CN-121348719-B - Plateau oxygen generation system and method based on multi-source fusion and collaborative optimization

CN121348719BCN 121348719 BCN121348719 BCN 121348719BCN-121348719-B

Abstract

The invention discloses a plateau oxygen generation system and a method based on multi-source fusion and collaborative optimization, wherein the system comprises a double oxygen outlet nozzle intelligent scheduling module, a temperature self-adaptive double-loop control module, an altitude self-adaptive parameter adjusting module, a Beidou emergency rescue module and a cross-module collaborative optimization controller, wherein the cross-module collaborative optimization controller is used for receiving state quantity and error quantity of each module in real time so as to construct a joint cost function; and solving the optimal control quantity of the joint cost function and synchronously issuing the optimal control quantity to each module to perform corresponding control. The invention realizes the comprehensive capability of two persons to simultaneously supply oxygen according to the needs, 0-5500m adjustment-free, full-dimension self-adaption and net-free Beidou emergency rescue in the portable oxygenerator for the first time, and has good practicability.

Inventors

WANG MI
SUN DEQUAN
YANG XIAOXI
YANG SHENGYUE
WANG DONGXU
TENG KAI
GUAN RUYI
QIU LIN

Assignees

西藏北斗森荣科技(集团)股份有限公司

Dates

Publication Date: 20260508
Application Date: 20250911

Claims (6)

1. A plateau oxygen generation system based on multisource fusion and collaborative optimization, comprising: The intelligent scheduling module of the double oxygen outlet nozzles is used for dynamically switching pulse/fixed-frequency oxygen supply modes based on negative pressure thresholds fa, fb and fc and preset sliding time window filtering, wherein fa, fb and fc respectively correspond to three statistical thresholds of instantaneous negative pressure peaks generated by a person at the oxygen outlet nozzle interface of the oxygen generator in the nasal inhalation stage; The temperature self-adaptive double-loop control module comprises a molecular sieve constant-temperature PID, a Smith pre-estimated controller and a compressor overheat-proof and saturation-proof PID controller; The elevation self-adaptive parameter adjusting module maintains the flow error to be less than or equal to 5% through an exponential model P (h) =P 0 ×e -h/8431 and a gain scheduling double-closed-loop PID, and P 0 =101.325 kPa; The Beidou emergency rescue module is used for sending the position and the physiological parameters in a short message format and executing a path planning when the altitude risk index HRI is more than or equal to 1.5; a cross-module co-optimization controller for performing: a) The method comprises the steps of receiving state quantity and error quantity of the double-oxygen-outlet-nozzle intelligent scheduling module, the temperature self-adaptive double-loop control module, the altitude self-adaptive parameter adjusting module and the Beidou emergency rescue module in real time; b) Based on the state quantity and the error quantity of the double-oxygen-outlet-nozzle intelligent scheduling module, the temperature self-adaptive double-loop control module, the altitude self-adaptive parameter adjusting module and the Beidou emergency rescue module, constructing a joint cost function J_total aiming at the satisfaction degree of oxygen demand of a user, power consumption, thermal safety margin of a compressor and positioning-communication reliability; c) Solving the optimal control quantity of the joint cost function J_total, and synchronously issuing the optimal control quantity to the double-oxygen-outlet-nozzle intelligent scheduling module, the temperature self-adaptive double-loop control module, the altitude self-adaptive parameter adjusting module and the Beidou emergency rescue module; d) Periodically updating the weight matrix and the constraint boundary of the joint cost function through a 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×|delta|+w4×|HRI|+w5×P_in+w6×ε, wherein w1, w2, w3, w4, w5 and w6 are weight values and are periodically updated by the cloud, and P_in is the total electric power drawn by the oxygenerator from the power supply in real time; The method comprises the steps of solving an optimal control quantity of a joint cost function J_total, and specifically adopting a rolling time domain optimization algorithm to solve the optimal control quantity u on line, wherein u is a real-time vector control quantity, u is defined as [ s, y, D 1 ,D 2 , n, T_cycle ] T, s is a fan rotating speed, y is a compressor duty ratio, D 1 is a solenoid valve duty ratio of one oxygen outlet nozzle in the double oxygen outlets, D 2 is a solenoid valve duty ratio of the other oxygen outlet nozzle, n is a compressor reference rotating speed correction quantity, and T_cycle is a molecular sieve switching period correction quantity.
2. The plateau oxygen generation system based on multi-source fusion and collaborative optimization according to claim 1, wherein the state quantity and error quantity of the dual oxygen outlet nozzle intelligent scheduling module, the temperature self-adaptive dual-loop control module, the altitude self-adaptive parameter adjusting module and the Beidou emergency rescue module are received in real time in the cross-module collaborative optimization controller, specifically, a temperature error e, a compressor overheat margin delta, a flow error e_L, a concentration error e_C, a positioning residual epsilon and an HRI value are received in real time, wherein the temperature error e is obtained by a molecular sieve constant temperature PID and Smith pre-estimated controller, the compressor overheat margin delta is obtained by a compressor overheat-resistant and saturation-resistant PID controller, the flow error e_L is obtained by an altitude self-adaptive parameter adjusting module, the concentration error e_C is obtained by an altitude self-adaptive parameter adjusting module, the positioning residual epsilon is obtained by a Beidou emergency rescue module, and the HRI value is a physiological risk index.
3. The plateau oxygen generation system based on multi-source fusion and collaborative optimization according to claim 1, wherein the constraint boundary refers to a group of real-time adjustable upper and lower limit intervals which must be met by a collaborative optimization controller when solving an optimal control quantity u, the constraint boundary is updated by cloud-end periodic OTA to ensure that the system is safe and efficient in a full life cycle, and the constraint boundary comprises an upper limit and a lower limit of a compressor safety T1, an upper limit and a lower limit of a temperature safety T2, an upper limit and a lower limit of a flow L, an upper limit and a lower limit of a concentration C, an upper limit and a lower limit of a fan rotating speed s and an upper limit and a lower limit of a compressor duty ratio y.
4. The plateau oxygen generating system based on multi-source fusion and collaborative optimization according to claim 2, wherein the temperature error e is obtained by subtracting a target 45 ℃ from a molecular sieve temperature sensor t2 sampled in real time; the overheat margin delta of the compressor is calculated by subtracting the real-time temperature from the safety threshold value of 85 ℃ after sampling by the temperature sensor t1 of the compressor; The flow error e_L is obtained by subtracting a flow value measured by a flow sensor in real time from a set flow 3L/min; The concentration error e_C is obtained by subtracting a concentration value measured by an oxygen concentration sensor in real time from a set concentration of 90%; the positioning residual epsilon is obtained by calculating the difference between the position estimation output by the Beidou positioning module and the Kalman filtering predicted value; The HRI value is the altitude risk index, hri=0.6× [ (100-SpO 2 )/10]+0.4×[(RR−20)/10];SpO 2 ) is the blood oxygen saturation, RR is the respiratory rate in units of times/min.
5. The system of claim 1, wherein the weight preset value in the joint cost function is w1=0.35, w2=0.25, w3=0.15, w4=0.15, w5=0.05, w6=0.05.
6. A method for preparing oxygen on a plateau based on multi-source fusion and collaborative optimization, which is characterized by being applied to the plateau oxygen preparing system as claimed in any one of claims 1 to 5, comprising: S0, system power-on self-checking, and loading a weight matrix w and a constraint boundary B which are recently issued by a cloud; s1, executing S1a-S1d in parallel with 100ms as a period: s1a, intelligent scheduling sub-process of double oxygen nozzles; s1b, temperature self-adaptive double-loop control sub-flow; s1c, altitude self-adaptive parameter adjustment sub-process; S1d, beidou emergency rescue sub-process; s2, collecting state quantity and error quantity of the S1a-S1d by a cross-module collaborative optimization controller, constructing a joint cost function J_total, and solving the optimal control quantity u on line; s3, broadcasting the optimal control u to each executing mechanism to realize synchronous actions of four modules; s4, updating w and B through cloud OTA every 24 hours, and returning to S0; The S1a intelligent scheduling sub-process of the double oxygen nozzles specifically comprises the following steps: a1 Sampling the negative pressure f1 and f2 of the raising mouth at 200 Hz; a2 Performing 100ms sliding median filtering to obtain f1 and f2; a3 Comparing the breathing state with preset threshold values fa, fb, fc, and judging the breathing state S1, S2; a4 According to the duty ratio D1 and D2 of the LUT output electromagnetic valve, realizing pulse/fixed frequency switching; a5 If abnormal negative pressure is detected within 30s, triggering module level alarm and uploading fault codes; The S1b temperature self-adaptive double-loop control sub-flow specifically comprises: b1 Collecting molecular sieve temperature t2 and compressor temperature t1; b2 Calculating an error e=t2-45 ℃, a margin δ=85 ℃ -t1; b3 Calculating the fan rotation speed s and the compressor duty ratio y by adopting a Smith pre-estimated controller and an anti-saturation PID; b4 If delta <5 ℃, immediately forcing y=30% and the fan to be at full speed until delta is more than or equal to 10 ℃; The S1c elevation self-adaptive parameter adjustment sub-process specifically comprises the following steps: c1 Reading an altitude barometric sensor, a flow sensor and an oxygen concentration sensor; c2 Calculating a feedforward reference rotation speed n0 with an exponential model P (h) =p 0 ×e (−h/8431) , P 0 =101.325 kPa; c3 Calculating a flow error e_l=3l/min-l_sensor, and a concentration error e_c=90% -c_sensor; c4 Gain scheduling PID output n, T_cycle, so that the flow error is less than or equal to 5%, and the concentration fluctuation is less than or equal to 1.5%; c5 If the Lyapunov function fails to judge stability, entering a derating mode; S1d Beidou emergency rescue sub-process comprises the following steps: d1 SpO 2 , respiratory rate RR, hri=0.6× (100-SpO 2 )/10+0.4× (RR-20)/10; d2 If HRI is more than or equal to 1.5 for 5 seconds, triggering SOS; d3 Outputting position estimation by the Beidou positioning module, and calculating positioning residual epsilon after Kalman filtering; d4 Assembling 18byte short messages and sending the 18byte short messages through Beidou RDSS; d5 And (3) the cloud end executes the A-path planning after receiving the short message and returns to the rescue path.

Description

Plateau oxygen generation system and method based on multi-source fusion and collaborative optimization Technical Field The invention relates to the technical field of plateau medical equipment, in particular to a shared portable oxygen generator which can supply oxygen to two users at the altitude of 0-5500m simultaneously and has the temperature/altitude self-adaptive control and Beidou emergency rescue capability, and a plateau oxygen generation method based on multi-source fusion and collaborative optimization. Background The existing portable oxygenerator exposes the following systematic defects under a plateau sharing scene: a) Continuous voyage bottleneck caused by integration of battery and host The traditional product adopts an integrated battery pack (18650 or polymer lithium battery), has fixed capacity and can not be hot plugged. The internal resistance of the battery is increased by 30-50% at a high altitude and at a low temperature (-10 ℃ below), the effective capacity is reduced by 25-40%, meanwhile, the peak current can reach 8-12A in a pulse oxygen supply mode, the duration is further shortened to be within 40 minutes, and the long-term requirements of high-altitude hiking and motorcade crossing are difficult to meet. B) Single oxygen supply mode and no double differential oxygen supply The mainstream model on the market only supports two fixed curves, namely 'continuous flow' or 'fixed pulse'. When two users (adults, vs children) inhale simultaneously, independent flow and pulse frequency cannot be adjusted according to the difference of the respective tidal volume and respiratory frequency, so that the phenomena of over-inhaling by one person and under-inhaling by one person are caused, and the maximum difference of the measured SpO 2 can reach 8%. C) The plateau environment needs to manually adjust parameters, has complex operation and is easy to be set by mistake Every 1000m altitude rise, the atmospheric pressure drops by about 12%, and the oxygen partial pressure drops simultaneously. The prior equipment needs a user to set flow, concentration and pulse frequency step by step through a knob or a key, not only depends on experience, but also is easy to produce misoperation in an anoxic environment. Experiments show that the manually set error rate can reach 22% at 4200m altitude, and 15% of users forget to correct again within 30 min. D) Lack of positioning alarm function required for shared scenarios Positioning deficiency, namely the existing oxygenerator has no alarm positioning function; communication loss, namely no cellular/satellite communication link exists, and SOS cannot be reported; The defects cause serious defects of reliability, usability and commercial sustainability of the existing products in the scenes of altitude sharing, emergency rescue, long-term operation and the like, and systematic innovation is needed urgently Disclosure of Invention The following presents a simplified summary of embodiments of the invention in order to provide a basic understanding of some aspects of the invention. It should be understood that the following summary is not an exhaustive overview of the invention. It is not intended to identify key or critical elements of the invention or to delineate the scope of the invention. Its purpose is to present some concepts in a simplified form as a prelude to the more detailed description that is discussed later. The invention aims to overcome the defects and provide a plateau oxygen production system and a plateau oxygen production method based on multi-source fusion and collaborative optimization, wherein the implementation scheme comprises four core modules: the intelligent dispatching module of the double oxygen nozzles, the Beidou emergency rescue module, the temperature self-adaptive double-loop control module, the altitude self-adaptive parameter adjusting module and the cross-module collaborative optimization controller integrate all the modules which are originally independently operated into a whole through cross-module joint state estimation, multi-target collaborative optimization and cloud closed loop calibration, so that a complete technical scheme of perception-prediction-decision-execution-learning is formed, and reliability, usability and commercial sustainability of the oxygenerator under the scenes such as plateau sharing, emergency rescue, long-term operation and the like are improved. According to a first aspect of the present application, there is provided a plateau oxygen generating system based on multi-source fusion and collaborative optimization, comprising: The intelligent scheduling module of the double oxygen outlet nozzles is used for dynamically switching pulse/fixed-frequency oxygen supply modes based on negative pressure thresholds fa, fb and fc and preset sliding time window filtering, wherein fa, fb and fc respectively correspond to three statistical thresholds of instantaneous negative pressure peaks generated by