A highland oxygen production system and method based on multi-source fusion and collaborative optimization

Abstract

The application discloses a highland oxygen production system and method based on multi-source fusion and collaborative optimization, which comprises a double-oxygen-nozzle intelligent scheduling module, a temperature self-adaptive double-loop control module, an altitude self-adaptive parameter adjustment module, a Beidou emergency rescue module and a cross-module collaborative optimization controller. The cross-module collaborative optimization controller is used for receiving state quantities and error quantities of each module in real time, so as to construct a joint cost function. The optimal control quantity of the joint cost function is solved and synchronously issued to each module for corresponding control. The application realizes the comprehensive ability of double-person simultaneous on-demand oxygen supply, 0-5500m free adjustment, full-dimension self-adaptation and Beidou emergency rescue without network in a portable oxygen generator for the first time, and has good practicability.

Description

Technical Field

[0001] This invention relates to the field of high-altitude medical equipment technology, and in particular to a shared portable oxygen generator capable of supplying oxygen to two users simultaneously at altitudes of 0-5500m, with temperature / altitude adaptive control and BeiDou emergency rescue capabilities, as well as a high-altitude oxygen generation method based on multi-source fusion and synergistic optimization. Background Technology

[0002] Existing portable oxygen concentrators exhibit the following systemic shortcomings in high-altitude shared environments:

[0003] a) Battery life bottleneck caused by integrating the battery with the main unit

[0004] Traditional products use integrated battery packs (18650 or polymer lithium batteries) with fixed capacity and no hot-swappable design. High altitude and low temperature (below -10℃) increase the battery's internal resistance by 30%-50% and reduce its effective capacity by 25%-40%. At the same time, the peak current in pulse oxygen supply mode can reach 8-12A, further shortening the range to less than 40 minutes, which is difficult to meet the needs of long-term activities such as high-altitude hiking and convoy crossings.

[0005] b) The oxygen supply mode is singular and cannot provide differentiated oxygen supply for two people.

[0006] Most mainstream models on the market only support two fixed curves: "continuous flow" or "fixed pulse". When two users (adult vs. child) are inhaling oxygen at the same time, the flow rate and pulse frequency cannot be adjusted independently according to the differences in their tidal volume and respiratory rate, resulting in the phenomenon of "one person over-inhaling and the other under-inhaling". The maximum difference in SpO2 measured in actual tests can reach 8%.

[0007] c) In high-altitude environments, parameters need to be manually adjusted, which is cumbersome and prone to missetting.

[0008] For every 1000m increase in altitude, atmospheric pressure decreases by approximately 12%, and the partial pressure of oxygen decreases accordingly. Existing equipment requires users to manually set flow rate, concentration, and pulse frequency step by step using knobs or buttons, which relies on experience and is prone to misoperation in hypoxic environments. Experiments show that the manual setting error rate can reach 22% at an altitude of 4200m, and 15% of users forget to recalibrate within 30 minutes.

[0009] d) Lack of location alarm function required for shared scenarios

[0010] Location tracking missing: Existing oxygen concentrators lack alarm location tracking functionality;

[0011] Communication gap: No cellular / satellite communication link, unable to report SOS;

[0012] The aforementioned deficiencies result in serious shortcomings in the reliability, usability, and commercial sustainability of existing products in scenarios such as high-altitude sharing, emergency rescue, and long-term operation, necessitating systemic innovation. Summary of the Invention

[0013] A brief overview of embodiments of the invention is provided below to provide a basic understanding of certain aspects of the invention. It should be understood that this overview is not an exhaustive summary of the invention. It is not intended to identify key or essential parts of the invention, nor is it intended to limit the scope of the invention. Its purpose is merely to present certain concepts in a simplified form as a prelude to the more detailed description that follows.

[0014] The purpose of this invention is to overcome the above-mentioned defects and provide a high-altitude oxygen generation system and method based on multi-source fusion and collaborative optimization. The implementation scheme includes four core modules: a dual-oxygen nozzle intelligent scheduling module, a Beidou emergency rescue module, a temperature adaptive dual-loop control module, an altitude adaptive parameter adjustment module, and a cross-module collaborative optimization controller. Through cross-module joint state estimation, multi-objective collaborative optimization, and cloud-based closed-loop calibration, the originally independently operating modules are integrated into one, forming a complete technical solution of "perception-prediction-decision-execution-learning", which improves the reliability, ease of use, and commercial sustainability of oxygen generators in scenarios such as high-altitude sharing, emergency rescue, and long-term operation.

[0015] According to a first aspect of this application, a high-altitude oxygen production system based on multi-source fusion and synergistic optimization is provided, comprising:

[0016] The dual oxygen outlet intelligent scheduling module is used to dynamically switch between pulse / fixed frequency oxygen supply modes based on negative pressure thresholds fa, fb, fc and preset sliding time window filtering; fa, fb, fc correspond to three statistical thresholds of the instantaneous negative pressure peak generated during nasal inhalation and at the oxygen outlet interface of the oxygen concentrator.

[0017] Temperature adaptive dual-loop control module, including molecular sieve constant temperature PID, Smith predictive controller and compressor anti-overheating and anti-saturation PID controller;

[0018] The altitude adaptive parameter adjustment module uses the exponential model P(h)=P0×e -h / 8431 and gain-controlled dual-closed-loop PID maintains flow error ≤5%;

[0019] The Beidou emergency rescue module is used to send location and physiological parameters in short message format and perform A* path planning when the high altitude danger index (HRI) is ≥1.5.

[0020] A cross-module collaborative optimization controller is used to execute:

[0021] a) Receive in real time the status and error quantities of the dual oxygen nozzle intelligent scheduling module, temperature adaptive dual-loop control module, altitude adaptive parameter adjustment module and Beidou emergency rescue module;

[0022] b) Based on the state and error quantities of the dual oxygen outlet intelligent scheduling module, temperature adaptive dual-loop control module, altitude adaptive parameter adjustment module and Beidou emergency rescue module, a joint cost function J_total is constructed with the user oxygen demand satisfaction, power consumption, compressor thermal safety margin and positioning-communication reliability as objectives.

[0023] c) Solve for the optimal control quantity of the joint cost function J_total and simultaneously send it to the dual oxygen outlet intelligent scheduling module, temperature adaptive dual-loop control module, altitude adaptive parameter adjustment module and Beidou emergency rescue module;

[0024] d) The weight matrix and constraint boundary of the joint cost function are periodically updated through the cloud parameter learning channel.

[0025] As a feasible solution, the cross-module collaborative optimization controller receives in real-time the status and error quantities of the dual oxygen nozzle intelligent scheduling module, the temperature adaptive dual-loop control module, the altitude adaptive parameter adjustment module, and the Beidou emergency rescue module. Specifically, it receives in real-time temperature error e, compressor overheat margin δ, flow error e_L, concentration error e_C, and positioning residual. ε and HRI value; where temperature error e is obtained by molecular sieve isothermal PID and Smith predictive controller; compressor overheat margin δ is obtained by compressor anti-overheating and anti-saturation PID controller; flow error e_L is obtained by altitude adaptive parameter adjustment module; concentration error e_C is obtained by altitude adaptive parameter adjustment module; positioning residual ε Data obtained from the BeiDou emergency rescue module; HRI (Hypoxia Risk Index) value is the physiological risk index.

[0026] As a feasible solution, the joint cost function J_total constructed in the cross-module collaborative optimization controller is expressed as follows:

[0027] Joint cost function J_total=w1×|e_L|+w2×|e_C|+w3×|δ|+w4×|HRI|+w5×P_in+w6× ε Among them, w1, w2, w3, w4, w5, and w6 are weight values, which are periodically updated by the cloud; P_in is the total electrical power that the oxygen concentrator draws from the power source in real time.

[0028] As a feasible solution, in the cross-module collaborative optimization controller, the optimal control quantity for solving the joint cost function J_total is specifically solved online using a rolling time-domain optimization algorithm; where u is the real-time vector control quantity, defined as [s, y, D1, D2, n, T_cycle]. T; s is the fan speed, y is the compressor duty cycle, D1 is the solenoid valve duty cycle of one of the dual oxygen outlets, D2 is the solenoid valve duty cycle of the other oxygen outlet, n is the compressor reference speed correction amount, and T_cycle is the molecular sieve switching cycle correction amount.

[0029] As a feasible solution, in the cross-module collaborative optimization controller, the constraint boundary refers to a set of real-time adjustable upper and lower limit intervals that the collaborative optimization controller must simultaneously satisfy when solving for the optimal control quantity u; the constraint boundary is periodically updated by OTA in the cloud to ensure that the system is both safe and efficient throughout its entire life cycle; the constraint boundary includes the upper and lower limits of compressor safety T1, temperature safety T2, flow rate L, concentration C, fan speed s, and compressor duty cycle y.

[0030] As a feasible solution, in the cross-module collaborative optimization controller, the temperature error e is obtained by subtracting the target 45°C from the real-time sampling of the molecular sieve temperature sensor t2.

[0031] The compressor overheat margin δ is calculated by subtracting the real-time temperature from the safety threshold of 85℃ after the compressor temperature sensor t1 samples the temperature.

[0032] The flow error e_L is obtained by subtracting the flow value measured in real time by the air flow sensor L from the set flow rate of 3L / min;

[0033] The concentration error e_C is obtained by subtracting 90% of the set concentration from the concentration value measured in real time by the oxygen concentration sensor n;

[0034] Positioning residual ε It is calculated from the difference between the position estimate output by the Beidou positioning module and the Kalman filter prediction value;

[0035] HRI is the high altitude risk index, HRI = 0.6 × [(100 - SpO2) / 10] + 0.4 × [(RR - 20) / 10]; SpO2 is blood oxygen saturation (%), and RR is respiratory rate, in breaths / min.

[0036] As a feasible approach, the weights in the joint cost function are preset to: w1 = 0.35, w2 = 0.25, w3 = 0.15, w4 = 0.15, w5 = 0.05, w6 = 0.05.

[0037] As a feasible solution, in the cross-module collaborative optimization controller, the negative pressure thresholds fa = -45.0 ± 2.5 Pa, fb = -90.0 ± 3.0 Pa, and fc = -135.0 ± 3.5 Pa.

[0038] As a feasible solution, in the cross-module collaborative optimization controller, the Smith predictor of the temperature adaptive dual-loop control module has a pure time delay T = 8s and a time constant T = 25s.

[0039] According to a second aspect of this application, a high-altitude oxygen production method based on multi-source fusion and synergistic optimization is provided, comprising:

[0040] S0: System power-on self-test, loading the weight matrix w and constraint boundary B recently issued from the cloud;

[0041] S1: Execute S1a-S1d in parallel with a period of 100ms.

[0042] S1a: Intelligent scheduling sub-process for dual oxygen outlet nozzles;

[0043] S1b: Temperature adaptive dual-loop control sub-process;

[0044] S1c: Altitude adaptive parameter adjustment subprocess;

[0045] S1d: Beidou emergency rescue sub-process;

[0046] S2: The cross-module collaborative optimization controller collects the state variables and error variables of S1a-S1d, constructs the joint cost function J_total, and solves the optimal control variable u online;

[0047] S3: Broadcast u to each actuator at once via CAN-FD bus to achieve synchronized operation of the four modules;

[0048] S4: Update w and B via cloud OTA every 24 hours and return to S0.

[0049] As a feasible solution, the S1a dual-oxygen nozzle intelligent scheduling sub-process specifically includes:

[0050] a1) The micro differential pressure sensor (SDP800) samples the negative pressure f1 and f2 of the nozzle at 200Hz;

[0051] a2) Perform a 100ms sliding median filter to obtain f1 and f2;

[0052] a3) Compare with the preset thresholds fa, fb, and fc to determine the respiratory states S1 and S2;

[0053] a4) Based on the duty cycles D1 and D2 of the output solenoid valve of the LUT, the pulse / fixed frequency switching is realized;

[0054] a5) If an abnormal negative pressure is detected within 30 seconds, a module-level alarm will be triggered and a fault code will be uploaded;

[0055] The S1b temperature adaptive dual-loop control sub-process specifically includes:

[0056] b1) Collect the molecular sieve temperature t2 and the compressor temperature t1;

[0057] b2) Calculation error e = t2 - 45℃, margin δ = 85℃ - t1;

[0058] b3) A Smith predictive controller (τ = 8s, T = 25s) and an anti-saturation PID controller are used to calculate the fan speed s and the compressor duty cycle y;

[0059] b4) If δ < 5℃, immediately force y = 30% and fan speed to full speed until δ ≥ 10℃;

[0060] The S1c altitude adaptive parameter adjustment subprocess specifically includes:

[0061] c1) Read the altitude and barometric pressure sensor h, flow sensor L, and oxygen concentration sensor n;

[0062] c2) Using the exponential model P(h)=P0×e (-h / 8431) Calculate the feedforward reference speed n0;

[0063] c3) Calculate the flow error e_L = 3L / min - L_sensor, and the concentration error e_C = 90% - C_sensor;

[0064] c4) Gain-controlled PID outputs n and T_cycle ensure that flow error ≤ 5% and concentration fluctuation ≤ 1.5%;

[0065] c5) If the Lyapunov function fails to determine stability (V>-0.05||x||), then... 2 (Continues for 3 seconds) Enters debit mode;

[0066] The S1d Beidou emergency rescue sub-process includes:

[0067] d1) Collect blood oxygen saturation SpO2 and respiratory rate RR, and calculate HRI: HRI=0.6×(100-SpO2) / 10+0.4×(RR-20) / 10;

[0068] d2) If HRI ≥ 1.5 lasts for 5 seconds, trigger SOS;

[0069] d3) The BeiDou positioning module outputs a position estimate, which is then used to calculate the positioning residual after Kalman filtering. ε ;

[0070] d4) Assemble an 18-byte short message and send it via BeiDou RDSS;

[0071] d5) After receiving the short message, the cloud executes A* path planning and returns the rescue path.

[0072] Furthermore, to avoid misjudgment of single-mouth flow rate caused by low pressure at high altitudes, step a3) compares the respiratory state S1 and S2 with preset thresholds fa, fb, and fc, and also includes:

[0073] The real-time output P(h) of the altitude-oxygen partial pressure model (P(h) in the S1c altitude adaptive parameter adjustment subprocess) is mapped to the flow rate correction coefficient α(h): α(h)=P(h) / P0, P0=101.325kPa;

[0074] Correct the baseline flow rate Q0 in the LUT using α(h):

[0075] Q′0 = α(h)·Q0, keeping the thresholds fa, fb, and fc unchanged, but scaling the determined target flow rate level by Q′0 to avoid misjudgment of single-nozzle flow rate caused by low pressure at high altitudes. In the altitude adaptive module, Q0 is a constant of "3L / min". Q0 is used for feedforward flow setting (usually fixed at 3.0L / min) for comparison between the PID outer loop and the error e_L. In this application, Q′0 is used as the reference flow rate before altitude correction in both the LUT and altitude feedforward links.

[0076] This invention, through the aforementioned overall scheme, designs a shared portable oxygen generator and its overall control method capable of simultaneously supplying oxygen to two users at altitudes ranging from 0 to 5500 meters, featuring temperature / altitude adaptive control and BeiDou emergency rescue capabilities. Compared with existing technologies, this invention achieves the following significant advantages:

[0077] 1. Dual oxygen nozzle intelligent scheduling module: Enables "simultaneous, on-demand oxygen supply for two people".

[0078] Traditional portable oxygen concentrators only support single-channel output. If a second user is forcibly connected, the flow rate can drop by more than 30%, resulting in a maximum SpO2 difference of >8%. This invention, through a micro-differential pressure sensor, 100ms sliding filter, and three-threshold judgment, can identify the inhalation intensity of two users in real time and output pulsed / fixed-frequency oxygen flow with independent solenoid valves at duty cycles D1 and D2 respectively. Real-world testing with 30 subjects at altitudes of 0-4500m showed: flow rate error ≤1.8%, false trigger rate ≤0.3%, and switching delay ≤62ms, achieving "dual-user, non-interference" oxygen supply for the first time in a portable device.

[0079] Furthermore, this application directly uses the P(h) output from the altitude-oxygen partial pressure model as the "flow rate correction coefficient" in the dual-outlet oxygen nozzle scheduling, that is, Q′0 is used as the baseline flow rate before altitude correction in the LUT and altitude feedforward links, thus solving the misjudgment caused by single-nozzle flow rate attenuation at high altitudes. Experiments show that the single-nozzle false trigger rate has been reduced from 1.2% to 0.3%.

[0080] 2. Temperature-adaptive dual-loop control module: Overcoming the bottlenecks of "low-temperature degradation" and "overheating shutdown".

[0081] Existing products only offer single-temperature zone control, resulting in a 25% decrease in molecular sieve efficiency at night in high-altitude areas (-10℃), and the compressor is prone to overheating and automatic shutdown. This invention employs a molecular sieve constant-temperature circuit (PID + Smith predictive controller τ=8s, T=25s) + a compressor overheat protection circuit (anti-saturation PID), forming a dual closed loop: the molecular sieve temperature remains stable at 45±2℃, ensuring constant efficiency; the compressor temperature is below 85℃, preventing shutdown; and the operating range is extended by 28% in extreme environments.

[0082] 3. Altitude adaptive parameter adjustment module: 0-5500m "one-click adjustment"

[0083] Traditional equipment requires users to manually set the flow rate / concentration step by step, resulting in a missetting rate as high as 22%. This invention uses an exponential model P(h) = P0 × e (-h / 8431 The system uses a feedforward mechanism and superimposed gain-controlled dual-loop PID controller to correct the compressor speed and molecular sieve switching cycle in real time: flow error ≤ ±5%, concentration fluctuation ≤ ±1.5%, and no manual intervention is required throughout the process, achieving "altitude-adaptive adjustment-free" for the first time.

[0084] 4. BeiDou Emergency Rescue Module: "Second-Level Rescue" in Off-Network Scenarios

[0085] Traditional devices rely solely on 4G, losing connection upon signal loss. This invention integrates a BeiDou RDSS module: it sends an SOS message within 3 seconds using an 18-byte compressed short message; and it generates a rescue route using A* path planning (30m resolution DEM). Field tests in high-altitude areas showed a 96% SOS delivery success rate in weak network scenarios, reducing rescue arrival time by 40%.

[0086] 5. Cross-module collaborative optimization controller: achieving "cross-domain integration" and "full lifecycle adaptation"

[0087] Existing controllers operate independently, failing to balance multiple objectives such as oxygen demand, power consumption, safety, and positioning reliability. This invention constructs a joint cost function J_total, using MPC to solve for the 6-dimensional optimal vector u online every 100ms: directly mapping altitude-oxygen partial pressure P(h) to a flow rate correction coefficient, resolving misjudgments of single-point flow at high altitudes; the cloud updates the weight matrix w and constraint boundary B via OTA every 24h, achieving "full lifecycle adaptation." After 500 hours of continuous operation, overall power consumption decreased by 15%, and the failure rate was <0.1%.

[0088] In summary, this invention is the first to achieve comprehensive capabilities in a portable oxygen generator, including simultaneous on-demand oxygen supply for two people, adjustment-free operation from 0 to 5500m, all-dimensional self-adaptation, and emergency rescue via BeiDou navigation without network connectivity, which is significantly superior to existing technologies. Attached Figure Description

[0089] The present invention can be better understood by referring to the description given below in conjunction with the accompanying drawings, in which the same or similar reference numerals are used throughout the drawings to denote the same or similar parts. These drawings, together with the following detailed description, are incorporated in and form part of this specification, and are used to further illustrate preferred embodiments of the invention and explain the principles and advantages of the invention. In the drawings:

[0090] Figure 1 This is a schematic diagram of a high-altitude oxygen generation system based on multi-source fusion and synergistic optimization, according to an embodiment of the present invention.

[0091] Figure 2 This is a flowchart of a high-altitude oxygen production method based on multi-source fusion and synergistic optimization, according to an embodiment of the present invention. Detailed Implementation

[0092] Embodiments of the present invention will now be described with reference to the accompanying drawings. Elements and features described in one drawing or embodiment of the invention may be combined with elements and features shown in one or more other drawings or embodiments. It should be noted that, for clarity, representations and descriptions of components and processes unrelated to the present invention and known to those skilled in the art have been omitted from the drawings and description.

[0093] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0094] Example 1

[0095] This invention provides a high-altitude oxygen production system based on multi-source fusion and synergistic optimization. (See also...) Figure 1 It includes a dual oxygen nozzle intelligent scheduling module, a temperature adaptive dual-loop control module, an altitude adaptive parameter adjustment module, a Beidou emergency rescue module, and a cross-module collaborative optimization controller.

[0096] The dual oxygen nozzle intelligent scheduling module includes an oxygen storage tank, a filter, a pressure stabilizing chamber, and two oxygen nozzle channels. Each oxygen nozzle channel includes a breathing sensor, a solenoid valve, an air delivery tube, and an oxygen nozzle. The oxygen storage tank, filter, and pressure stabilizing chamber are connected in sequence and then divided into two solenoid valves that are connected to the two oxygen nozzle channels respectively. The breathing sensor is located above the pressure stabilizing chamber and on the outer wall of the air delivery tube, and is connected to the solenoid valve, air delivery tube, and oxygen nozzle in sequence.

[0097] The currently sensed respiratory negative pressure values f1 and f2 corresponding to the two oxygen outlets respectively. Each oxygen outlet has three grades of respiratory negative pressure thresholds fa, fb, and fc, and fa < fb < fc. Both of the two oxygen outlets have two oxygen supply modes m1 and m2. m1 is pulse oxygen supply, and m2 is fixed-frequency oxygen supply. The oxygen supply mode of each oxygen outlet can be set independently. The negative pressure generated during the current inhalation is collected by a negative pressure sensor and compared with the three negative pressure thresholds. To prevent the interference of air flow, time window filtering needs to be added.

[0098] This intelligent scheduling module for the two oxygen outlets is used to dynamically switch between pulse / fixed-frequency oxygen supply modes based on the negative pressure thresholds fa, fb, and fc and the preset sliding time window filtering; fa, fb, and fc respectively correspond to three statistical thresholds of the instantaneous negative pressure peaks generated by a person during the nasal inhalation stage and at the oxygen outlet interface of the oxygen generator.

[0099] fa (≈ -45 Pa): The minimum detectable negative pressure generated during shallow breathing / small tidal volume inhalation. Values below this are considered "no effective inhalation".

[0100] fb (≈ -90 Pa): The typical negative pressure peak during normal adult resting breathing. Reaching or exceeding this value is determined as "normal inhalation demand".

[0101] fc (≈ -135 Pa): The maximum negative pressure peak that appears during deep breathing / high-load exercise. Exceeding this value indicates that the user is in a "high oxygen demand" state.

[0102] fa, fb, and fc divide the continuous negative pressure signal into four physiological demand intervals, which are used to trigger different oxygen supply strategies (valve closing, fixed-frequency, low pulse, high pulse). In this embodiment, the negative pressure thresholds are fa = -45.0 ± 2.5 Pa, fb = -90.0 ± 3.0 Pa, and fc = -135.0 ± 3.5 Pa, which are measured by 30 subjects at an altitude of 0 - 4500 m under four breathing modes with a 95% confidence interval.

[0103] The temperature adaptive dual-loop control module includes a compressor, a molecular sieve tower, a fan, a drive board for driving the compressor and the fan, a molecular sieve constant temperature PID, a Smith predictor controller, and a compressor anti-overheating and anti-saturation PID controller. Pressure sensors are provided at the compressor exhaust port and the upper end of the molecular sieve, and temperature sensors for detecting the ambient temperature are provided at the air inlet of the machine shell, the compressor chamber, and the molecular sieve. The Smith predictor controller has a transfer function of G(s) = e (-8s) / (25s+1)The pure time delay is 8s, and the time constant is 25s. The compressor generates heat during operation; if it overheats, it will automatically shut down. The fan dissipates heat by accelerating the airflow within the oxygen concentrator. Therefore, based on the working principle and temperature control requirements of the oxygen concentrator, a dual-loop adaptive temperature control algorithm was designed. This algorithm dynamically coordinates the fan speed and compressor load to ensure a constant temperature for the molecular sieve (40-50℃) while preventing the compressor from overheating and shutting down.

[0104] The altitude adaptive parameter adjustment module includes an altitude sensor for measuring air pressure, a flow sensor for measuring airflow, a concentration sensor for measuring oxygen concentration, a compressor, and a molecular sieve switching valve. The flow sensor is located at the molecular sieve inlet, and the concentration sensor is located at the molecular sieve outlet. The compressor receives a speed correction command n. The molecular sieve switching valve can be a 24V solenoid valve assembly (3-position 4-way).

[0105] The altitude adaptive parameter adjustment module uses the exponential model P(h)=P0×e -h / 8431 The gain-controlled dual-closed-loop PID maintains a flow rate error of ≤5%.

[0106] The percentage of oxygen in the air varies with altitude. The higher the altitude, the more pronounced the symptoms of hypoxia and the greater the demand for oxygen. This altitude-adaptive parameter adjustment module automatically adjusts the working parameters of the oxygen generator according to the altitude changes, thereby ensuring that the flow rate and oxygen concentration remain stable.

[0107] The Beidou emergency rescue module is used to send location and physiological parameters in short message format and perform A* path planning when the plateau danger index (HRI) is ≥1.5.

[0108] The Beidou emergency rescue module uses meter-level Beidou positioning module and can also be equipped with 5G communication module and voice interaction module. The voice interaction module includes a microphone and speaker. The speaker can play a prompt tone when SOS is requested.

[0109] A cross-module collaborative optimization controller is used to execute:

[0110] a) Receive in real time the status and error quantities of the dual oxygen nozzle intelligent scheduling module, temperature adaptive dual-loop control module, altitude adaptive parameter adjustment module and Beidou emergency rescue module;

[0111] b) Based on the state and error quantities of the dual oxygen outlet intelligent scheduling module, temperature adaptive dual-loop control module, altitude adaptive parameter adjustment module and Beidou emergency rescue module, a joint cost function J_total is constructed with the user oxygen demand satisfaction, power consumption, compressor thermal safety margin and positioning-communication reliability as objectives.

[0112] c) Solve for the optimal control quantity of the joint cost function J_total and simultaneously send it to the dual oxygen outlet intelligent scheduling module, temperature adaptive dual-loop control module, altitude adaptive parameter adjustment module and Beidou emergency rescue module;

[0113] d) The weight matrix and constraint boundary of the joint cost function are periodically updated through the cloud parameter learning channel.

[0114] Specifically, the cross-module collaborative optimization controller includes a main control MCU, which can be implemented using STM32H743. The main control MCU also interacts with the cloud and updates weights and constraints every 24 hours via OTA.

[0115] In this embodiment, the real-time reception of the status and error quantities of the dual oxygen nozzle intelligent scheduling module, temperature adaptive dual-loop control module, altitude adaptive parameter adjustment module, and Beidou emergency rescue module in the aforementioned cross-module collaborative optimization controller specifically includes real-time reception of temperature error e, compressor overheat margin δ, flow error e_L, concentration error e_C, and positioning residual. ε and HRI value; where temperature error e is obtained by molecular sieve isothermal PID and Smith predictive controller; compressor overheat margin δ is obtained by compressor anti-overheating and anti-saturation PID controller; flow error e_L is obtained by altitude adaptive parameter adjustment module; concentration error e_C is obtained by altitude adaptive parameter adjustment module; positioning residual ε Data obtained from the BeiDou emergency rescue module; HRI value is the physiological risk index.

[0116] Specifically, the corresponding explanations for each symbol are as follows:

[0117]

[0118] The temperature error e is obtained by subtracting the target temperature of 45℃ from the real-time sampling of the molecular sieve temperature sensor t2.

[0119] The compressor overheat margin δ is calculated by subtracting the real-time temperature from the safety threshold of 85℃ after the compressor temperature sensor t1 samples the temperature.

[0120] The flow error e_L is obtained by subtracting the flow value measured in real time by the air flow sensor L from the set flow rate of 3L / min;

[0121] The concentration error e_C is obtained by subtracting 90% of the set concentration from the concentration value measured in real time by the oxygen concentration sensor n;

[0122] Positioning residual ε It is calculated from the difference between the position estimate output by the Beidou positioning module and the Kalman filter prediction value;

[0123] HRI is the high altitude risk index, HRI = 0.6 × [(100 - SpO2) / 10] + 0.4 × [(RR - 20) / 10]; SpO2 is blood oxygen saturation (%), and RR is respiratory rate, in breaths / min.

[0124] The constructed joint cost function J_total is expressed as follows:

[0125] The joint cost function J_total=w1×|e_L|+w2×|e_C|+w3×|δ|+w4×|HRI|+w5×P_in+w6×ε; where w1, w2, w3, w4, w5 and w6 are weight values, which are periodically updated by the cloud; P_in is the total electrical power that the oxygen concentrator draws from the power source in real time. P_in is calculated by sampling the bus voltage and the current of each actuator in real time: P_in=V_bus×(I_fan+I_comp+I_valve+I_comm) / 150W (per unit), where V_bus is the bus voltage of the whole machine (24V or 12V DC), I_fan is the real-time current of the fan motor, I_comp is the real-time current of the compressor motor, I_valve is the real-time current of the solenoid valve coil, and I_comm is the total current of the communication and control circuit.

[0126] The weights in the joint cost function are preset to: w1 = 0.35, w2 = 0.25, w3 = 0.15, w4 = 0.15, w5 = 0.05, and w6 = 0.05. The parameter values ​​in the joint cost function formula are as follows:

[0127]

[0128] The optimal control quantity for solving the joint cost function J_total is specifically obtained by using a rolling time-domain optimization algorithm to solve for the optimal control quantity u online; where u is the real-time vector control quantity, defined as [s, y, D1, D2, n, T_cycle]. T ; s is the fan speed, y is the compressor duty cycle, D1 is the solenoid valve duty cycle of one of the dual oxygen outlets, D2 is the solenoid valve duty cycle of the other oxygen outlet, n is the compressor base speed correction amount, and T_cycle is the molecular sieve switching cycle correction amount. The corresponding explanations for each symbol are as follows:

[0129]

[0130] Constraint boundaries refer to a set of real-time adjustable upper and lower limit intervals that the collaborative optimization controller must simultaneously satisfy when solving for the optimal control quantity u; the parameters of the constraint boundaries are shown in the table below:

[0131]

[0132] The constraint boundaries include the upper and lower limits of compressor safety T1, temperature safety T2, flow rate L, concentration C, fan speed s, and compressor duty cycle y. These constraint boundaries are periodically updated via OTA (Over-The-Air) updates in the cloud to ensure the system remains both safe and efficient throughout its entire lifecycle.

[0133] Example 2

[0134] See Figure 2 This invention provides a high-altitude oxygen production method based on multi-source fusion and synergistic optimization, comprising:

[0135] SO: The system performs a power-on self-test and loads the weight matrix w and constraint boundary B recently issued from the cloud.

[0136] S1: Execute S1a-S1d in parallel with a period of 100ms.

[0137] S1a: Intelligent scheduling sub-process for dual oxygen outlet nozzles, including:

[0138] a1) The micro differential pressure sensor (SDP800) samples the negative pressure f1 and f2 of the nozzle at 200Hz;

[0139] a2) Perform a 100ms sliding median filter to obtain f1 and f2;

[0140] a3) Compare with the preset thresholds fa, fb, and fc to determine the respiratory states S1 and S2;

[0141] In addition, to avoid misjudgment of single-nozzle flow rate caused by low pressure at high altitudes, this step also includes:

[0142] The real-time output P(h) of the altitude-oxygen partial pressure model (P(h) in the S1c altitude adaptive parameter adjustment subprocess) is mapped to the flow rate correction coefficient α(h): α(h)=P(h) / P0, P0=101.325kPa;

[0143] Correct the baseline flow rate Q0 in the LUT using α(h):

[0144] Q′0 = α(h)·Q0, keeping the thresholds fa, fb, and fc unchanged, but scaling the determined target flow rate level by Q′0 to avoid misjudgment of single-nozzle flow rate caused by low pressure at high altitudes. In the altitude adaptive module, Q0 is a constant of "3L / min". Q0 is used for feedforward flow setting (usually fixed at 3.0L / min) for comparison between the PID outer loop and the error e_L. In this application, Q′0 is used as the reference flow rate before altitude correction in both the LUT and altitude feedforward links.

[0145] a4) Based on the duty cycles D1 and D2 of the output solenoid valve of the LUT, the pulse / fixed frequency switching is realized;

[0146] a5) If an abnormal negative pressure is detected within 30 seconds, a module-level alarm will be triggered and a fault code will be uploaded;

[0147] S1b: Temperature adaptive dual-loop control sub-process, including:

[0148] b1) Collect the molecular sieve temperature t2 and the compressor temperature t1;

[0149] b2) Calculation error e = t2 - 45℃, margin δ = 85℃ - t1;

[0150] b3) A Smith predictive controller (τ = 8s, T = 25s) and an anti-saturation PID controller are used to calculate the fan speed s and the compressor duty cycle y;

[0151] b4) If δ < 5℃, immediately force y = 30% and fan speed to full speed until δ ≥ 10℃;

[0152] S1c: Altitude adaptive parameter adjustment subprocess, including:

[0153] c1) Read the altitude and barometric pressure sensor h, flow sensor L, and oxygen concentration sensor n;

[0154] c2) Using the exponential model P(h)=P0×e (-h / 8431) Calculate the feedforward reference speed n0;

[0155] c3) Calculate the flow error e_L = 3L / min - L_sensor, and the concentration error e_C = 90% - C_sensor;

[0156] c4) Gain-controlled PID outputs n and T_cycle ensure that flow error ≤ 5% and concentration fluctuation ≤ 1.5%;

[0157] c5) If the Lyapunov function fails to determine stability (V>-0.05||x||), then... 2 (Continues for 3 seconds) Enters debit mode;

[0158] S1d: Beidou emergency rescue sub-process, including:

[0159] d1) Collect blood oxygen saturation SpO2 and respiratory rate RR, and calculate HRI: HRI=0.6×(100-SpO2) / 10+0.4×(RR-20) / 10;

[0160] d2) If HRI ≥ 1.5 lasts for 5 seconds, trigger SOS;

[0161] d3) The BeiDou positioning module outputs a position estimate, which is then used to calculate the positioning residual after Kalman filtering. ε ;

[0162] d4) Assemble an 18-byte short message and send it via BeiDou RDSS;

[0163] d5) After receiving the short message, the cloud executes A* path planning and returns the rescue path.

[0164] S2: The cross-module collaborative optimization controller collects the state variables and error variables of S1a-S1d, constructs the joint cost function J_total, and solves the optimal control variable u online;

[0165] S3: Broadcast u to each actuator at once via CAN-FD bus to achieve synchronized operation of the four modules;

[0166] S4: Update w and B via cloud OTA every 24 hours and return to S0.

[0167] When in use, the user first powers on the device, and the system loads the latest w and b from the cloud. Single or dual-user connection: Insert half of the oxygen outlet into the left and right oxygen outlets respectively. Scan the code / affix an NFC card to complete identity binding (optional, for shared billing). The system adaptively supplies oxygen: ① Normal user inhalation → The device identifies the breathing intensity within 100ms and automatically switches between pulse / fixed frequency. ② Real-time screen display: Spo2, RR, remaining battery power, and altitude indicator. ③ If the temperature light flashes orange, it indicates that the molecular sieve is heating; if it flashes blue, it indicates that the compressor is reducing load. One-button SOS (only for emergency use): ① Press the SOS button twice consecutively → 3 beeps + red screen flash. ② Within 3 seconds, a Beidou short message is sent, and the cloud pushes the rescue route to the driver / family member's mobile phone. ③ Before the rescue arrives, the device automatically enters "emergency high flow" mode.

[0168] Furthermore, unlike existing technologies, this system can also proactively issue alarms based on detected anomalies:

[0169] The user's location coordinates are obtained in real time through the BeiDou positioning module, and the raw observation values ​​are subjected to sliding window weighted filtering and residual analysis. ε To determine the reliability of the positioning, if ε > ε0 Then switch to BeiDou Doppler assisted positioning mode;

[0170] The altitude sickness risk index (HRI) is generated based on a physiological parameter abnormality detection model. When the HRI ≥ θ1 and continues for a preset time, an SOS signal is triggered.

[0171] The user ID, longitude, latitude, altitude, HRI, and timestamp are encapsulated into an 18-byte BeiDou short message and sent via the RDSS link. The 18-byte short message format is: Byte0: 0xAA start; Bytes 1-2: User ID; Bytes 3-4: Longitude × 10 4 Byte5-6: Latitude × 104 ; Byte7: SpO2; Byte8: RR; Byte9: HRI×100; Byte10-11: Altitude; Byte12-13: CRC16; Byte14-17: UnixTime lower 32 bits.

[0172] After receiving the short message, the cloud-based rescue platform uses the A* algorithm combined with an elevation map to plan a rescue route and then sends the route to the rescue terminal.

[0173] Application Scenario 1:

[0174] Process 1: Driver A connects to oxygen outlet nozzle 1, and passenger B connects to oxygen outlet nozzle 2.

[0175] Process 1: Power-on self-test of the device → Load cloud w = [0.35, 0.25, 0.15, 0.15, 0.05, 0.05] and constraint boundary B (T2∈[40, 50]℃, etc.).

[0176] Process 2, Dual Oxygen Outlet Module:

[0177] A was detected to be in a "deep breathing" state (f1≈-140Pa), D1 duty cycle 70%, peak flow rate 4.2L / min;

[0178] B was detected as being in a "normal breathing" state (f2≈-90Pa), D2 duty cycle was 45%, and peak flow rate was 2.6L / min;

[0179] Flow error <1.8%, no cross-flow.

[0180] Process 3, Temperature Loop:

[0181] With an outside temperature of -8℃, t2 stabilizes at 45.3℃; t1 peaks at 82℃ → no forced load reduction is required.

[0182] Process 4, Altitude Loop:

[0183] At 4500m, α(h) = 0.58, Q′0 = 1.74 L / min; e_L and e_C are both < ±2%.

[0184] Process 5: During use, driver A experienced altitude sickness: SpO2 dropped to 84%, RR 28 times / min → HRI = 1.6.

[0185] SOS will be triggered after 5 seconds;

[0186] The Beidou emergency rescue module sends an 18-byte message within 3 seconds: {ID=A, Lon=94.3°, Lat=30.1°, Alt=4505m, HRI=160, SpO2=84, RR=28}.

[0187] Step 6: The cloud receives the message, and A* path planning generates a rescue route (2.1km in a straight line, estimated arrival time 12 minutes).

[0188] The cloud sends out the "Emergency High Flow Mode" command: u update: D1=90%, D2=70%, fan at full speed, compressor at 95%.

[0189] Driver A's SpO2 recovered to 91% within 22 minutes, HRI dropped to 0.8, and SOS was automatically deactivated.

[0190] The above process enables uninterrupted oxygen supply for two people for 60 minutes; the equipment's peak power consumption is 135W, with 42% battery remaining; the rescue vehicle arrives in 11 minutes, achieving "zero-wait" rescue.

[0191] It should be emphasized that the term "including / comprises" as used herein refers to the presence of a feature, element, step, or component, but does not exclude the presence or addition of one or more other features, elements, steps, or components.

[0192] Furthermore, the method of the present invention is not limited to being executed in the chronological order described in the specification, but may also be executed in other chronological orders, in parallel, or independently. Therefore, the execution order of the method described in this specification does not constitute a limitation on the technical scope of the present invention.

[0193] Although the invention has been disclosed above through the description of specific embodiments, it should be understood that all the embodiments and examples described above are exemplary and not restrictive. Those skilled in the art can design various modifications, improvements, or equivalents to the invention within the spirit and scope of the appended claims. These modifications, improvements, or equivalents should also be considered to be included within the protection scope of the invention.

Claims

1. A high-altitude oxygen production system based on multi-source fusion and synergistic optimization, characterized in that, include: The dual oxygen nozzle intelligent scheduling module is used to dynamically switch between pulse / fixed frequency oxygen supply modes based on negative pressure thresholds fa, fb, fc and preset sliding time window filtering. fa, fb, and fc correspond to three statistical thresholds for the instantaneous negative pressure peak generated during nasal inhalation and at the oxygen outlet of the oxygen concentrator, respectively. Temperature adaptive dual-loop control module, including molecular sieve constant temperature PID, Smith predictive controller and compressor anti-overheating and anti-saturation PID controller; The altitude adaptive parameter adjustment module uses the exponential model P(h)=P0×e -h / 8431 The gain-controlled dual-loop PID maintains a flow error of ≤5%, P0=101.325kPa; The Beidou emergency rescue module is used to send location and physiological parameters in short message format and perform A* path planning when the high altitude danger index (HRI) is ≥1.

5. A cross-module collaborative optimization controller is used to execute: a) Receive in real time the status and error quantities of the dual oxygen nozzle intelligent scheduling module, temperature adaptive dual-loop control module, altitude adaptive parameter adjustment module and Beidou emergency rescue module; b) Based on the state and error quantities of the dual oxygen outlet intelligent scheduling module, temperature adaptive dual-loop control module, altitude adaptive parameter adjustment module and Beidou emergency rescue module, a joint cost function J_total is constructed with the user oxygen demand satisfaction, power consumption, compressor thermal safety margin and positioning-communication reliability as objectives. c) Solve for the optimal control quantity of the joint cost function J_total and simultaneously send it to the dual oxygen outlet intelligent scheduling module, temperature adaptive dual-loop control module, altitude adaptive parameter adjustment module and Beidou emergency rescue module; d) Periodically update the weight matrix and constraint boundary of the joint cost function through the cloud parameter learning channel; The constructed joint cost function J_total is expressed as follows: The joint cost function J_total = w1 × |e_L| + w2 × |e_C| + w3 × |δ| + w4 × |HRI| + w5 × P_in + w6 × ε; where w1, w2, w3, w4, w5, and w6 are weight values, which are periodically updated by the cloud; P_in is the total electrical power that the oxygen concentrator draws from the power source in real time; The optimal control quantity for solving the joint cost function J_total is specifically obtained by using a rolling time-domain optimization algorithm to solve for the optimal control quantity u online. Here, u is the real-time vector control quantity, defined as [s,y,D1,D2,n,T_cycle]ᵀ; s is the fan speed, y is the compressor duty cycle, D1 is the solenoid valve duty cycle of one of the dual oxygen outlets, D2 is the solenoid valve duty cycle of the other oxygen outlet, n is the compressor reference speed correction, and T_cycle is the molecular sieve switching cycle correction.

2. The high-altitude oxygen production system based on multi-source fusion and synergistic optimization according to claim 1, characterized in that, In the cross-module collaborative optimization controller, the real-time reception of state and error quantities from the dual oxygen nozzle intelligent scheduling module, temperature adaptive dual-loop control module, altitude adaptive parameter adjustment module, and Beidou emergency rescue module specifically includes real-time reception of temperature error e, compressor overheat margin δ, flow error e_L, concentration error e_C, positioning residual ε, and HRI value. Specifically, temperature error e is obtained from the molecular sieve constant temperature PID and Smith prediction controller; compressor overheat margin δ is obtained from the compressor anti-overheating and anti-saturation PID controller; flow error e_L is obtained from the altitude adaptive parameter adjustment module; concentration error e_C is obtained from the altitude adaptive parameter adjustment module; positioning residual ε is obtained from the Beidou emergency rescue module; and HRI value is the physiological risk index.

3. The high-altitude oxygen production system based on multi-source fusion and synergistic optimization according to claim 1, characterized in that, The constraint boundary refers to a set of real-time adjustable upper and lower limit intervals that the collaborative optimization controller must simultaneously satisfy when solving for the optimal control quantity u. The constraint boundary is periodically updated by OTA in the cloud to ensure that the system is both safe and efficient throughout its entire life cycle. The constraint boundary includes the upper and lower limits of compressor safety T1, temperature safety T2, flow rate L, concentration C, fan speed s, and compressor duty cycle y.

4. The high-altitude oxygen production system based on multi-source fusion and synergistic optimization according to claim 2, characterized in that, The temperature error e is obtained by subtracting the target 45°C from the real-time sampling of the molecular sieve temperature sensor t2. The compressor overheat margin δ is calculated by subtracting the real-time temperature from the safety threshold of 85℃ after the compressor temperature sensor t1 samples the temperature. The flow error e_L is obtained by subtracting the flow value measured in real time by the flow sensor from the set flow rate of 3L / min; The concentration error e_C is obtained by subtracting 90% of the set concentration from the concentration value measured in real time by the oxygen concentration sensor; The positioning residual ε is calculated from the difference between the position estimate output by the BeiDou positioning module and the Kalman filter prediction value; HRI is the high altitude risk index, HRI=0.6×[(100−SpO2) / 10]+0.4×[(RR−20) / 10]; SpO2 is blood oxygen saturation, and RR is respiratory rate, in breaths / min.

5. The high-altitude oxygen production system based on multi-source fusion and synergistic optimization according to claim 1, characterized in that, In the joint cost function, the weights are preset to: w1=0.35, w2=0.25, w3=0.15, w4=0.15, w5=0.05, w6=0.

05.

6. A high-altitude oxygen production method based on multi-source fusion and synergistic optimization, characterized in that, The high-altitude oxygen generation system described in any one of claims 1-5 comprises: S0: System power-on self-test, loading the weight matrix w and constraint boundary B recently issued from the cloud; S1: Execute S1a–S1d in parallel with a period of 100ms. S1a: Intelligent scheduling sub-process for dual oxygen outlet nozzles; S1b: Temperature adaptive dual-loop control sub-process; S1c: Altitude adaptive parameter adjustment subprocess; S1d: Beidou emergency rescue sub-process; S2: The cross-module collaborative optimization controller collects the state variables and error variables of S1a–S1d, constructs the joint cost function J_total, and solves the optimal control variable u online; S3: Broadcast the optimal control u to each actuator to achieve synchronized action of the four modules; S4: Update w and B via cloud OTA every 24 hours and return to S0; The S1a dual oxygen nozzle intelligent scheduling sub-process specifically includes: a1) Sample the negative pressure f1 and f2 of the nozzle at 200Hz; a2) Perform a 100ms sliding median filter to obtain f1 and f2; a3) Compare with the preset thresholds fa, fb, and fc to determine the respiratory states S1 and S2; a4) Based on the duty cycles D1 and D2 of the output solenoid valve of the LUT, the pulse / fixed frequency switching is realized; a5) If an abnormal negative pressure is detected within 30 seconds, a module-level alarm will be triggered and a fault code will be uploaded; The S1b temperature adaptive dual-loop control sub-process specifically includes: b1) Collect the molecular sieve temperature t2 and the compressor temperature t1; b2) Calculate the error e = t2 − 45℃ and the margin δ = 85℃ − t1; b3) The fan speed s and compressor duty cycle y are calculated using a Smith predictive controller and an anti-saturation PID controller; b4) If δ < 5℃, immediately force y = 30% and fan speed to full until δ ≥ 10℃; The S1c altitude adaptive parameter adjustment subprocess specifically includes: c1) Read the altitude and barometric pressure sensor, flow sensor, and oxygen concentration sensor; c2) Using the exponential model P(h) = P0 × e (−h / 8431) Calculate the feedforward reference speed n0, P0 = 101.325 kPa; c3) Calculate the flow error e_L = 3L / min − L_sensor, and the concentration error e_C = 90% − C_sensor; c4) Gain-controlled PID outputs n and T_cycle ensure flow error ≤ 5% and concentration fluctuation ≤ 1.5%; c5) If the Lyapunov function fails to determine stability, it enters the derating mode; The S1d Beidou emergency rescue sub-process includes: d1) Collect SpO2 and respiratory rate RR, and calculate HRI = 0.6 × (100 - SpO2) / 10 + 0.4 × (RR - 20) / 10; d2) If HRI ≥ 1.5 lasts for 5 seconds, trigger SOS; d3) The Beidou positioning module outputs the position estimate, and the positioning residual ε is calculated after Kalman filtering; d4) Assemble an 18-byte short message and send it via BeiDou RDSS; d5) After receiving the short message, the cloud executes A* path planning and returns the rescue path.

Images