensor node indicating the abnormal environment belongs is used as the abnormal environment node to be adjusted to which the sensor node belongs; In response to the presence of at least two adjacent sensor node data indicating an abnormal environment, a target adjacent sensor node is selected from the adjacent sensor nodes indicating an abnormal environment, and the abnormal environment node to be adjusted to which the target adjacent sensor node belongs is used as the abnormal environment node to be adjusted to which the sensor node belongs.

[0022] It should be noted that if the data of all sensor nodes adjacent to the current sensor node do not show environmental anomalies (that is, no adjacent nodes have abnormalities); in this case, a new abnormal environment node to be adjusted needs to be recreated; this new node will be used as the abnormal environment node to be adjusted to which the current sensor node belongs; in other words, the current node will be considered as a new abnormal environment area alone; if no adjacent sensor node shows abnormalities, then the current node needs to be processed separately, which may mean that it has some abnormality that has not been detected by other nodes, or the environmental state cannot be directly inferred through adjacent nodes; if among the sensor nodes adjacent to the current sensor node, only one sensor node’s data indicates environmental anomalies; in this case, the current sensor node will belong to the abnormal environment node to be adjusted of the only adjacent sensor node that shows abnormalities; in other words, the current The sensor node will follow the abnormal sensor node and become part of the same abnormal environment. When only one adjacent node shows an abnormality, it can usually be assumed that the abnormality of this node is significant enough to affect the current node. In this way, by associating adjacent nodes, it can be reasonably inferred that the current node may also be in a similar abnormal state. If the data of at least two sensor nodes adjacent to the current sensor node show that the environment is abnormal, in this case, one of the two abnormal adjacent nodes is selected as the target adjacent sensor node. The current sensor node will belong to the abnormal environment node to be adjusted to which the target adjacent sensor node belongs. When multiple adjacent nodes are abnormal, it may be necessary to decide which abnormal node to associate the current node with based on certain criteria. Selecting a target node as a reference is usually determined based on factors such as environmental impact, distance, and correlation between adjacent nodes.

[0023] In an optional embodiment, the association relationship between the abnormal environment nodes to be adjusted obtained by environmental analysis is recorded through an association relationship table; wherein, the association relationship table includes at least one data record corresponding to the abnormal environment node to be adjusted; wherein, the data record indicates that the abnormal environment node to be adjusted is a source abnormal area, or the data record indicates that the abnormal environment node to be adjusted is associated with an adjacent abnormal environment node to be adjusted among the abnormal environment nodes to be adjusted determined before the abnormal environment node to be adjusted; if the abnormal environment node to be adjusted is not associated with all abnormal environment nodes to be adjusted determined before the abnormal environment node to be adjusted, then the abnormal environment node to be adjusted is the source abnormal area.

[0024] It should be noted that when conducting environmental analysis, there may be some kind of connection between abnormal environment nodes (referring to those areas or locations where environmental anomalies are detected); these associations will be stored in an association table; the role of the association table is to clearly record the relationships between these nodes to help analyze the impact and mutual relationships between abnormal nodes; each data record in the association table will be associated with an abnormal environment node to be adjusted; that is, each abnormal environment node to be adjusted will correspond to at least one record in the table; this ensures that all abnormal environment nodes to be adjusted are tracked and recorded, regardless of whether they are associated with other nodes; abnormal environment nodes to be adjusted are source abnormal areas: if an abnormal environment node to be adjusted is not associated with any previously determined abnormal environment node to be adjusted, then it is regarded as a source abnormal area; the source abnormal area refers to There is no other adjacent abnormal area associated with it, which may be the starting point or initial area of ​​abnormal spread; the abnormal environment node to be adjusted is associated with a neighboring abnormal environment node to be adjusted: if the abnormal node is adjacent to and associated with a previously identified abnormal node (that is, an "existing abnormal environment node to be adjusted"), then the record will indicate that this node is associated with another node; in other words, it may be part of an already identified abnormal area, or it is connected to an existing abnormal area; if the current abnormal environment node to be adjusted has no association with any previously identified abnormal environment node to be adjusted, then the node is considered to be the source abnormal area; this means that the node may be the starting point of the abnormality, or it is an independent abnormal area that is not affected by other areas; in this case, it will be regarded as a new abnormal source and requires special attention.

[0025] In an optional embodiment, merging the abnormal environment nodes to be adjusted that have a correlation relationship among the abnormal environment nodes to be adjusted obtained through the current environment analysis includes: For any abnormal environment node to be adjusted that has been analyzed, recursively search the data records in the association table to determine the source abnormal area associated with the abnormal environment node to be adjusted, and update the data records in the association table. The updated data record indicates that the analyzed abnormal environment node to be adjusted is the source abnormal area, or that the analyzed abnormal environment node to be adjusted is associated with one of the source abnormal areas. Based on the updated association table, the abnormal area labels of each abnormal environment node to be adjusted that has been analyzed are updated. For a non-source abnormal area, the abnormal area label of the source abnormal area associated with the non-source abnormal area is used to update the abnormal area label of the non-source abnormal area. After the abnormal area labels of the abnormal environment nodes to be adjusted are updated, the abnormal environment nodes to be adjusted with the same abnormal area labels are merged into one abnormal environment node to be adjusted.

[0026] It should be noted that, for a currently analyzed abnormal environment node to be adjusted, the system needs to recursively search the data records in the association table to determine the source abnormal area associated with the node; through recursive search, the system continuously tracks the relationship between nodes and gradually finds the source abnormal area associated with the abnormal node; the purpose of recursive search is to determine in the association table which source abnormal area the abnormal node belongs to; once the association is determined, the data record in the association table will be updated; the updated record indicates that: the analyzed node is the source abnormal area, or the analyzed node is associated with a source abnormal area, which means that the node belongs to a part of the source abnormal area; through recursive search and updated association table, each analyzed node needs to be searched again. The abnormal area label of the analyzed abnormal environment node to be adjusted is updated; if an abnormal node is not the source abnormal area (that is, it is associated with other nodes), it will inherit the label of its associated source abnormal area; specifically, the label of the non-source abnormal area will be updated to the label of its associated source abnormal area; in this way, all nodes associated with the same source abnormal area will share the same label; once the labels of all abnormal environment nodes to be adjusted are updated, the system will merge the nodes with the same label; specifically, nodes with the same label indicate that they belong to the same abnormal area; these abnormal nodes with the same label will be classified as the same "abnormal environment node to be adjusted", which means that these nodes are considered to be part of the same abnormal area.

[0027] In an optional embodiment, the environmental regulation intensity indicates the execution intensity of the environmental regulation type, and the energy storage requirement indicates the degree of demand of the node to be regulated for the energy storage converter in the local CT machine room environment corresponding to the local environmental topology.

[0028] It should be noted that the intensity of environmental regulation refers to the "intensity" applied by a certain type of environmental regulation during its execution; environmental regulation types may refer to different regulation methods, such as temperature control, humidity regulation, power load regulation, etc.; environmental regulation intensity reflects the strength or degree of regulation activities; for example, if it is temperature regulation, the regulation intensity may be related to the amplitude of temperature regulation - such as small temperature adjustments or large temperature changes; higher regulation intensity usually means that a larger energy or resource input is required, or that the change to the environment is more significant; environmental regulation intensity determines the effect of environmental regulation behavior and the resource consumption during implementation; higher regulation intensity may require more energy and time to complete the regulation task; energy storage demand indicates the amount of energy to be regulated. The degree to which a node requires an energy storage system (such as an energy storage inverter). In the local environmental topology, an energy storage inverter is a device used to manage and regulate electrical energy. In particular, in a power system, it is usually responsible for storing electrical energy and providing power when needed. The energy storage demand reflects the degree to which a node to be regulated (such as a device, region, or system) relies on energy storage devices. This demand may be related to factors such as the node's power consumption requirements, load fluctuations, and the charging and discharging capabilities of the energy storage device. The energy storage demand helps determine how much energy storage resources (such as batteries or energy storage inverters) are needed to ensure the smooth progress of environmental regulation. If a node to be regulated has a high demand for electricity or energy, then the node's energy storage demand is high, meaning that more energy storage support is needed.

[0029] In an optional embodiment, performing environmental adjustment on an abnormal environment node to be adjusted based on real-time environmental parameter information and a local environment topology map to obtain environmental adjustment information includes: In response to the real-time environmental parameter information being the temperature and humidity sensor node data, local environmental parameters are determined based on a pre-trained environmental prediction model and the temperature and humidity sensor node data; environmental adjustment information is obtained based on the pre-trained environmental adjustment model, the local environmental parameters, and a local environmental topology map of the abnormal node to be adjusted; In response to the real-time environmental parameter information being power load sensor node data, feature extraction is performed on the power load sensor node data to obtain power load features; and environmental adjustment information is obtained based on the power load features, a pre-trained power load prediction model, and a local environmental topology map; In response to the energy storage demand being greater than a preset energy storage demand and the environmental regulation intensity being less than a preset regulation intensity, generating an energy storage converter charge and discharge instruction matching the environmental regulation type; In response to the energy storage demand being greater than a preset energy storage demand and the environmental regulation intensity being greater than or equal to a preset regulation intensity, an energy storage expansion warning for a local CT machine room environment corresponding to the local environmental topology map is generated.

[0030] It should be noted that the system obtains data from temperature and humidity sensor nodes, which reflect the temperature and humidity in the environment in real time. Based on a pre-trained environmental prediction model (usually the result of machine learning or data modeling), the system uses the temperature and humidity sensor data to predict and determine local environmental parameters. For example, these predictions may include future temperature and humidity changes. Combining local environmental parameters (such as temperature and humidity) with the environmental topology map of the abnormal nodes to be adjusted (which may refer to the node location and impact range related to environmental adjustment), the pre-trained environmental adjustment model generates corresponding adjustment measures or adjustment instructions. In short, the system predicts and adjusts the environment to ensure that it meets the set requirements. Through power load sensor nodes, the system obtains real-time power load information. These sensors monitor the power consumption of each device or area. Feature extraction is performed on the data of power load sensor nodes to extract useful information or trends from the raw data, such as load fluctuation patterns and peak loads. Based on the extracted power load features, the system uses a pre-trained power load prediction model to predict future load demand. These predictions help determine future power supply demand and ensure that adjustment measures can adapt to changes. Combining power load features, prediction models, and local environmental topology maps, the system generates corresponding environmental adjustment information. This may involve adjusting the power load. Adjustment or optimization, such as adjusting equipment or increasing energy storage to cope with future load demands; the energy storage demand reflects the degree of demand for electricity by energy storage equipment (such as batteries, energy storage converters) in the current system; when the energy storage demand is greater than the preset energy storage demand, it means that the current energy storage demand is high; the environmental regulation intensity indicates the degree of intervention in the environment during the environmental regulation process; if the environmental regulation intensity is less than the preset regulation intensity, the system will generate energy storage converter charge and discharge instructions that match the current environmental regulation type; in other words, when the demand is high and the regulation intensity is low, the system will instruct the energy storage converter to charge or discharge to maintain the balance of environmental regulation; the energy storage converter charge and discharge instruction indicates The energy storage device performs corresponding charging or discharging operations to meet the regulation demand; an energy storage demand greater than the preset energy storage demand indicates a high energy storage demand, which may mean that the existing energy storage resources are insufficient to meet the demand; when the environmental regulation intensity is high (i.e., greater intervention is required), more energy storage resources may be required to support environmental regulation; if the energy storage demand is already high and the regulation intensity is also strong, the system will generate an energy storage expansion warning; this means that the existing energy storage resources may not be able to meet future regulation needs, and the capacity of the energy storage device needs to be increased or expanded to avoid energy shortages affecting environmental regulation; if the existing energy storage device cannot meet the demand, the system will issue a warning, prompting the need to increase energy storage resources.

[0031] In an optional embodiment, the environmental adjustment model includes a first convolution layer, a first feature fusion layer, a second convolution layer, a first dimensionality reduction layer, a second feature fusion layer, a third convolution layer, a second dimensionality reduction layer, a third feature fusion layer, a global pooling layer and a decision layer.

[0032] In an optional embodiment, the environmental adjustment information is obtained based on a pre-trained environmental adjustment model, local environmental parameters, and a local environmental topology map of the abnormal node to be adjusted, including: Performing convolution processing on the local environment parameters through the first convolution layer to obtain the first environment feature; The first environmental feature and the node feature of the local environmental topology map are input into the first feature fusion layer. The first feature fusion layer enhances the correlation between the first environmental feature and the node feature through a gated fusion mechanism to obtain the second environmental feature. Performing deep convolution processing on the second environmental feature through the second convolution layer to determine the interaction feature of the connection relationship between the second environmental feature and the local CT room environment in the local environmental topology map, thereby obtaining a first topological interaction feature; The first topological interaction feature is compressed by the first dimensionality reduction layer, retaining the key topology-environment correlation information to obtain the reduced-dimensional topological feature; The reduced-dimensional topological features are input into the second feature fusion layer. Based on the feature information of the abnormal node to be adjusted, the weights of the node features and the reduced-dimensional topological features are dynamically adjusted through the attention mechanism to obtain the first fusion feature. The first fusion feature is nonlinearly transformed through the third convolution layer to obtain the second fusion feature; the second fusion feature is spatially compressed through the second dimensionality reduction layer to obtain the fusion feature after dimensionality reduction; The fused features after dimensionality reduction are input into the third feature fusion layer. Based on the interaction between the edge features and node features of the local environment topology graph, the global environment topology features are obtained. The global environment topology features are globally averaged through the global pooling layer to obtain the decision representation vector. The decision representation vector is input into the decision layer, and the environmental regulation type, environmental regulation intensity and energy storage requirement are output respectively.

[0033] It should be noted that the first convolution layer performs a convolution operation on the input environmental parameters to extract preliminary feature information to form a "first environmental feature"; the convolution layer can help capture the patterns or laws of the local environment, such as local features of temperature and humidity changes; the first feature fusion layer enhances the correlation between the first environmental feature and the node feature through a gated fusion mechanism; the gating mechanism is similar to a gated neural network, which can dynamically adjust the fusion method between features according to the different importance of the input features; after fusion, a second environmental feature is generated, which is a high-level feature that combines environmental parameters and topological information; the second convolution layer performs further convolution processing on the second environmental feature, with the goal of extracting the interaction features between the local CT room environment and other nodes in the topological map; through this deep convolution, the system can capture deeper connections between environmental features and topological nodes, and generate first topological interaction features; the first dimensionality reduction layer compresses the dimension of the feature through dimensionality reduction operations, retaining only the most critical topological and environmental correlation information; this step reduces the amount of computation while ensuring that important features Information will not be lost, and the dimensionality reduction topological features are generated; the second feature fusion layer combines the features of the abnormal nodes to be adjusted and the dimensionality reduction topological features; the attention mechanism is used to dynamically adjust the weights of these two features, so that the model can pay more attention to specific nodes or areas according to the current state; this step generates the first fusion feature, which combines the topological information and the features of specific nodes; the third convolutional layer performs a nonlinear transformation on the first fusion feature to further extract deep features and patterns; this helps the model capture more complex environments and topological interactions and generate the second fusion feature; the second dimensionality reduction layer compresses the spatial dimension of the second fusion feature, reduces the dimension of the feature, and retains important fusion features; this step generates the dimensionality reduction fusion feature, which provides a concise feature representation that contains the core information of the environment and topology; The third feature fusion layer interacts the fusion features after dimensionality reduction with the edge features and node features of the local environmental topology map to generate global environmental topological features. This step further enhances the global perception capability of the model by fusing global information (such as the relationship between nodes). The global pooling layer performs global average pooling on the global environmental topological features, simplifies the feature expression, and generates a smaller decision representation vector with global information. Global pooling can effectively compress information and retain the most representative features. The final decision layer outputs the environmental regulation type, regulation intensity, and energy storage requirement based on the decision representation vector. These outputs are used to guide the specific actions of the environmental regulation system, such as deciding what type of regulation to adopt, the intensity of regulation, and energy storage requirement.

[0034] For example 2, please refer to Figure 2 The present invention provides a technical solution: an intelligent adjustment system for a CT room environment based on the Internet of Things, which is applicable to the above-mentioned intelligent adjustment method for a CT room environment based on the Internet of Things, including: An environmental positioning unit 1 is configured to locate at least one environmental node to be adjusted in a CT machine room based on an Internet of Things sensor network, and obtain a set of information about the node to be adjusted; wherein the information about the node to be adjusted includes the type of the node to be adjusted, the location of the node to be adjusted, and a local environmental topology map; the local environmental topology map represents the topological structure of the local CT machine room environment where the node to be adjusted is located; Environmental sensing unit 2, used to collect real-time environmental parameter information; wherein the real-time environmental parameter information includes at least one of temperature and humidity sensor node data and power load sensor node data; An environment analysis unit 3 is used to perform environment analysis on the nodes to be adjusted based on real-time environment parameter information to obtain abnormal environment nodes to be adjusted; The environmental adjustment unit 4 is used to adjust the environment of the abnormal environment node to be adjusted based on the real-time environmental parameter information and the local environmental topology map to obtain environmental adjustment information; wherein the environmental adjustment information includes the environmental adjustment type, environmental adjustment intensity and energy storage requirement.

[0035] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. A method for intelligently adjusting the CT room environment based on the Internet of Things, characterized in that: include: Locating at least one node in the CT room's environment to be adjusted based on the Internet of Things sensor network to obtain a set of information about the node to be adjusted; wherein the information about the node to be adjusted includes the type of the node to be adjusted, the location of the node to be adjusted, and a local environment topology map; the local environment topology map represents the topological structure of the local CT room environment where the node to be adjusted is located; Collecting real-time environmental parameter information; wherein the real-time environmental parameter information includes at least one of temperature and humidity sensor node data and power load sensor node data; Performing environmental analysis on the node to be adjusted based on the real-time environmental parameter information to obtain an abnormal environment node to be adjusted; Based on the real-time environmental parameter information and the local environmental topology map, environmental adjustment is performed on the abnormal environment node to be adjusted to obtain environmental adjustment information; wherein the environmental adjustment information includes environmental adjustment type, environmental adjustment intensity and energy storage requirement.

2. The method for intelligently adjusting the CT room environment based on the Internet of Things according to claim 1, characterized in that: Performing an environmental analysis on the node to be adjusted based on the real-time environmental parameter information to obtain an abnormal environment node to be adjusted, including: Traversing each sensor node data in the real-time environmental parameter information, for the currently traversed sensor node data; determining whether the number of abnormal environment nodes to be adjusted obtained from the current environmental analysis is greater than a preset number threshold; If not, determining the abnormal environment node to be adjusted to which the sensor node belongs based on the adjacent sensor node data of the sensor node in the traversed sensor node data; If so, a merging operation is performed on the abnormal environment nodes to be adjusted that have an associated relationship among the abnormal environment nodes to be adjusted obtained by the current environmental analysis. At the same time, after performing the merging operation, the abnormal environment node to be adjusted to which the adjacent sensor node data belongs is determined, and the abnormal environment node to be adjusted to which the sensor node belongs is determined based on the abnormal environment node to which the adjacent sensor node data belongs.

3. The method for intelligently adjusting the CT room environment based on the Internet of Things according to claim 2, characterized in that: Determining the abnormal environment node to be adjusted to which the sensor node belongs based on the adjacent sensor node data of the sensor node in the traversed sensor node data includes: In response to the adjacent sensor node data not indicating an environmental anomaly, re-creating an abnormal environment node to be adjusted, and using the re-created abnormal environment node to be adjusted as the abnormal environment node to be adjusted to which the sensor node belongs; In response to the presence of only one adjacent sensor node data indicating an abnormal environment among the adjacent sensor node data, taking the abnormal environment node to be adjusted to which the adjacent sensor node indicating the abnormal environment belongs as the abnormal environment node to be adjusted to which the sensor node belongs; In response to the presence of at least two adjacent sensor node data in the adjacent sensor node data indicating an abnormal environment, a target adjacent sensor node is selected from the adjacent sensor nodes indicating an abnormal environment, and the abnormal environment node to be adjusted to which the target adjacent sensor node belongs is used as the abnormal environment node to be adjusted to which the sensor node belongs.

4. The method for intelligently adjusting the CT room environment based on the Internet of Things according to claim 3 is characterized in that: The association relationship between the abnormal environment nodes to be adjusted obtained through environmental analysis is recorded through an association relationship table; wherein, the association relationship table includes at least one data record corresponding to the abnormal environment node to be adjusted; wherein, the data record indicates that the abnormal environment node to be adjusted is a source abnormal area, or the data record indicates that the abnormal environment node to be adjusted is associated with an adjacent abnormal environment node to be adjusted among the abnormal environment nodes to be adjusted determined before the abnormal environment node to be adjusted. If the abnormal environment node to be adjusted is not associated with all abnormal environment nodes to be adjusted determined before the abnormal environment node to be adjusted, then the abnormal environment node to be adjusted is the source abnormal area.

5. The method for intelligently adjusting the CT room environment based on the Internet of Things according to claim 4 is characterized in that: Merge the abnormal environment nodes to be adjusted that have a related relationship among the abnormal environment nodes to be adjusted obtained through the current environment analysis, including: For any abnormal environment node to be adjusted that has been analyzed, recursively search the data records in the association table to determine the source abnormal area associated with the abnormal environment node to be adjusted, and update the data record in the association table. The updated data record indicates that the analyzed abnormal environment node to be adjusted is the source abnormal area, or that the analyzed abnormal environment node to be adjusted is associated with one of the source abnormal areas. Based on the updated association table, the abnormal area labels of each abnormal environment node to be adjusted that has been analyzed are updated. For a non-source abnormal area, the abnormal area label of the source abnormal area associated with the non-source abnormal area is used to update the abnormal area label of the non-source abnormal area. After the abnormal area labels of the abnormal environment nodes to be adjusted are updated, the abnormal environment nodes to be adjusted with the same abnormal area labels are merged into one abnormal environment node to be adjusted.

6. The method for intelligently adjusting the CT room environment based on the Internet of Things according to claim 5, characterized in that: The environmental regulation intensity indicates the execution intensity of the environmental regulation type, and the energy storage requirement indicates the degree of demand of the node to be regulated for the energy storage converter in the local CT machine room environment corresponding to the local environmental topology.

7. The method for intelligently adjusting the CT room environment based on the Internet of Things according to claim 6, characterized in that: Performing environmental adjustment on the abnormal environment node to be adjusted based on the real-time environmental parameter information and the local environment topology map to obtain environmental adjustment information includes: In response to the real-time environmental parameter information being temperature and humidity sensor node data, determining local environmental parameters based on a pre-trained environmental prediction model and the temperature and humidity sensor node data; obtaining the environmental adjustment information based on a pre-trained environmental adjustment model, the local environmental parameters, and a local environmental topology map of the abnormal node to be adjusted; In response to the real-time environmental parameter information being power load sensor node data, feature extraction is performed on the power load sensor node data to obtain power load features; and environmental adjustment information is obtained based on the power load features, a pre-trained power load prediction model, and the local environmental topology map; In response to the energy storage requirement being greater than a preset energy storage requirement and the environmental regulation intensity being less than a preset regulation intensity, generating an energy storage converter charge and discharge instruction matching the environmental regulation type; In response to the energy storage requirement being greater than the preset energy storage requirement and the environmental regulation intensity being greater than or equal to the preset regulation intensity, an energy storage expansion warning for the local CT machine room environment corresponding to the local environmental topology is generated.

8. The method for intelligently adjusting the CT room environment based on the Internet of Things according to claim 7, characterized in that: The environmental adjustment model includes a first convolution layer, a first feature fusion layer, a second convolution layer, a first dimensionality reduction layer, a second feature fusion layer, a third convolution layer, a second dimensionality reduction layer, a third feature fusion layer, a global pooling layer and a decision layer.

9. The method for intelligently adjusting the CT room environment based on the Internet of Things according to claim 8, characterized in that: The environmental adjustment information is obtained based on a pre-trained environmental adjustment model, the local environmental parameters, and a local environmental topology map of the abnormal node to be adjusted, including: Performing convolution processing on the local environmental parameter through the first convolution layer to obtain a first environmental feature; Inputting the first environmental feature and the node feature of the local environmental topology map into the first feature fusion layer, the first feature fusion layer enhancing the correlation between the first environmental feature and the node feature through a gated fusion mechanism to obtain a second environmental feature; performing deep convolution processing on the second environmental feature through the second convolution layer to determine an interaction feature of a connection relationship between the second environmental feature and the local CT room environment in the local environmental topology map, thereby obtaining a first topological interaction feature; Performing dimensionality reduction on the first topological interaction feature through the first dimensionality reduction layer, retaining key topology-environment association information, and obtaining a reduced-dimensionality topological feature; Input the reduced-dimensionality topological features into the second feature fusion layer, and dynamically adjust the weights of the node features and the reduced-dimensionality topological features through the attention mechanism according to the feature information of the abnormal node to be adjusted, to obtain the first fused features; Performing a nonlinear transformation on the first fused features through the third convolutional layer to obtain a second fused feature; performing spatial dimension compression on the second fused features through the second dimensionality reduction layer to obtain a fused feature after dimensionality reduction; Inputting the fusion features after dimension reduction into the third feature fusion layer, obtaining the global environment topology features based on the interaction between the edge features and the node features of the local environment topology graph; performing global averaging on the global environment topology features through the global pooling layer to obtain a decision representation vector; The decision representation vector is input into the decision layer, and the environmental regulation type, environmental regulation intensity and energy storage requirement are output respectively.

10. An Internet of Things-based intelligent CT room environment adjustment system, which is applicable to the Internet of Things-based CT room environment intelligent adjustment method according to any one of claims 1 to 9, characterized in that: include: An environmental positioning unit is configured to locate at least one to-be-adjusted environmental node in a CT machine room based on an Internet of Things sensor network, and obtain a set of to-be-adjusted point information; wherein the to-be-adjusted node information includes the to-be-adjusted node type, the to-be-adjusted node location, and a local environmental topology map; the local environmental topology map represents the topological structure of the local CT machine room environment where the to-be-adjusted node is located; An environmental sensing unit, configured to collect real-time environmental parameter information; wherein the real-time environmental parameter information includes at least one of temperature and humidity sensor node data and power load sensor node data; An environment analysis unit, configured to perform an environment analysis on the node to be adjusted based on the real-time environment parameter information to obtain an abnormal environment node to be adjusted; An environmental adjustment unit is used to perform environmental adjustment on the abnormal environment node to be adjusted based on the real-time environmental parameter information and the local environmental topology map to obtain environmental adjustment information; wherein the environmental adjustment information includes environmental adjustment type, environmental adjustment intensity and energy storage requirement.

Images