CN110529419A - The pressure output control method of noninvasive ventilator blower - Google Patents

The pressure output control method of noninvasive ventilator blower Download PDF

Info

Publication number
CN110529419A
CN110529419A CN201910823394.3A CN201910823394A CN110529419A CN 110529419 A CN110529419 A CN 110529419A CN 201910823394 A CN201910823394 A CN 201910823394A CN 110529419 A CN110529419 A CN 110529419A
Authority
CN
China
Prior art keywords
pressure
neural network
pressure output
blower
network algorithm
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201910823394.3A
Other languages
Chinese (zh)
Inventor
江松林
戴泽远
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Suzhou Beth Medical Technology Co Ltd
Original Assignee
Suzhou Beth Medical Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Suzhou Beth Medical Technology Co Ltd filed Critical Suzhou Beth Medical Technology Co Ltd
Priority to CN201910823394.3A priority Critical patent/CN110529419A/en
Publication of CN110529419A publication Critical patent/CN110529419A/en
Pending legal-status Critical Current

Links

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
    • A61M16/00—Devices for influencing the respiratory system of patients by gas treatment, e.g. mouth-to-mouth respiration; Tracheal tubes
    • A61M16/0003—Accessories therefor, e.g. sensors, vibrators, negative pressure
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
    • F04D—NON-POSITIVE-DISPLACEMENT PUMPS
    • F04D27/00—Control, e.g. regulation, of pumps, pumping installations or pumping systems specially adapted for elastic fluids
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61M—DEVICES FOR INTRODUCING MEDIA INTO, OR ONTO, THE BODY; DEVICES FOR TRANSDUCING BODY MEDIA OR FOR TAKING MEDIA FROM THE BODY; DEVICES FOR PRODUCING OR ENDING SLEEP OR STUPOR
    • A61M2205/00—General characteristics of the apparatus
    • A61M2205/33—Controlling, regulating or measuring
    • A61M2205/3331—Pressure; Flow
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F05—INDEXING SCHEMES RELATING TO ENGINES OR PUMPS IN VARIOUS SUBCLASSES OF CLASSES F01-F04
    • F05D—INDEXING SCHEME FOR ASPECTS RELATING TO NON-POSITIVE-DISPLACEMENT MACHINES OR ENGINES, GAS-TURBINES OR JET-PROPULSION PLANTS
    • F05D2270/00—Control
    • F05D2270/70—Type of control algorithm
    • F05D2270/706—Type of control algorithm proportional-integral-differential
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F05—INDEXING SCHEMES RELATING TO ENGINES OR PUMPS IN VARIOUS SUBCLASSES OF CLASSES F01-F04
    • F05D—INDEXING SCHEME FOR ASPECTS RELATING TO NON-POSITIVE-DISPLACEMENT MACHINES OR ENGINES, GAS-TURBINES OR JET-PROPULSION PLANTS
    • F05D2270/00—Control
    • F05D2270/70—Type of control algorithm
    • F05D2270/709—Type of control algorithm with neural networks

Abstract

The invention discloses the pressure output control methods of noninvasive ventilator blower, it is related to ventilator air-blower control technical field, the pressure output control method of the noninvasive ventilator blower, by neural network algorithm to carrying out data processing between deviation and control amount, the pressure control data precision obtained is higher, the practical airway pressure force value kurtosis of user is small simultaneously, extremum is small, it fluctuates small, airway pressure waveform is being exhaled, steady state is mutually presented in suction, it is obviously improved the accuracy of ventilator blower pressure output, the wind pressure that ventilator blower is exported is quick, accurately switch between breath pressure (EPAP) and pressure of inspiration(Pi) (IPAP), auto-control is made according to patient airway flow and pressure amplitude, the stress level for enabling ventilator blower to export suits the needs of the state of an illness and the requirement for the treatment of in multiple angles.

Description

The pressure output control method of noninvasive ventilator blower
Technical field
The present invention relates to ventilator air-blower control technical fields, specially the pressure output controlling party of noninvasive ventilator blower Method.
Background technique
Noninvasive ventilator during the work time, passes through differential pressure pickup and the value of feedback judgement breathing phase of pressure sensor The control of switching and blower output pressure provides the positive pressure of different level in patient breaths and exhalation process to carry out Respiratory Therapy It treats.
The blower that noninvasive ventilator drives frequently with brshless DC motor is as gas generating unit, the method for controlling motor It main start-up and shut-down control including real-time detection rotor-position and is controlled, the precision of control and is resisted dry by the revolving speed of PWM pressure regulation The performance disturbed is poor, and this patent proposes PID blower pressure output control method neural network based that can be adaptive.
Summary of the invention
(1) the technical issues of solving
In view of the deficiencies of the prior art, the present invention provides the pressure output control methods of noninvasive ventilator blower, solve It the start-up and shut-down control of real-time detection rotor-position and is controlled by the revolving speed of PWM pressure regulation, the precision of control and the property of resisting interference The poor problem of energy.
(2) technical solution
To achieve the above objectives, the technical solution adopted by the present invention is that: the pressure output controlling party of noninvasive ventilator blower Method includes the following steps；
S1, processing is acquired to the data of breath pressure in actual use and pressure of inspiration(Pi)；
S2, the neural network algorithm model for establishing breathing blower；
The application of S3, neural network algorithm model.
Preferably, in the S1 step, PID controller is directly initiated, obtains the relational expression between control amount and deviation are as follows:
Preferably, wherein E (t) is deviation, and U (t) is control amount, KP, KI, KDRespectively ratio, integral and differential coefficient.
Preferably, it when sampling is packed up short enough, needs by PID control process discretization, using incremental digital PID: Δ U (n)=KpΔE(n)+KIE(n)+KD[E (n) -2E (n-1)+E (n-2)], wherein n is sampling sequence number, and E (n) is the inclined of n times sampling Difference；U (n) is the control amount exported when n times sample, and when gas flow mutates, the flow velocity that flow sensor measures arrive is missed Difference can increase suddenly, then reduce rapidly, and need to introduce neural network algorithm.
Preferably, in the S2 step, the neural network algorithm model between deviation and control amount is established using MATLAB And it is debugged, YSTo input setting value, by output valve YoWith input value YSIt is handled by neural network, obtains Xi(t)(i =l, 2,3) three quantity of states, finally obtain Yo by learning regulation neuron weight ω i (t) (i=l, 2,3).
Preferably, the algorithm of the neural network algorithm model for the breathing blower established in the S2 step are as follows: X1=E (n)- E(n-1),X2=E (n), X3=E (n) -2E (n-1)+E (n-2), U (n)=ω1X1+ω2X2+ω3X3；
The excitation function of neural network algorithm can be logarithmic function logsig, tangent function tangsig and purely linear letter Number or in which any combination, ωiAutomatic adjusument can be carried out in actual moving process, and is changed according to following algorithm Become, wherein ηPηIηDRespectively ratio, integral, the learning rate of differential,
ω1(k+1)=ω1(k)+ηpu(k)e(k)x1(k)
ω2(k+1)=ω2(k)+ηIu(k)e(k)x2(k)
ω3(k+1)=ω3(k)+ηDu(k)e(k)x3(k),
One group of weight of any non-zero is given, such as: (ω1,ω2,ω3)=(0.1,0.2,0.7), using MATLAB into Row debugging, the final size for determining learning rate coefficient, thus obtains the neural network algorithm model of deviation and control amount.
Preferably, the data acquired in S1 are inputted in corresponding neural network algorithm model, by neural network algorithm Model is calculated with PID, then data that you can get it are precisely controlled the pressure output of ventilator blower.
Preferably, the data that neural network algorithm obtains carry out control to blower and guarantee the practical airway pressure force value of user Kurtosis is small, and extremum is small, fluctuates small, and steady state is mutually presented exhaling, inhales in airway pressure waveform, hence it is evident that improvement ventilator blower pressure The accuracy of power output.
(3) beneficial effect
The beneficial effects of the present invention are:
The pressure output control method of the noninvasive ventilator blower, regulatory PID control method occur to dash forward to when gas flow When change, flow sensor measures to flow rate error can increase, then reduce rapidly suddenly the case where be difficult to handle, will affect and exhale Handoff procedure is accurate between atmospheric pressure and pressure of inspiration(Pi), by neural network algorithm to carrying out data between deviation and control amount Processing, the pressure control data precision obtained is higher, while the practical airway pressure force value kurtosis of user is small, and extremum is small, fluctuation Small, steady state is mutually presented exhaling, inhale in airway pressure waveform, hence it is evident that improve the accuracy of ventilator blower pressure output, so that The wind pressure of ventilator blower output can quickly and accurately switch between breath pressure (EPAP) and pressure of inspiration(Pi) (IPAP), Auto-control is made according to patient airway flow and pressure amplitude, the stress level for enabling ventilator blower to export is multiple Angle suits the needs of the state of an illness and the requirement for the treatment of.
Detailed description of the invention
Fig. 1 is neural network algorithm PID control structural schematic diagram of the present invention；
Fig. 2 is regulatory PID control and PID control change curve schematic diagram neural network based.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other Embodiment shall fall within the protection scope of the present invention.
As shown in Figs. 1-2, the present invention provides a kind of technical solution: the pressure output control method of noninvasive ventilator blower, Include the following steps；
S1, processing is acquired to the data of breath pressure in actual use and pressure of inspiration(Pi)；
S2, the neural network algorithm model for establishing breathing blower；
The application of S3, neural network algorithm model.
In S1 step, PID controller is directly initiated, obtains the relational expression between control amount and deviation are as follows:
Wherein E (t) is deviation, and U (t) is control amount, KP, KI, KDRespectively ratio, integral and differential coefficient.
When sampling is packed up short enough, need by PID control process discretization, using incremental digital PID: Δ U (n)=Kp ΔE(n)+KIE(n)+KD[E (n) -2E (n-1)+E (n-2)], wherein n is sampling sequence number, and E (n) is the deviation of n times sampling；U(n) For the control amount exported when n times sampling, when gas flow mutates, the flow rate error that flow sensor measures arrive can be unexpected It increases, then reduces rapidly, need to introduce neural network algorithm.
In S2 step, the neural network algorithm model established between deviation and control amount using MATLAB is simultaneously debugged, YSTo input setting value, by output valve YoWith input value YSIt is handled by neural network, obtains Xi(t) (i=l, 2,3) three A quantity of state finally obtains Yo by learning regulation neuron weight ω i (t) (i=l, 2,3).
The algorithm of the neural network algorithm model for the breathing blower established in S2 step are as follows: X1=E (n)-E (n-1), X2=E (n),X3=E (n) -2E (n-1)+E (n-2), U (n)=ω1X1+ω2X2+ω3X3；
The excitation function of neural network algorithm can be logarithmic function logsig, tangent function tangsig and purely linear letter Number or in which any combination, ωiAutomatic adjusument can be carried out in actual moving process, and is changed according to following algorithm Become, wherein ηPηIηDRespectively ratio, integral, the learning rate of differential,
ω1(k+1)=ω1(k)+ηpu(k)e(k)x1(k)
ω2(k+1)=ω2(k)+ηIu(k)e(k)x2(k)
ω3(k+1)=ω3(k)+ηDu(k)e(k)x3(k),
One group of weight of any non-zero is given, such as: (ω1,ω2,ω3)=(0.1,0.2,0.7), using MATLAB into Row debugging, the final size for determining learning rate coefficient, thus obtains the neural network algorithm model of deviation and control amount.
The data acquired in S1 are inputted in corresponding neural network algorithm model, by neural network algorithm model with PID is calculated, then data that you can get it are precisely controlled the pressure output of ventilator blower.
The data that neural network algorithm obtains carry out control to blower and guarantee that the practical airway pressure force value kurtosis of user is small, Extremum is small, fluctuates small, and steady state is mutually presented exhaling, inhales in airway pressure waveform, hence it is evident that improves ventilator blower pressure and exports Accuracy.
Above-described specific embodiment has carried out further the purpose of the present invention, technical scheme and beneficial effects It is described in detail, it should be understood that being not limited to this hair the foregoing is merely a specific embodiment of the invention Bright, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should be included in the present invention Protection scope within.

Claims (8)

1. the pressure output control method of noninvasive ventilator blower, it is characterised in that: include the following steps；
S1, processing is acquired to the data of breath pressure in actual use and pressure of inspiration(Pi)；
S2, the neural network algorithm model for establishing breathing blower；
The application of S3, neural network algorithm model.
2. the pressure output control method of noninvasive ventilator blower according to claim 1, it is characterised in that: the S1 step In rapid, PID controller is directly initiated, obtains the relational expression between control amount and deviation are as follows:
3. the pressure output control method of noninvasive ventilator blower according to claim 2, it is characterised in that: wherein E (t) For deviation, U (t) is control amount, KP, KI, KDRespectively ratio, integral and differential coefficient.
4. the pressure output control method of noninvasive ventilator blower according to claim 2, it is characterised in that: when sampling is received Rise it is short enough, need by PID control process discretization, using incremental digital PID: Δ U (n)=KpΔE(n)+KIE(n)+KD [E (n) -2E (n-1)+E (n-2)], wherein n is sampling sequence number, and E (n) is the deviation of n times sampling；U (n) is that n times export when sampling Control amount, when gas flow mutate when, flow sensor measures to flow rate error can increase suddenly, then rapidly drop It is low, it needs to introduce neural network algorithm.
5. the pressure output control method of noninvasive ventilator blower according to claim 1, it is characterised in that: the S2 step In rapid, the neural network algorithm model established between deviation and control amount using MATLAB is simultaneously debugged, YSFor input setting Value, by output valve YoWith input value YSIt is handled by neural network, obtains Xi(t) (i=l, 2,3) three quantity of states, pass through Learning regulation neuron weight ω i (t) (i=l, 2,3) finally obtains Yo.
6. the pressure output control method of noninvasive ventilator blower according to claim 1, it is characterised in that: the S2 step The algorithm of the neural network algorithm model for the breathing blower established in rapid are as follows: X1=E (n)-E (n-1), X2=E (n), X3=E (n) -2E (n-1)+E (n-2), U (n)=ω1X1+ω2X2+ω3X3；
The excitation function of neural network algorithm can be logarithmic function logsig, tangent function tangsig and purely linear function or Any combination therein, ωiAutomatic adjusument can be carried out in actual moving process, and is changed according to following algorithm, Middle ηPηIηDRespectively ratio, integral, the learning rate of differential, ω1(k+1)=ω1(k)+ηpu(k)e(k)x1(k)
ω2(k+1)=ω2(k)+ηIu(k)e(k)x2(k)
ω3(k+1)=ω3(k)+ηDu(k)e(k)x3(k),
One group of weight of any non-zero is given, such as: (ω1,ω2,ω3)=(0.1,0.2,0.7), it is adjusted using MATLAB Examination, the final size for determining learning rate coefficient, thus obtains the neural network algorithm model of deviation and control amount.
7. the pressure output control method of noninvasive ventilator blower according to claim 6, it is characterised in that: will be adopted in S1 The data of collection input in corresponding neural network algorithm model, are calculated, can be obtained with PID by neural network algorithm model Then data out are precisely controlled the pressure output of ventilator blower.
8. the pressure output control method of noninvasive ventilator blower according to claim 7, it is characterised in that: neuron net The data that network algorithm obtains carry out control to blower and guarantee that the practical airway pressure force value kurtosis of user is small, and extremum is small, and fluctuation is small, Steady state is mutually presented exhaling, inhale in airway pressure waveform, hence it is evident that improves the accuracy of ventilator blower pressure output.
CN201910823394.3A 2019-09-02 2019-09-02 The pressure output control method of noninvasive ventilator blower Pending CN110529419A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910823394.3A CN110529419A (en) 2019-09-02 2019-09-02 The pressure output control method of noninvasive ventilator blower

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910823394.3A CN110529419A (en) 2019-09-02 2019-09-02 The pressure output control method of noninvasive ventilator blower

Publications (1)

Publication Number Publication Date
CN110529419A true CN110529419A (en) 2019-12-03

Family

ID=68666220

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910823394.3A Pending CN110529419A (en) 2019-09-02 2019-09-02 The pressure output control method of noninvasive ventilator blower

Country Status (1)

Country Link
CN (1) CN110529419A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113738662A (en) * 2020-05-28 2021-12-03 埃布姆-派斯特穆尔芬根股份有限两合公司 Method for operating a fan system and fan system having a back-curved centrifugal fan

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102799107A (en) * 2012-08-28 2012-11-28 成都乐创自动化技术股份有限公司 Stepper motor proportion integration differentiation (PID) parameter self-matching method based on micro-neural network
CN103736185A (en) * 2013-12-13 2014-04-23 科迈（常州）电子有限公司 Output pressure adjusting method of domestic respirator fan
CN105125215A (en) * 2015-10-08 2015-12-09 湖南明康中锦医疗科技发展有限公司 Neural network based breathing machine state analytic method and device
CN105843036A (en) * 2016-04-09 2016-08-10 北京工业大学 Sewage treatment process control method based on neural network
CN105973544A (en) * 2016-06-20 2016-09-28 中国民航大学 Airplane full static pressure test system based on adaptive PID method and method
CN206190577U (en) * 2016-08-30 2017-05-24 沈阳新松医疗科技股份有限公司 Noninvasive ventilator fan control ware
US20170326316A1 (en) * 2003-08-14 2017-11-16 New York University System and Method for Diagnosis and Treatment of a Breathing Pattern of a Patient
CN108066860A (en) * 2016-11-16 2018-05-25 北京航天长峰股份有限公司 A kind of lung ventilator leak compensation method
CN108211075A (en) * 2017-12-15 2018-06-29 湖南明康中锦医疗科技发展有限公司 The method for stabilizing voltage of lung ventilator wind turbine
CN108619597A (en) * 2017-03-17 2018-10-09 深圳市大雅医疗技术有限公司 A kind of pressure compensating method of lung ventilator runner

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170326316A1 (en) * 2003-08-14 2017-11-16 New York University System and Method for Diagnosis and Treatment of a Breathing Pattern of a Patient
CN102799107A (en) * 2012-08-28 2012-11-28 成都乐创自动化技术股份有限公司 Stepper motor proportion integration differentiation (PID) parameter self-matching method based on micro-neural network
CN103736185A (en) * 2013-12-13 2014-04-23 科迈（常州）电子有限公司 Output pressure adjusting method of domestic respirator fan
CN105125215A (en) * 2015-10-08 2015-12-09 湖南明康中锦医疗科技发展有限公司 Neural network based breathing machine state analytic method and device
CN105843036A (en) * 2016-04-09 2016-08-10 北京工业大学 Sewage treatment process control method based on neural network
CN105973544A (en) * 2016-06-20 2016-09-28 中国民航大学 Airplane full static pressure test system based on adaptive PID method and method
CN206190577U (en) * 2016-08-30 2017-05-24 沈阳新松医疗科技股份有限公司 Noninvasive ventilator fan control ware
CN108066860A (en) * 2016-11-16 2018-05-25 北京航天长峰股份有限公司 A kind of lung ventilator leak compensation method
CN108619597A (en) * 2017-03-17 2018-10-09 深圳市大雅医疗技术有限公司 A kind of pressure compensating method of lung ventilator runner
CN108211075A (en) * 2017-12-15 2018-06-29 湖南明康中锦医疗科技发展有限公司 The method for stabilizing voltage of lung ventilator wind turbine

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
王军琴: "基于神经网络的增量PID算法及仿真研究", 《现代电子技术》 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113738662A (en) * 2020-05-28 2021-12-03 埃布姆-派斯特穆尔芬根股份有限两合公司 Method for operating a fan system and fan system having a back-curved centrifugal fan

Similar Documents

Publication Publication Date Title
CN103893865B (en) A kind of method of lung ventilator turbine volume controlled ventilation
CN103028170B (en) A kind of respiratory pressure fuzzy control type respirator and respiratory pressure fuzzy control method
CN102770070B (en) For the system and method for real-time assessment mechanics of lung
EP3845259B1 (en) Ventilator apparatus
RU2656528C2 (en) System and method for controlling airway gas parameters during high frequency positive pressure ventilation
JP2014506163A (en) System and method for providing forced inhalation exhalation to a subject
CN206152034U (en) Follow control system who breathes apparatus of oxygen supply
JP2014506163A5 (en)
CN102802710A (en) Servo ventilation using negative pressure support
WO2014101548A1 (en) Turbine ventilator pressure-controlled ventilation method
CN106139338A (en) Lung ventilator
CN109718440A (en) Reduce the method and system of Breathing Suppotion equipment man-machine confrontation
CN103191503A (en) Pressure control device and pressure control method of respirator
CN114177451B (en) Control method for single-breathing cycle pressure-capacity double-control mode of breathing machine
CN110529419A (en) The pressure output control method of noninvasive ventilator blower
JP2015520648A (en) Cough assistance and measurement system and method
CN110755077B (en) Lung elasticity coefficient measuring method and system based on breathing device
EP3383464B1 (en) Method of co2 measurement during non-invasive ventilation
CN106267493B (en) Medical breathing machine and its method of continuous measuring and calculating resistance of respiratory tract and compliance
Živčák et al. A portable BVM-based emergency mechanical ventilator
CN114939212A (en) Man-machine synchronous breathing device
US20230169888A1 (en) System for simulating the breathing of a living being
CN203379444U (en) Breathing machine highly synchronous with breathing
CN102178993B (en) Medical breathing machine in air duct positive-pressure high-frequency ventilating mode
CN204950950U (en) Multiple breathing gas strength of mainstream formula and synchronous monitoring devices of breathing atmospheric pressure

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20191203