Intelligent monitoring and regulation method and system for oxygen cabin with internet of things

Abstract

The present application relates to the technical field of Internet of Things, in particular to an intelligent monitoring and regulating method and system for an oxygen cabin in Internet of Things, comprising the following steps: obtaining cabin pressure, oxygen concentration, flow, mask pressure difference, blood oxygen and wireless message arrival time by using Internet of Things nodes at the cabin body, oxygen supply pipeline, oxygen exhaust pipeline and wearing mask, forming an oxygen cabin credible sampling sequence, calculating pressure-oxygen coupling numerical value, synthesizing cabin patient linkage state sequence, constructing dynamic regulation boundary, generating regulation offset numerical value and outputting closed-loop control instruction. In the present application, the use order of readings is constrained by the credibility of wireless messages, the synchronous change of cabin pressure and oxygen concentration is reflected by the pressure-oxygen coupling numerical value, the wearing closed state is identified by the relationship between mask pressure difference and flow, the dynamic regulation boundary is generated by stable segments, the oxygen supply valve, oxygen exhaust valve and node reporting frequency are adjusted according to the cabin patient state linkage, and the regulation deviation caused by communication jitter and cabin pressure inertia is reduced.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) technology, and in particular to an IoT-based intelligent monitoring and control method and system for oxygenation chambers. Background Technology

[0002] The Internet of Things (IoT) technology involves connecting embedded nodes, wireless communication links, gateways, and execution devices into a single control network to collect, transmit, and control the status variables, operation commands, and device feedback of physical objects. Traditional IoT-based intelligent monitoring and control methods for oxygen chambers involve installing pressure transmitters, oxygen concentration probes, temperature resistors, and humidity probes on the oxygen chamber body, and flow meters, solenoid valves, and mechanical pressure regulating valves on the oxygen supply and exhaust pipelines. A control cabinet near the chamber receives electrical signals from the probes and drives the oxygen supply valves, exhaust valves, and audible and visual alarms via relays. On-duty personnel manually confirm the chamber door sealing, mask wearing, and oxygen supply and exhaust status based on display readings and paper inspection records.

[0003] Existing oxygen chamber monitoring and control methods mostly rely on pressure gauges, oxygen concentration probes, flow meters, and manual inspection records to determine the operating status separately. There is a lack of synchronous conversion between changes in chamber pressure and changes in oxygen concentration. When masks are loose, pipeline flow is still used as an indication of sufficient oxygen supply. Retransmission or delay of wireless node messages can easily mix in sampling signals and control data. Valve opening is often adjusted according to fixed safety ranges. When chamber pressure inertia and personnel breathing fluctuations are superimposed, oxygen supply and exhaust actions are prone to repeated corrections. There is a time misalignment between alarm judgment and control action. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing an IoT-based intelligent monitoring and control method and system for oxygenated chambers.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: an IoT-based intelligent monitoring and control method for oxygenated chambers, comprising the following steps: By deploying IoT nodes in the oxygen chamber, oxygen supply pipeline, oxygen exhaust pipeline, and mask wearing area, the chamber pressure reading, oxygen concentration reading, pipeline flow reading, mask differential pressure reading, blood oxygen reading, and wireless message arrival time are directly obtained. The readings within the same sampling time slot are analyzed according to the wireless message arrival time and the number of node retransmissions to generate a reliable sampling sequence for the oxygen chamber. Based on the change of the product of chamber pressure reading and oxygen concentration reading in the oxygen chamber reliable sampling sequence, the pressure-oxygen coupling value is calculated. Based on the duration of the same direction of the mask pressure difference reading and the pipeline flow reading, the wearing and sealing status is determined. The pressure-oxygen coupling value and the wearing and sealing status are combined into a chamber-patient linkage status sequence. Stable segments in the cabin-patient linkage state sequence are screened, and a dynamic control boundary is constructed using the median difference of the stable segments and the adjacent sampling slope. The current pressure oxygen coupling value, blood oxygen reading and the dynamic control boundary are compared to generate a control offset value. The oxygen supply valve opening increment, oxygen exhaust valve opening increment, and node reporting interval are calculated based on the control offset value. The reporting frequency of the oxygen supply valve, oxygen exhaust valve, and IoT node is adjusted, and a closed-loop control command is output based on the directional consistency between the oxygen pressure coupling value and the control offset value.

[0006] As a further aspect of the present invention, the process of generating the reliable sampling sequence of the oxygen chamber specifically includes: Collect the node identifier, sampling time, arrival time of the wireless message, number of retransmissions of the node and corresponding readings sent by the IoT node, and classify the corresponding readings into the same sampling time slot according to the sampling time and the preset sampling period, and establish a time slot message record according to the node identifier; Compare the arrival delay between the arrival time of each wireless message in the time slot message record and the start time of the corresponding sampling time slot. Prioritize arranging them according to the arrival delay from smallest to largest. If the arrival delays are the same, arrange them according to the number of retransmissions of the node from smallest to largest to generate a reliable message sort. The corresponding readings that meet the preset delay range and preset retransmission count limit in the trusted order of the messages are selected. For repeated wireless messages from the same node, the corresponding readings with the earlier trusted order are retained and aligned according to the sampling time slot and node identifier to generate a trusted sampling sequence for the oxygen chamber.

[0007] As a further aspect of the present invention, the generation process of the cabin-disease linkage state sequence specifically includes: Read the chamber pressure reading and the oxygen concentration reading in the same sampling time slot of the oxygen chamber reliable sampling sequence, multiply the chamber pressure reading and the oxygen concentration reading, compare the direction and magnitude of change of the product result of adjacent sampling time slots, and generate a pressure-oxygen coupling value. Extract the mask pressure difference reading and the pipeline flow reading in the same sampling time slot, determine whether the increase or decrease direction of the mask pressure difference reading and the pipeline flow reading is consistent in the continuous sampling time slot, accumulate the duration of the same direction, and compare the duration of the same direction with a preset sealing duration threshold to obtain the wearing sealing state; The pressure oxygen coupling value is associated with the sampling time and the wearing airtight state. A linkage state item is established based on the value change and state change within the continuous sampling time slot. The linkage state items are arranged according to the sampling time to generate the cabin-disease linkage state sequence.

[0008] As a further aspect of the present invention, the process of generating the adjustment offset value specifically includes: The variation range of the pressure oxygen coupling value and blood oxygen reading in the cabin-patient linkage state sequence between consecutive state items is detected, and the wearing airtight state corresponding to the consecutive state item is read. State items in which both types of variation range are within a preset stable range and the wearing airtight state is continuous and consistent are divided into stable segments. Calculate the median of the pressure-oxygen coupling value and the median of the blood oxygen reading within the stable segment. Scale the data based on the median of the pressure-oxygen coupling value and the median of the blood oxygen reading. Determine the boundary center and boundary width based on the median difference of the scaling results and the adjacent sampling slope to construct a dynamic control boundary. The positional relationship between the current pressure oxygen coupling value and the proportionalized result of the blood oxygen reading and the dynamic control boundary is compared. The offset direction and offset magnitude are determined based on the direction and distance beyond the boundary, and the control offset value is generated.

[0009] As a further aspect of the present invention, the process of generating the closed-loop control command specifically includes: Read the offset direction and offset magnitude of the control offset value, determine the opening adjustment direction of the oxygen supply valve and the oxygen exhaust valve according to the offset direction, match the opening change level and data acquisition density according to the offset magnitude, and calculate the oxygen supply valve opening increment, the oxygen exhaust valve opening increment and the node reporting interval. The oxygen supply valve opening increment and the oxygen exhaust valve opening increment are compared with the preset valve opening range. The increment that causes the valve opening to exceed the preset valve opening range is corrected to the corresponding range boundary. The reporting frequency of the Internet of Things node is adjusted according to the node reporting interval to establish the control execution parameters. The pressure-oxygen coupling value after the control is executed is collected, and the direction of the change of the value before and after the control is executed is compared with the direction of the offset of the control offset value. The consistency of the direction is determined. If the direction is consistent, the control execution parameter is retained. If the direction is opposite, the control execution parameter is cancelled. If the direction of the change of the value is stable, the control execution parameter is reduced, and a closed-loop control command is output.

[0010] As a further aspect of the present invention, the process of generating the trusted message ordering specifically includes: Extract the sampling time, arrival time and sampling time slot start time of each wireless message from the time slot message record, calculate the transmission delay of the arrival time relative to the sampling time, and determine the delay confidence level according to the proportion of the transmission delay to the preset sampling period. The number of wireless packets with the same node identifier in the same sampling time slot and the number of times the node retransmits are counted. The difference between the number of times the node retransmits and the preset upper limit of the number of retransmissions is compared, and the retransmission reliability level is determined by combining the arrival order of the duplicate wireless packets. According to the hierarchical order of the delay trust level, the retransmission trust level, and the arrival order, the wireless packets in the same sampling time slot are compared level by level. For wireless packets with the same trust conditions, a fixed order is maintained according to the node identifier to generate a packet trust order.

[0011] As a further aspect of the present invention, the process for determining the wearing seal state specifically includes: Acquire the mask pressure differential reading and the pipeline flow reading within a continuous sampling time slot, compare the changes in adjacent readings with a preset fluctuation dead zone, mark the changes within the preset fluctuation dead zone as the stable direction, and mark the changes outside the preset fluctuation dead zone as the rising or falling direction, and generate a reading direction sequence. Compare the mask pressure difference reading direction and the pipeline flow reading direction corresponding to the same sampling time slot in the reading direction sequence. Accumulate the same-direction duration for sampling time slots with the same direction or both being stable directions, and clear the same-direction duration for sampling time slots with opposite directions to obtain a closed continuous record. The continuous sealing record is compared with the preset continuous sealing threshold. The mask pressure difference reading is detected to see if it is within the preset sealing pressure difference range. When the continuous threshold is reached and the pressure difference is within the preset sealing pressure difference range, the sealing state is determined. When the continuous threshold is not reached or the pressure difference is exceeded, the non-sealing state is determined, thus obtaining the wearing sealing state.

[0012] As a further aspect of the present invention, the process of constructing the dynamic control boundary specifically includes: Obtain the pressure oxygen coupling value and blood oxygen reading aligned by sampling time within each stable segment, calculate the median of the pressure oxygen coupling value and the median of the blood oxygen reading, and scale each sampled value based on the median of the pressure oxygen coupling value and the median of the blood oxygen reading to establish a stable scaling sequence. Calculate the difference between the hyperbaric oxygen ratio and the blood oxygen ratio in the same sampling time slot in the stable ratio sequence, sort them according to the size of the difference, extract the middle position difference when the number of differences is odd, and calculate the average of the two middle differences when the number of differences is even to obtain the median difference. At the same time, compare the change of the ratio value of adjacent sampling time slots with the sampling interval to obtain the adjacent sampling slope. Calculate the boundary center corresponding to the median difference and the boundary width corresponding to the median of the absolute values ​​of the adjacent sampling slopes, and move the boundary center according to the changing direction of the current adjacent sampling slopes to establish a dynamically adjustable boundary.

[0013] As a further aspect of the present invention, the process for determining directional consistency specifically includes: The pressure-oxygen coupling value, the control offset value, and the control execution parameters before the control are executed are obtained. Based on the offset direction of the control offset value and the opening adjustment direction of the oxygen supply valve and the oxygen exhaust valve, the expected change direction of the pressure-oxygen coupling value is determined. The pressure-oxygen coupling values ​​are collected in three consecutive sampling time slots after the control is executed. Abnormal values ​​that exceed the preset upper limit of adjacent change range are removed. The increase or decrease relationship between the retained values ​​and the values ​​before the control is executed is compared to obtain the actual change direction. The expected change direction is compared with the actual change direction. If the directions are consistent, the control execution parameters are retained. If the directions are opposite, the control execution parameters are canceled. When the actual change direction is in a stable direction, the valve opening increment is reduced and the node reporting interval is shortened, and a closed-loop control command is output.

[0014] An IoT-based intelligent monitoring and control system for an oxygenation chamber, comprising: The reliable sampling module acquires chamber pressure readings, oxygen concentration readings, pipeline flow readings, mask differential pressure readings, blood oxygen readings, wireless message arrival times, and node retransmission counts. It analyzes the readings within the same sampling time slot according to the wireless message arrival times and node retransmission counts to generate a reliable sampling sequence for the oxygen chamber. The linkage synthesis module calculates the pressure oxygen coupling value and determines the wearing and sealing status based on the oxygen chamber reliable sampling sequence, and synthesizes the pressure oxygen coupling value and the wearing and sealing status into a chamber-patient linkage status sequence. The boundary offset module filters out stable segments in the cabin-patient linkage state sequence, constructs a dynamic control boundary using the median difference and the adjacent sampling slope, compares the current pressure oxygen coupling value, blood oxygen reading and the dynamic control boundary, and generates a control offset value. The closed-loop control module calculates the oxygen supply valve opening increment, oxygen exhaust valve opening increment, and node reporting interval based on the control offset value, adjusts the reporting frequency of the oxygen supply valve, oxygen exhaust valve, and the IoT node, compares whether the change direction of the oxygen pressure coupling value and the control offset value is consistent, and outputs closed-loop control commands.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, by sorting the chamber pressure reading, oxygen concentration reading, and wireless message arrival time within the same sampling time slot, the interference of retransmission delay on the control basis is filtered out. The pressure-oxygen coupling value is formed by the change of the product of chamber pressure and oxygen concentration, so that the oxygen supply judgment simultaneously reflects the chamber pressure inertia and oxygen content fluctuation. The wearing and sealing status is identified by the continuous relationship between mask pressure difference and pipeline flow, avoiding misjudgment of oxygen supply caused by loose wearing. Dynamic control boundary is generated by the median difference of stable segments and the slope of adjacent sampling, so that the opening of oxygen supply valve, oxygen exhaust valve, and node reporting interval are adjusted according to the chamber fault linkage status, reducing the risk of reciprocating correction and alarm misalignment. Attached Figure Description

[0016] Figure 1 This is a flowchart of the IoT-based intelligent monitoring and control process for the oxygenation chamber of this invention. Figure 2 This is a schematic diagram illustrating the reliable sampling and abnormal message suppression of the present invention; Figure 3 This is a schematic diagram illustrating the formation of the cabin-hazard linkage state in this invention; Figure 4 This is a schematic diagram of the dynamic control boundary and control offset state of the present invention; Figure 5 This is a schematic diagram of the coordination and feedback of the closed-loop control module of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the software-based technical solution is described in detail below with reference to system architecture diagrams and embodiments. It should be understood that the specific embodiments described herein are only for explaining the technical solutions of this invention and do not constitute a limitation on the scope of protection.

[0018] In the description of this invention, the system architecture relationships or data processing flows indicated by terms such as "layer," "module," "interface," "data flow," "client," and "server" are all defined based on the architecture diagram or flowchart corresponding to the embodiments. This way of describing is only used to clearly illustrate the logical relationships between the elements in the technical solution, and not to limit the physical deployment form. The term "multiple" includes two or more technical units, including but not limited to multiple data nodes, processing threads, service instances, or functional components and other scalable elements. The specific number is determined according to the actual business scenario and needs to be specifically specified.

[0019] Please see Figures 1 to 5This embodiment provides an IoT-based intelligent monitoring and control method for oxygen chambers. During the actual operation of the oxygen chamber, with the recipient wearing a mask and the oxygen supply and exhaust pipelines jointly maintaining the chamber environment, IoT nodes deployed at the oxygen chamber body, oxygen supply pipelines, oxygen exhaust pipelines, and the mask wearing location continuously generate sensor messages. The local controller of the oxygen chamber receives these sensor messages and drives the oxygen supply and exhaust valves. The IoT nodes include chamber pressure acquisition nodes, oxygen concentration acquisition nodes, pipeline flow acquisition nodes, mask pressure difference acquisition nodes, and blood oxygen acquisition nodes. After sampling is triggered, each node writes its node identifier, sampling time, sensor reading, wireless message arrival time, and retransmission count fields into the message payload. The wireless message arrival time is a timing field recorded by the local controller of the oxygen chamber when it completes message reception and passes integrity verification. The retransmission count field is the repeated transmission status recorded by the IoT node during the current sampling message transmission process. The local controller of the oxygen chamber processes sensor messages in the order of reliable sampling, linkage state synthesis, dynamic boundary generation, and closed-loop control, forming oxygen supply valve control field, oxygen exhaust valve control field, and node reporting interval field.

[0020] S1: Receive sensor messages sent by IoT nodes. Based on the sampling time, classify chamber pressure readings, oxygen concentration readings, pipeline flow readings, mask differential pressure readings, and blood oxygen readings into the corresponding sampling time slots. Then, evaluate the reliable order of repeated messages within the same sampling time slot according to the arrival time of the wireless message and the retransmission count field. Retain readings that meet the conditions of message integrity, timing validity, and node uniqueness to generate a reliable oxygen chamber sampling sequence. The reliable oxygen chamber sampling sequence is a structured record arranged according to sampling time slots. Each record carries a sampling time slot identifier, node identifier, reading type, reading content, reliable status, and abnormal status. Subsequent steps only call records with a valid reliable status. Abnormal status records are entered into the acquisition anomaly log, retaining the node identifier, anomaly category, and recovery status for the next sampling time slot to perform missing data compensation and node communication status judgment.

[0021] S101: Collect sensor messages sent by each IoT node, parse the node identifier, reading type, sampling time, and retransmission count fields in the message header, perform format verification on the sensor readings in the message payload, and record the arrival time of the wireless message after the integrity field at the end of the message passes verification. The node identifier is jointly defined by the IoT node's factory configuration record and the oxygen chamber equipment registration record, and the reading type maintains a fixed correspondence with the node's installation location. Format verification checks the field length, field type, and reading status identifier. Messages that cannot be parsed are written to a parsing error record and are not included in the sampling time slot grouping. Messages that fail integrity verification are marked as transmission damaged, and the oxygen chamber local controller returns a retransmission request to the corresponding node; if a complete message is not obtained before the start of the next sampling time slot, a missing status is written to the corresponding reading position, and no other reading type is used as a substitute.

[0022] S102: The sampling time slot is determined based on the sampling time and sampling period configuration fields in the sensor message. The sampling period configuration field is stored in the node configuration area of ​​the oxygen chamber local controller and is uniformly distributed during the oxygen chamber control strategy update process. Messages with sampling times within the same time slot boundary are grouped into the same time slot message record. The time slot message record is indexed according to the node identifier and associated with the message arrival time, retransmission count field, and reading status. When a sampling time is missing, the most recent valid clock synchronization status of the IoT node is read; when the synchronization status is valid, the time slot assignment is reconstructed according to the message sequence identifier and the adjacent valid sampling time, and marked as timing repair status; when the synchronization status fails, the message is marked as timing invalid status, triggering a node clock calibration request.

[0023] S103: Extract the sampling time, radio message arrival time, and timeslot start identifier of each message from the timeslot message record to form a transmission delay state. The transmission delay state describes the timing offset of a message from the formation of the node sampling record to the confirmation of reception by the oxygen chamber local controller. The delay judgment boundary is derived from the wireless communication protocol configuration table, the node transmit queue configuration, and the oxygen chamber local controller receive buffer state. The transmission delay state is mapped to normal delay, acceptable delay, and failure delay levels, and the level mapping is performed according to the boundary order in the protocol configuration table. During receive buffer congestion or wireless channel recovery, the delay judgment boundary calls the configuration item corresponding to the current communication state. After communication recovery, it switches back to the stable operation configuration to avoid uniformly judging consecutive messages during the communication recovery phase as failures.

[0024] S104: Aggregate message records with the same node identifier within the same sampling time slot, read the retransmission count field and arrival order status of each message, and form a retransmission reliability level. The retransmission judgment boundary is derived from the IoT node communication driver configuration and the oxygen chamber local controller's receiving policy table. When the retransmission count is in the allowed state and the message integrity passes the verification, it is retained as a candidate. When the retransmission count exceeds the allowed state, it is marked as a communication instability candidate. When multiple candidate messages carry the same sampling time and the same reading content, the record with the higher arrival delay level and the higher retransmission reliability level is retained. When the reading content is inconsistent, the continuity of the node sequence identifier and the sensor self-check status are compared. The record with continuous sequence and valid self-check status is retained, and the inconsistent content is written to the duplicate message conflict log.

[0025] S105: Candidate packets are sorted hierarchically according to transmission delay level, retransmission reliability level, and arrival order. Transmission delay level has the highest comparison priority. Retransmission reliability level is compared when transmission delay levels are the same. Arrival order is compared when all the aforementioned levels are the same. When all comparison conditions are the same, the sorting remains stable according to the fixed node order in the node registry. The packet reliability sorting is an ordered index of candidate packets within the same sampling time slot. The index records do not change the original packet content, but only record the retention priority and the reason for exclusion. The screening process reads the delay judgment boundary, retransmission judgment boundary, and sensor self-test status, retains the node readings that are ranked higher and meet all valid conditions, and marks the remaining duplicate packets of the same node as redundant.

[0026] S106: Align the retained readings according to the sampling time slot and node identifier to form a reliable sampling sequence for the oxygen chamber. When the equipment starts up, no historical time slots have been formed yet. The first sampling time slot only performs message integrity, node registration status, and sensor self-test status verification, and writes the valid readings into the initial reliable record. Subsequent time slots gradually add delay levels and duplicate message consistency checks. When a certain type of reading is temporarily missing, the missing state is retained in the reliable sampling sequence for the oxygen chamber, and the most recent valid reading is used as a continuity reference. The most recent valid reading does not replace the current reading in the formation of control quantities. Continuous missing states trigger node communication anomaly flags and corresponding valve control freeze flags. After the sensor self-test recovers and the continuous messages meet the reliable conditions again, the freeze flag is cleared, and the continuous sequence is re-established starting from the recovered reliable readings.

[0027] S2: Read the chamber pressure reading, oxygen concentration reading, pipeline flow rate reading, mask differential pressure reading, and blood oxygen reading from the same sampling time slot in the reliable sampling sequence of the oxygen chamber. Multiply the chamber pressure reading and oxygen concentration reading, and compare the direction and magnitude of change in adjacent sampling time slots to form a pressure-oxygen coupling value. Simultaneously, compare the directional relationship and continuity of the mask differential pressure reading and pipeline flow rate reading within continuous sampling time slots to form the wearing and sealing status. Then, according to the sampling time, associate the pressure-oxygen coupling value and the wearing and sealing status to generate a chamber-patient linkage status sequence. The chamber-patient linkage status sequence is a status record arranged according to the sampling time. Each status record carries the pressure-oxygen coupling value, wearing and sealing status, blood oxygen reading, valid reading status, and communication anomaly status, serving as input for stable segment selection.

[0028] S201: Extract chamber pressure and oxygen concentration readings from the reliable sampling sequence of the oxygen chamber, and check whether the corresponding records belong to the same sampling time slot and whether the reliable status is valid. The chamber pressure reading comes from the chamber pressure acquisition node, and the oxygen concentration reading comes from the in-chamber oxygen concentration acquisition node. The two readings retain their respective physical acquisition calibers and are not directly compared across dimensions. The product processing converts the two readings into the same pressure-oxygen coupling evaluation caliber. The output includes the coupling result of the current sampling time slot, the direction of change relative to the previous valid sampling time slot, and the change amplitude level. The boundary of the change amplitude level comes from the oxygen chamber equipment calibration record, treatment control strategy table, and sensor range status. The boundary update is written to the configuration area along with the equipment maintenance calibration results. When any reading is in a missing state or the sensor is in a distorted state, no new valid pressure-oxygen coupling value is formed. The status record inherits the previous valid change direction and adds a current invalid mark.

[0029] S202: Extract the mask differential pressure reading and tubing flow reading from the continuous sampling time slots, and compare the changes in adjacent readings with the fluctuation dead zone field. The fluctuation dead zone field describes the negligible fluctuation boundary formed by the sensor resolution, tubing pulsation, and wearer breathing disturbances. It is derived from the mask differential pressure sensor calibration record, flow sensor calibration record, and oxygen chamber control strategy table, and is updated according to the sensor maintenance status. Changes within the fluctuation dead zone are marked as stable directions, changes exceeding the fluctuation dead zone and showing an increasing trend are marked as rising directions, and changes showing a decreasing trend are marked as falling directions, thus forming mask differential pressure direction records and tubing flow direction records. When a sudden change in reading is accompanied by a sensor self-test anomaly, the direction of change is marked as an invalid direction; when a sudden change in reading is not accompanied by a self-test anomaly, the direction record is retained and a transient disturbance flag is written, which is then filtered by the continuous status judgment.

[0030] S203: Align the mask pressure differential direction record and pipeline flow direction record according to the sampling time, and compare the directional status within the same sampling time slot. If the directions are the same or both are stable directions, continue the same-direction continuous state; if the directions are opposite, either direction is invalid, or the sampling time slots are discontinuous, interrupt the same-direction continuous state and record the reason for the interruption. The transient disturbance indicator is only interrupted when the continuous state has not reached the airtightness judgment boundary. When the continuous state has stabilized and subsequent readings have returned to the original direction, retain the previous valid airtightness state and write the transient disturbance into the status supplementary field. When there is a lack of continuous direction history during the equipment startup phase, set the first valid direction record as the starting point of the continuous state, and output the formal state after a continuous record that meets the airtightness judgment rules is formed.

[0031] S204: Compare the continuous state in the same direction with the closed continuous boundary field, and simultaneously check whether the mask differential pressure reading is within the closed differential pressure reference range. The closed continuous boundary field and the closed differential pressure reference range are derived from the mask structural parameters, wearing interface calibration records, and oxygen chamber control strategy table. The configuration content is called according to the mask model and sensor maintenance status. When the continuous boundary is met in the continuous direction and the mask differential pressure is within the reference range, the closed state is output; when the continuous boundary is not met, the mask differential pressure exceeds the reference range, or the differential pressure sensor is invalid, the non-closed state is output; when communication is briefly interrupted and the previous valid state is the closed state, the closed state to be confirmed is output and the valve incremental expansion operation is frozen. The wearing closed state record also carries the state source, state update time, and abnormal reason, for subsequent linkage state synthesis and closed-loop control judgment and calling.

[0032] S205: A linked status item is formed by associating the pressure oxygen coupling value, the wearer's airtightness status, and the blood oxygen reading at the sampling time. The linked status item is a combined record of the cabin environment and the wearer's physiological readings within the same sampling time slot. Fields include the sampling time slot identifier, pressure oxygen coupling value, direction of coupling change, wearer's airtightness status, blood oxygen reading, the validity status of each field, and the control freeze status. The status item is written into the cabin-patient linked status sequence according to the sampling time. When a valid blood oxygen reading is missing in a sampling time slot, the blood oxygen missing identifier is retained, while the pressure oxygen coupling value and wearer's airtightness status are still written as valid. After the blood oxygen reading recovers, the status association continues to be established from the recovery time slot, without supplementing the estimated reading for historically missing time slots. The cabin-patient linked status sequence is written to a local circular cache, and abnormal states are synchronously written to the traceability log.

[0033] S3: Perform continuity and stability checks on the cabin-patient linkage state sequence, selecting state items where changes in pressure oxygen coupling values, blood oxygen readings, and the wearing of a sealed suit all meet the stability judgment rules, forming stable segments. For the pressure oxygen coupling values ​​and blood oxygen readings in the stable segments, establish benchmarks and convert them to a unified proportional caliber. Then, read the median difference and adjacent sampling slopes between the proportionalization results to form a dynamic control boundary that updates with the operating status. Compare the current proportionalization results of the pressure oxygen coupling values ​​and blood oxygen readings with the dynamic control boundary to determine the offset direction and offset amplitude level, generating a control offset value. The control offset value is a state field that can be directly invoked by the control rules, carrying the offset direction, offset amplitude level, boundary version, and data validity status.

[0034] S301: Read the continuous state items in the cabin-patient linkage state sequence, and check the change level of the pressure oxygen coupling value, the change level of the blood oxygen reading, the continuity of the wearing closure state, and the continuity of the sampling time slot. The stability determination rules are stored in the control strategy table. The rule boundaries are derived from the oxygen chamber equipment calibration records, the historical effective sequences after the wearer enters the stable oxygen supply stage, and the sensor noise status. When both the pressure oxygen coupling change and the blood oxygen change are in the stable category, the wearing closure state is consistent, and the sampling time slot is continuous, the corresponding state item is assigned to the current stable segment; the current segment ends when any condition changes to the unstable category. The pending closure state and the control freeze state are not included in the stable segment to avoid communication anomalies or wearing anomalies changing the dynamic boundary. The stable segment records the start and end time slots, the status of the effective fields, and the boundary candidate identifier for reference processing.

[0035] S302: Extract the pressure-oxygen coupling numerical sequence and blood oxygen reading sequence from the stable segment, arrange them according to numerical order, and determine the robust benchmark corresponding to the intermediate position. When the sequence sorting result corresponds to a single intermediate position, directly read the data of the intermediate position; when it corresponds to adjacent intermediate positions, read the average state of the adjacent intermediate data as the benchmark. The pressure-oxygen coupling benchmark and the blood oxygen benchmark are stored separately in the current stable segment record, and the two benchmarks do not directly replace each other. Scale each pressure-oxygen coupling value within the segment relative to the pressure-oxygen coupling benchmark, and scale each blood oxygen reading relative to the blood oxygen benchmark, outputting a stable proportional sequence with a dimensionless change caliber. When the benchmark is invalid or the effective record of the segment is insufficient, stop the boundary update of the current segment and use the previous effective dynamic control boundary.

[0036] S303: Align the pressure oxygen ratio and blood oxygen ratio states in the stable ratio sequence according to the sampling time to form a ratio difference record for the same time slot. The ratio difference record describes the direction and relative degree of deviation between the changes in pressure oxygen and blood oxygen in the cabin under a unified caliber. Arrange all ratio difference records in order of magnitude, and determine the median difference value according to the central position rule to reduce the impact of transient abnormal records on the boundary center. Read the ratio state changes and the continuous state of the time slots in adjacent sampling time slots to form adjacent sampling slope records; if the time slot is interrupted or any ratio state is invalid, no slope record is formed. The slope records are divided into rising, stable, and falling states according to the direction of change, and mapped to slope levels according to the degree of change. The mapping boundary comes from the historical distribution of stable segments and the control strategy table.

[0037] S304: A dynamically adjustable boundary is established by using the deviation position corresponding to the median difference as the boundary center, the robust intermediate state of the absolute change in adjacent sampling slopes as the source of the boundary width, and adjusting the movement direction of the boundary center according to the current slope direction. The dynamically adjustable boundary records the upper boundary state, lower boundary state, boundary center, boundary width level, update slot, and source segment identifier. When the slope is increasing, the boundary center moves along the corresponding direction of change; when the slope is decreasing, the boundary center moves along the opposite direction of change; when the slope is stable, the boundary center is maintained and only the boundary width state is updated. If the boundary width source record is insufficient, a stable segment is interrupted, or the wearing closure state changes to a non-closure state, boundary updates are paused and the previous valid boundary version is retained.

[0038] S305: Reads the pressure oxygen coupling value and blood oxygen reading of the current sampling time slot, calls the most recent effective pressure oxygen coupling reference and blood oxygen reference for proportionalization, and then compares them with the current effective dynamic control boundary. When the current proportional state is within the boundary, it outputs the state within the boundary and the direction of no offset; when it exceeds the upper boundary, it outputs the upper offset direction; when it exceeds the lower boundary, it outputs the lower offset direction. The offset amplitude level is mapped according to the relative distance after the current proportional state crosses the boundary. The level boundary comes from the control strategy table and maintains a fixed correspondence with the valve opening change level. When there is a directional conflict between the pressure oxygen proportional state and the blood oxygen proportional state, it first checks the wearing closure status and the effective blood oxygen acquisition status. If the wearing is not closure or the blood oxygen is invalid, it outputs the control freeze status. In other cases, it calls the offset conflict rule and retains the offset level with more limited control amplitude.

[0039] S306: Write the offset direction, offset amplitude level, dynamic control boundary version, wearing closure status, sensor reading validity status, and control freeze status into the control offset value record. During the equipment startup phase, before a stable segment is formed, the initial control boundary in the oxygen chamber equipment calibration record is invoked. The initial boundary is only used for comparison after chamber pressure, oxygen concentration, blood oxygen, and wearing closure status have all passed validity checks. After a stable segment is formed, switch to the dynamic control boundary, retaining the time sequence records of the initial boundary version and the dynamic boundary version during the switch. If the dynamic boundary record is corrupted or configuration reading fails, restore the boundary version that most recently passed integrity verification and write it into the boundary recovery status. Subsequent closed-loop control is executed according to the restricted opening change level.

[0040] S4: Read the recorded control offset values, determine the oxygen supply valve opening increment, oxygen exhaust valve opening increment, and node reporting interval according to the offset direction and magnitude, and adjust the reporting frequency of the oxygen supply valve, oxygen exhaust valve, and IoT node. After control is executed, re-collect the pressure oxygen coupling values ​​to form the actual change direction, and compare it with the expected change direction corresponding to the control offset values. If the directions are consistent, retain the current control execution parameters; if the directions are opposite, cancel the current control execution parameters. When the actual change is in a stable state, reduce the valve opening increment and shorten the node reporting interval, and output a closed-loop control command. The closed-loop control command carries the valve target state, node reporting configuration, execution validity period, boundary version, direction consistency status, and abnormal feedback path.

[0041] S401: Reads the offset direction, offset amplitude level, wearing airtight state, and control freeze state from the control offset value, and calls the valve control mapping table to determine the opening adjustment direction. The valve control mapping table stores the correspondence between the offset direction and the actions of the oxygen supply valve and oxygen exhaust valve. The table entries are derived from the flow structure of the oxygen supply pipeline and oxygen exhaust pipeline, valve actuator response calibration records, and oxygen chamber control strategy. When the offset direction requires increasing the oxygen pressure state, the oxygen supply valve enters the increasing opening direction, and the oxygen exhaust valve enters the decreasing opening direction; when the offset direction requires decreasing the oxygen pressure state, the oxygen supply valve enters the decreasing opening direction, and the oxygen exhaust valve enters the increasing opening direction; when there is no offset direction, the current valve target state is maintained. When the wearing airtight state is non-airtight or the control freeze state is effective, the oxygen supply expansion action is stopped, and the wearing check control status is output.

[0042] S402: Based on the offset magnitude level, the oxygen supply valve opening change level and the oxygen exhaust valve opening change level are read from the valve control mapping table. Simultaneously, the corresponding data acquisition density status is read from the node reporting strategy table, forming the oxygen supply valve opening increment, the oxygen exhaust valve opening increment, and the node reporting interval. The closer the offset magnitude level is to the boundary state, the more restricted the opening change level; the further the offset magnitude level deviates from the dynamic control boundary, the denser the node reporting interval configuration becomes. The node reporting strategy table is jointly defined by wireless communication capacity, node power status, receive buffer status, control offset magnitude level, and control freeze status. When wireless communication is congested, priority is given to reporting chamber pressure, oxygen concentration, mask pressure difference, and blood oxygen reports. Pipeline auxiliary status reports maintain their original configuration and are written with a scheduling delay flag.

[0043] S403: Compare the oxygen supply valve opening increment and the oxygen exhaust valve opening increment with the valve opening range field. The valve opening range field is derived from the valve actuator equipment parameters, installation and commissioning records, and the current fault isolation status. If the increment causes the target opening to exceed the allowable boundary, the target opening is corrected to the corresponding boundary, and the boundary correction status is written into the control execution parameters. When there are interlock restrictions between the oxygen supply valve and the oxygen exhaust valve, the actions are issued in stages according to the action priority specified in the oxygen chamber control strategy table. The next valve action is issued only after the previous valve action is confirmed. The node reporting interval is sent to the IoT node through the node configuration message. After the node returns the configuration confirmation status, the new reporting strategy is enabled; if no confirmation status is returned, the previous valid configuration is maintained and the communication retry record is triggered.

[0044] S404: Combine the corrected target status of the oxygen supply valve, the target status of the oxygen exhaust valve, the node reporting interval, the dynamic control boundary version, and the control offset value identifier into control execution parameters and write them to the execution queue. Before sending the action command, the valve control interface checks the actuator's online status, current position feedback status, and fault lockout status. After passing the checks, it sends the action command and waits for position feedback. If the position feedback matches the target status, it marks the execution as confirmed; if the position feedback is not updated or returns to a fault status, it marks the execution as failed, stops subsequent incremental actions, and maintains the actuator's current confirmed position. Control execution parameters, sending status, feedback status, and abnormal reasons are written to the control traceability log. The log records are provided to the maintenance interface for reading according to device access permissions.

[0045] S405: Determine the expected change direction of the oxygen pressure coupling value based on the control offset direction and the valve opening adjustment direction. Control combinations that increase oxygen pressure correspond to the expected upward direction, control combinations that decrease oxygen pressure correspond to the expected downward direction, and control combinations that maintain oxygen pressure correspond to the expected stable direction. The expected change direction is recorded and linked to the control execution parameter identifier and execution confirmation status. A direction consistency judgment is entered only when both the oxygen supply valve and the oxygen exhaust valve receive valid feedback. If the valve execution or feedback is inconsistent, the expected change direction is marked as pending confirmation, and the automatic addition of opening increments based on oxygen pressure changes is paused.

[0046] S406: After the control is executed, read the pressure oxygen coupling values ​​within the preset continuous sampling window, and check the reliability, sealing, and communication status corresponding to each sampling time slot. If the change in adjacent pressure oxygen coupling values ​​exceeds the effective change boundary of the sensor and is accompanied by a self-test anomaly, it is marked as an abnormal value and excluded from the actual direction judgment record; if the change exceeds the boundary but the sensor self-test is effective, the value is retained and marked as a rapid change state. When the continuous effective records show an increasing trend, the actual upward direction is output; when they show a decreasing trend, the actual downward direction is output; when the change is in the fluctuation dead zone, the actual stable direction is output. When there are insufficient effective records, the observation state is extended and limited control is maintained, and the actual direction is not estimated by calling missing records.

[0047] S407: Compare the expected change direction with the actual change direction. If the directions are the same, retain the control execution parameters and write the current valve status into the effective control state. If the directions are opposite, cancel the control execution parameters, send a rollback command to the valve control interface to restore the effective state before execution, and increase the sensor message acquisition density. If the actual change direction is stable, adjust the oxygen supply valve opening increment and the oxygen exhaust valve opening increment to a more restricted level, while shortening the node reporting interval and continuing to observe the pressure-oxygen coupling status. If the rollback command is not confirmed by the actuator, enter the valve fault lockout state, stop automatic incremental adjustment, and output the manual inspection status. The direction judgment result, rollback status, and fault lockout status are jointly written into the closed-loop control command.

[0048] S408: Sends closed-loop control commands to the oxygen supply valve control interface, oxygen exhaust valve control interface, and IoT node configuration interface. Each interface returns action confirmation, position feedback, or configuration confirmation, respectively. When all feedback is valid, the current closed-loop cycle is completed, and the latest reliable oxygen chamber sampling sequence, chamber-hazard linkage state sequence, dynamic control boundary, and control execution parameters are used as the historical state for the next cycle. If some feedback is missing, the confirmed state is retained, and unconfirmed objects maintain the previous valid control configuration and enter the retry queue. When the oxygen chamber local controller restarts, it reads the most recently verified historical state and the actual valve position feedback. If the historical state matches the actual position, closed-loop processing continues; if they do not match, the initial control state is established based on the actual position, and the expected change direction is regenerated.

[0049] The IoT-based intelligent monitoring and control system for oxygen chambers, corresponding to the aforementioned method, includes a trusted sampling module, a linkage synthesis module, a boundary offset module, and a closed-loop control module. The trusted sampling module receives sensor messages via an IoT communication interface, handling field parsing, time slot grouping, trusted sorting, anomaly marking, and trusted sequence output for steps S1 and its sub-steps, and outputs the oxygen chamber's trusted sampling sequence to the linkage synthesis module. The linkage synthesis module handles the pressure-oxygen coupling processing, direction sequence generation, occupancy closure judgment, and state item association for steps S2 and its sub-steps, and outputs the chamber's linkage state sequence to the boundary offset module.

[0050] The boundary offset module handles the stable segment selection, scaling, median difference determination, adjacent sampling slope generation, dynamic control boundary update, and control offset status output formed by steps S3 and its sub-steps. The dynamic control boundary and control offset values ​​are transmitted to the closed-loop control module via an internal data interface. The closed-loop control module handles the valve incremental mapping, node reporting strategy selection, execution range correction, control command issuance, feedback confirmation, and direction consistency judgment formed by steps S4 and its sub-steps. It completes closed-loop feedback through the valve control interface and node configuration interface. All modules share a read-only device configuration area and a controlled state storage area. The configuration area stores node registration records, communication protocol configurations, sensor calibration records, valve control mapping tables, and control strategy tables. The state storage area stores reliable sampling states, linkage states, boundary versions, control execution states, and anomaly traceability records. Modules only transmit data objects that have completed state verification. When any module outputs an invalid state, the downstream module processes it according to the corresponding freeze rule, inheritance rule, or recovery rule, directly driving the valve action without bypassing data validity checks.

[0051] The above embodiments illustrate preferred embodiments of the present invention. Any equivalent adjustments to the technical solution based on software engineering methods are within the scope of protection, including but not limited to: implementing algorithm logic using different programming languages, refactoring functional modules into services, adjusting data interaction protocols, and optimizing resource scheduling strategies. Any implementation scheme derived from reasonable modifications to the data processing flow, service call chain, or system architecture layer without departing from the core technology of the present invention should be considered within the scope of protection defined by the claims of the present invention.

Claims

1. An IoT-based intelligent monitoring and control method for oxygenated chambers, characterized in that: Includes the following steps: By deploying IoT nodes in the oxygen chamber, oxygen supply pipeline, oxygen exhaust pipeline, and mask wearing area, the chamber pressure reading, oxygen concentration reading, pipeline flow reading, mask differential pressure reading, blood oxygen reading, and wireless message arrival time are directly obtained. The readings within the same sampling time slot are analyzed according to the wireless message arrival time and the number of node retransmissions to generate a reliable sampling sequence for the oxygen chamber. Based on the change of the product of chamber pressure reading and oxygen concentration reading in the oxygen chamber reliable sampling sequence, the pressure-oxygen coupling value is calculated. Based on the duration of the same direction of the mask pressure difference reading and the pipeline flow reading, the wearing and sealing status is determined. The pressure-oxygen coupling value and the wearing and sealing status are combined into a chamber-patient linkage status sequence. Stable segments in the cabin-patient linkage state sequence are screened, and a dynamic control boundary is constructed using the median difference of the stable segments and the adjacent sampling slope. The current pressure oxygen coupling value, blood oxygen reading and the dynamic control boundary are compared to generate a control offset value. The oxygen supply valve opening increment, oxygen exhaust valve opening increment, and node reporting interval are calculated based on the control offset value. The reporting frequency of the oxygen supply valve, oxygen exhaust valve, and IoT node is adjusted, and a closed-loop control command is output based on the directional consistency between the oxygen pressure coupling value and the control offset value.

2. The IoT-based intelligent monitoring and control method for oxygenated chambers according to claim 1, characterized in that, The process of generating the reliable sampling sequence of the oxygen chamber specifically includes: Collect the node identifier, sampling time, arrival time of the wireless message, number of retransmissions of the node and corresponding readings sent by the IoT node, and classify the corresponding readings into the same sampling time slot according to the sampling time and the preset sampling period, and establish a time slot message record according to the node identifier; Compare the arrival delay between the arrival time of each wireless message in the time slot message record and the start time of the corresponding sampling time slot. Prioritize arranging them according to the arrival delay from smallest to largest. If the arrival delays are the same, arrange them according to the number of retransmissions of the node from smallest to largest to generate a reliable message sort. The corresponding readings that meet the preset delay range and preset retransmission count limit in the trusted order of the messages are selected. For repeated wireless messages from the same node, the corresponding readings with the earlier trusted order are retained and aligned according to the sampling time slot and node identifier to generate a trusted sampling sequence for the oxygen chamber.

3. The IoT-based intelligent monitoring and control method for oxygenated chambers according to claim 1, characterized in that, The generation process of the cabin-disease linkage state sequence specifically includes: Read the chamber pressure reading and the oxygen concentration reading in the same sampling time slot of the oxygen chamber reliable sampling sequence, multiply the chamber pressure reading and the oxygen concentration reading, compare the direction and magnitude of change of the product result of adjacent sampling time slots, and generate a pressure-oxygen coupling value. Extract the mask pressure difference reading and the pipeline flow reading in the same sampling time slot, determine whether the increase or decrease direction of the mask pressure difference reading and the pipeline flow reading is consistent in the continuous sampling time slot, accumulate the duration of the same direction, and compare the duration of the same direction with a preset sealing duration threshold to obtain the wearing sealing state; The pressure oxygen coupling value is associated with the sampling time and the wearing airtight state. A linkage state item is established based on the value change and state change within the continuous sampling time slot. The linkage state items are arranged according to the sampling time to generate the cabin-disease linkage state sequence.

4. The IoT-based intelligent monitoring and control method for oxygenated chambers according to claim 1, characterized in that, The process of generating the adjustment offset value specifically includes: The variation range of the pressure oxygen coupling value and blood oxygen reading in the cabin-patient linkage state sequence between consecutive state items is detected, and the wearing airtight state corresponding to the consecutive state item is read. State items in which both types of variation range are within a preset stable range and the wearing airtight state is continuous and consistent are divided into stable segments. Calculate the median of the pressure-oxygen coupling value and the median of the blood oxygen reading within the stable segment. Scale the data based on the median of the pressure-oxygen coupling value and the median of the blood oxygen reading. Determine the boundary center and boundary width based on the median difference of the scaling results and the adjacent sampling slope to construct a dynamic control boundary. The positional relationship between the current pressure oxygen coupling value and the proportionalized result of the blood oxygen reading and the dynamic control boundary is compared. The offset direction and offset magnitude are determined based on the direction and distance beyond the boundary, and the control offset value is generated.

5. The IoT-based intelligent monitoring and control method for oxygenated chambers according to claim 1, characterized in that, The process of generating the closed-loop control command specifically includes: Read the offset direction and offset magnitude of the control offset value, determine the opening adjustment direction of the oxygen supply valve and the oxygen exhaust valve according to the offset direction, match the opening change level and data acquisition density according to the offset magnitude, and calculate the oxygen supply valve opening increment, the oxygen exhaust valve opening increment and the node reporting interval. The oxygen supply valve opening increment and the oxygen exhaust valve opening increment are compared with the preset valve opening range. The increment that causes the valve opening to exceed the preset valve opening range is corrected to the corresponding range boundary. The reporting frequency of the Internet of Things node is adjusted according to the node reporting interval to establish the control execution parameters. The pressure-oxygen coupling value after the control is executed is collected, and the direction of the change of the value before and after the control is executed is compared with the direction of the offset of the control offset value. The consistency of the direction is determined. If the direction is consistent, the control execution parameter is retained. If the direction is opposite, the control execution parameter is cancelled. If the direction of the change of the value is stable, the control execution parameter is reduced, and a closed-loop control command is output.

6. The IoT-based intelligent monitoring and control method for oxygenated chambers according to claim 2, characterized in that, The process of generating the trusted message ordering specifically includes: Extract the sampling time, arrival time and sampling time slot start time of each wireless message from the time slot message record, calculate the transmission delay of the arrival time relative to the sampling time, and determine the delay confidence level according to the proportion of the transmission delay to the preset sampling period. The number of wireless packets with the same node identifier in the same sampling time slot and the number of times the node retransmits are counted. The difference between the number of times the node retransmits and the preset upper limit of the number of retransmissions is compared, and the retransmission reliability level is determined by combining the arrival order of the duplicate wireless packets. According to the hierarchical order of the delay trust level, the retransmission trust level, and the arrival order, the wireless packets in the same sampling time slot are compared level by level. For wireless packets with the same trust conditions, a fixed order is maintained according to the node identifier to generate a packet trust order.

7. The IoT-based intelligent monitoring and control method for oxygenated chambers according to claim 3, characterized in that, The process for determining the airtight wearing condition specifically includes: Acquire the mask pressure differential reading and the pipeline flow reading within a continuous sampling time slot, compare the changes in adjacent readings with a preset fluctuation dead zone, mark the changes within the preset fluctuation dead zone as the stable direction, and mark the changes outside the preset fluctuation dead zone as the rising or falling direction, and generate a reading direction sequence. Compare the mask pressure difference reading direction and the pipeline flow reading direction corresponding to the same sampling time slot in the reading direction sequence. Accumulate the same-direction duration for sampling time slots with the same direction or both being stable directions, and clear the same-direction duration for sampling time slots with opposite directions to obtain a closed continuous record. The continuous sealing record is compared with the preset continuous sealing threshold. The mask pressure difference reading is detected to see if it is within the preset sealing pressure difference range. When the continuous threshold is reached and the pressure difference is within the preset sealing pressure difference range, the sealing state is determined. When the continuous threshold is not reached or the pressure difference is exceeded, the non-sealing state is determined, thus obtaining the wearing sealing state.

8. The IoT-based intelligent monitoring and control method for oxygenation chambers according to claim 4, characterized in that, The process of constructing the dynamic control boundary specifically includes: Obtain the pressure oxygen coupling value and blood oxygen reading aligned by sampling time within each stable segment, calculate the median of the pressure oxygen coupling value and the median of the blood oxygen reading, and scale each sampled value based on the median of the pressure oxygen coupling value and the median of the blood oxygen reading to establish a stable scaling sequence. Calculate the difference between the hyperbaric oxygen ratio and the blood oxygen ratio in the same sampling time slot in the stable ratio sequence, sort them according to the size of the difference, extract the middle position difference when the number of differences is odd, and calculate the average of the two middle differences when the number of differences is even to obtain the median difference. At the same time, compare the change of the ratio value of adjacent sampling time slots with the sampling interval to obtain the adjacent sampling slope. Calculate the boundary center corresponding to the median difference and the boundary width corresponding to the median of the absolute values ​​of the adjacent sampling slopes, and move the boundary center according to the changing direction of the current adjacent sampling slopes to establish a dynamically adjustable boundary.

9. The IoT-based intelligent monitoring and control method for oxygenated chambers according to claim 5, characterized in that, The process for determining directional consistency specifically includes: The pressure-oxygen coupling value, the control offset value, and the control execution parameters before the control are executed are obtained. Based on the offset direction of the control offset value and the opening adjustment direction of the oxygen supply valve and the oxygen exhaust valve, the expected change direction of the pressure-oxygen coupling value is determined. The pressure-oxygen coupling values ​​are collected in three consecutive sampling time slots after the control is executed. Abnormal values ​​that exceed the preset upper limit of adjacent change range are removed. The increase or decrease relationship between the retained values ​​and the values ​​before the control is executed is compared to obtain the actual change direction. The expected change direction is compared with the actual change direction. If the directions are consistent, the control execution parameters are retained. If the directions are opposite, the control execution parameters are canceled. When the actual change direction is in a stable direction, the valve opening increment is reduced and the node reporting interval is shortened, and a closed-loop control command is output.

10. An IoT-based intelligent monitoring and control system for oxygenated chambers, characterized in that: The IoT-based intelligent monitoring and control system for oxygenated chambers is used to implement the IoT-based intelligent monitoring and control method for oxygenated chambers as described in any one of claims 1-9. The IoT-based intelligent monitoring and control system for oxygenated chambers includes: The reliable sampling module acquires chamber pressure readings, oxygen concentration readings, pipeline flow readings, mask differential pressure readings, blood oxygen readings, wireless message arrival times, and node retransmission counts. It analyzes the readings within the same sampling time slot according to the wireless message arrival times and node retransmission counts to generate a reliable sampling sequence for the oxygen chamber. The linkage synthesis module calculates the pressure oxygen coupling value and determines the wearing and sealing status based on the oxygen chamber reliable sampling sequence, and synthesizes the pressure oxygen coupling value and the wearing and sealing status into a chamber-patient linkage status sequence. The boundary offset module filters out stable segments in the cabin-patient linkage state sequence, constructs a dynamic control boundary using the median difference and adjacent sampling slopes, compares the current pressure oxygen coupling value, blood oxygen reading and the dynamic control boundary, and generates a control offset value. The closed-loop control module calculates the oxygen supply valve opening increment, oxygen exhaust valve opening increment, and node reporting interval based on the control offset value, adjusts the reporting frequency of the oxygen supply valve, oxygen exhaust valve, and the IoT node, compares whether the change direction of the oxygen pressure coupling value and the control offset value is consistent, and outputs closed-loop control commands.

Images