CN117055638A - Fan rotating speed control method and device of medical ventilation equipment and fan system - Google Patents
Fan rotating speed control method and device of medical ventilation equipment and fan system Download PDFInfo
- Publication number
- CN117055638A CN117055638A CN202311112343.2A CN202311112343A CN117055638A CN 117055638 A CN117055638 A CN 117055638A CN 202311112343 A CN202311112343 A CN 202311112343A CN 117055638 A CN117055638 A CN 117055638A
- Authority
- CN
- China
- Prior art keywords
- fan
- rotating speed
- neural network
- error
- preset
- 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
Links
- 238000009423 ventilation Methods 0.000 title claims abstract description 59
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000013528 artificial neural network Methods 0.000 claims abstract description 82
- 210000002569 neuron Anatomy 0.000 claims description 32
- 238000005259 measurement Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 7
- 230000007547 defect Effects 0.000 abstract description 5
- 230000006870 function Effects 0.000 description 14
- 230000008569 process Effects 0.000 description 13
- 230000004913 activation Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 7
- 238000005070 sampling Methods 0.000 description 6
- 125000003275 alpha amino acid group Chemical group 0.000 description 5
- 230000008859 change Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 3
- 238000009529 body temperature measurement Methods 0.000 description 2
- 150000001875 compounds Chemical class 0.000 description 2
- 230000001276 controlling effect Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000011478 gradient descent method Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000008054 signal transmission Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 206010002091 Anaesthesia Diseases 0.000 description 1
- 238000012935 Averaging Methods 0.000 description 1
- 230000003213 activating effect Effects 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000037005 anaesthesia Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- RFHAOTPXVQNOHP-UHFFFAOYSA-N fluconazole Chemical compound C1=NC=NN1CC(C=1C(=CC(F)=CC=1)F)(O)CN1C=NC=N1 RFHAOTPXVQNOHP-UHFFFAOYSA-N 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000005399 mechanical ventilation Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 210000000225 synapse Anatomy 0.000 description 1
- 238000002560 therapeutic procedure Methods 0.000 description 1
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
- A61M16/0057—Pumps therefor
- A61M16/0066—Blowers or centrifugal pumps
-
- 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
- A61M16/021—Devices for influencing the respiratory system of patients by gas treatment, e.g. mouth-to-mouth respiration; Tracheal tubes operated by electrical means
- A61M16/022—Control means therefor
- A61M16/024—Control means therefor including calculation means, e.g. using a processor
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D13/00—Control of linear speed; Control of angular speed; Control of acceleration or deceleration, e.g. of a prime mover
- G05D13/62—Control of linear speed; Control of angular speed; Control of acceleration or deceleration, e.g. of a prime mover characterised by the use of electric means, e.g. use of a tachometric dynamo, use of a transducer converting an electric value into a displacement
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B30/00—Energy efficient heating, ventilation or air conditioning [HVAC]
- Y02B30/70—Efficient control or regulation technologies, e.g. for control of refrigerant flow, motor or heating
Abstract
The embodiment of the specification discloses a fan rotating speed control method, a device and a fan system of medical ventilation equipment. The scheme may include: acquiring a preset target rotating speed of a fan and an actual rotating speed of the fan at the current moment, and acquiring fan rotating speed control parameters of the medical ventilation equipment through a preset neural network based on the target rotating speed of the fan and the actual rotating speed of the fan; then inputting the fan rotating speed control parameters to an incremental PID controller to obtain fan voltage control parameters at the next moment; and then adjusting the rotating speed of the fan of the medical ventilation equipment according to the fan voltage control parameter. Therefore, the BP neural network can better realize the regulation of the PID regulator, overcomes the defects of larger inertia and delay of the conventional PID regulator, and ensures that the fan system can realize stable, accurate and efficient control of the fan rotating speed, thereby realizing stable, accurate and efficient control of the fan ventilation flow applied to the medical ventilation equipment.
Description
Technical Field
The application relates to the technical field of medical equipment, in particular to the technical field of medical ventilation equipment, and particularly relates to a fan rotating speed control method, device and fan system of the medical ventilation equipment.
Background
In the use of medical devices such as ventilators and anesthesia machines, mechanical ventilation is required for patients, and it is very important for patients undergoing treatment to control the ventilation amount, and precise and stable control of ventilation flow is required.
In practical applications, the ventilation flow is provided to the medical ventilator by the blower system, and therefore, it is necessary to provide an accurate and stable blower rotational speed control method to achieve accurate and stable control of the ventilation flow of the ventilated medical ventilator.
Disclosure of Invention
The embodiment of the specification provides a fan rotating speed control method, a device and a fan system of medical ventilation equipment, so as to provide an accurate and stable fan rotating speed control method, and further realize accurate and stable control of ventilation flow of the ventilation medical equipment.
In order to solve the above technical problems, the embodiments of the present specification are implemented as follows:
the fan rotating speed control method of the medical ventilation equipment provided by the embodiment of the specification comprises the following steps:
acquiring a preset target rotating speed of a fan and an actual rotating speed of the fan at the current moment;
based on the target rotating speed of the fan and the actual rotating speed of the fan, obtaining fan rotating speed control parameters of the medical ventilation equipment through a preset neural network;
inputting the fan rotating speed control parameters to an incremental PID controller to obtain fan voltage control parameters at the next moment;
and adjusting the rotating speed of the fan of the medical ventilation equipment according to the fan voltage control parameter.
The fan rotational speed control device of medical ventilation equipment that this description embodiment provided includes:
the data acquisition module is used for acquiring a preset target rotating speed of the fan and an actual rotating speed of the fan at the current moment;
the neural network module is used for obtaining fan rotating speed control parameters of the medical ventilation equipment through a preset neural network based on the fan target rotating speed and the fan actual rotating speed;
the PID control module is used for inputting the fan rotating speed control parameters to the incremental PID controller to obtain fan voltage control parameters at the next moment;
and the rotating speed adjusting module is used for adjusting the rotating speed of the fan of the medical ventilation equipment according to the fan voltage control parameter.
The embodiment of the specification provides a fan system applied to medical ventilation equipment, which is characterized by comprising a sensing measurement module, a controller and a motor module;
the sensing measurement module is used for collecting the actual rotating speed of the fan;
the controller comprises a neural network unit and a PID control unit; the fan rotating speed control method is used for receiving the actual rotating speed of the fan acquired by the sensor module and executing the fan rotating speed control method of the medical ventilation equipment in the embodiment of the specification;
the motor module is used for responding to the controller to control the fan and adjusting the rotating speed of the fan.
One embodiment of the present disclosure can achieve at least the following advantages: obtaining a fan speed control parameter of the medical ventilation equipment through a preset neural network by obtaining a preset fan target speed and a fan actual speed at the current moment and based on the fan target speed and the fan actual speed; then inputting the fan rotating speed control parameters to an incremental PID controller to obtain fan voltage control parameters at the next moment; and then the fan rotating speed of the medical ventilation equipment is adjusted according to the fan voltage control parameter, so that the regulation of the PID regulator can be better realized by using the BP neural network, the defects of larger inertia and delay of the conventional PID regulator are overcome, the fan system can realize stable, accurate and efficient control of the fan rotating speed, and further realize stable, accurate and efficient control of the fan ventilation flow applied to the medical ventilation equipment.
Drawings
In order to more clearly illustrate the embodiments of the present description or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some embodiments described in the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of a blower system for a medical ventilator according to an embodiment of the present disclosure;
fig. 2 is a schematic flow chart of a fan speed control method applied to a medical ventilation device according to an embodiment of the present disclosure;
fig. 3 is a schematic structural diagram of a BP neural network applied to a fan controller of a medical ventilator according to an embodiment of the present disclosure;
fig. 4 is a schematic diagram of a control principle of a PID controller based on a BP neural network applied to a blower of a medical ventilator according to an embodiment of the present disclosure;
fig. 5 is a schematic diagram of an execution flow of a PID control algorithm based on a BP neural network applied to a medical ventilator in a practical application scenario provided in an embodiment of the present disclosure;
fig. 6 is a fan speed control apparatus corresponding to the fan speed control method of the medical ventilator of fig. 2 according to an embodiment of the present disclosure.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of one or more embodiments of the present specification more clear, the technical solutions of one or more embodiments of the present specification will be clearly and completely described below in connection with specific embodiments of the present specification and corresponding drawings. It will be apparent that the described embodiments are only some, but not all, of the embodiments of the present specification. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without undue burden, are intended to be within the scope of one or more embodiments herein.
In order to facilitate understanding of the embodiments of the present application, an application scenario related to the present application is described below, with reference to fig. 1.
A schematic structural diagram of a blower system applied to a medical ventilator according to an embodiment of the present specification is shown in fig. 1.
In practice, the blower system may be used to provide ventilation flow to a ventilation medical device. In embodiments of the present description, the medical ventilator device may include a continuous positive airway pressure (Continuous Positive Airway Pressure, CPAP) device, a high flow humidification therapy device, or the like, and is not limited to the examples given herein.
As shown in fig. 1, the fan system applied to the medical ventilator according to the embodiment of the present disclosure may include an algorithm control module 102, a fan driving module 103 connected to the algorithm control module 102, a fan 104 connected to the fan driving module 103, and a fan rotation speed measuring sensor 105 connected to the fan 104, where the fan rotation speed measuring sensor 105 is further connected to the algorithm control module.
In actual application, in one aspect, the algorithm control module 102 may receive a fan target speed input by a user. Alternatively, it may be entered by the user through the user interface module 101 into the algorithm control module. On the other hand, the algorithm control module 102 may receive the actual rotational speed of the blower from the blower rotational speed measurement sensor 105 through real-time measurement of the blower 104. In the algorithm control module 102, fan voltage control parameters input to the fan drive module 103 may be adjusted based on a fan target speed and a fan actual speed.
In alternative embodiments, the blower 104 may comprise a brushless turbine blower. The fan driving module 103 may be a BLDC (Brushless Direct Current Motor) brushless dc motor driving circuit. The fan voltage control parameters input to the fan driving module 103 may specifically include a PWM duty cycle.
In an alternative embodiment, a fan temperature measurement sensor 106 coupled to the fan 104 and the algorithm control module 102 may also be included in the fan system. Specifically, the fan temperature measurement sensor 106 may collect fan temperature information during the operation of the fan 104, and send the fan temperature information to the algorithm control module 102, where a high temperature early warning program may be preset in the algorithm control module 102. For example, when the fan temperature is greater than a preset temperature threshold, the algorithm control module 102 may control the fan 104 to stop running through the fan driving module 103.
In addition, in the embodiment of the present disclosure, a power module 107 connected to the user interface module 101, the algorithm control module 102, and the fan driving module 103 may be further included in the fan system.
In the embodiment of the present disclosure, the algorithm control module 102 may specifically include a PID control algorithm based on a neural network, and may include a neural network unit and a PID control unit. Specifically, the controller can utilize the BP neural network to realize on-line setting of control parameters of the PID controller, so that the PID controller outputs stable actual control output quantity, and accurate and stable control of a fan applied to the medical ventilation equipment is realized, and further accurate and stable control of ventilation flow of the medical ventilation equipment is realized.
In the prior art, the traditional mode of controlling the rotation speed of the brushless turbine fan is to adopt an incremental PID or a position PID. The traditional PID control has the defects of insufficient control precision, static error, low convergence speed, low robustness and the like.
In the embodiment of the present specification, a neural network (for example, BP neural network) is added on the basis of the conventional control PID mode. Because the neural network has the capability of approaching any nonlinear function, the PID control scheme established by adopting the neural network structure can overcome the defects of insufficient control precision, static error, slow convergence speed, weak robustness and the like of the traditional PID control scheme.
The fan rotational speed control method of the medical ventilator provided in the embodiment of the present specification will be described from the viewpoint of the controller 103.
Referring to fig. 2, a flow chart of a fan speed control method applied to a medical ventilator according to an embodiment of the present disclosure is shown.
In practical applications, the method shown in fig. 2 may be implemented by a computing service device having data processing and program running functions. Optionally, the computing service device may also have network communication functionality.
In fig. 2, the fan speed control method of the medical ventilator may include the steps of:
step 202: and acquiring a preset target rotating speed of the fan and an actual rotating speed of the fan at the current moment.
The target rotation speed of the fan can be the rotation speed of the fan determined according to the actual ventilation flow demand of the medical ventilation equipment. The fan target rotation speed may be preset, specifically, the fan target rotation speed may be a fixed value calculated in advance according to a preset actual ventilation flow demand, or the fan target rotation speed may be a value calculated in real time according to an actual working condition according to a rotation speed function predetermined according to the preset actual ventilation flow demand.
The actual rotation speed of the fan can be the rotation speed of the fan actually measured by the sensing measurement module in the operation process of the fan. In the running process of the fan, the sensing measurement module can continuously measure the actual rotating speed of the fan.
In an ideal situation, the actual fan speed may be consistent (e.g., equal) to the target fan speed. In practical applications, however, there is often a deviation between the actual rotational speed of the fan and the target rotational speed of the fan, i.e., a rotational speed error.
Step 204: and obtaining the fan rotating speed control parameter of the medical ventilation equipment through a preset neural network based on the fan target rotating speed and the fan actual rotating speed.
In the embodiments of the present description, a PID controller is used to control the fan speed. In practical application, in order to obtain a better control effect of PID control, three control actions of proportion, integral and derivative are required to be adjusted to form a relationship of mutual coordination and mutual restriction in control quantity. The relationship between the proportional, integral and differential control actions is not typically a simple "linear combination", and it is desirable to find the preferred combination from a non-linear combination that varies infinitely. Given the ability of neural networks to have any nonlinear expression, in embodiments of the present description, PID control with an optimal combination can be achieved through learning of system performance.
In step 204, the fan speed control parameters may include a scaling factor k for input to an incremental PID controller p Integral coefficient k i And differential coefficient k d 。
In an alternative embodiment, the parameter k is established p 、k i 、k d In the self-learning PID controller process, the preset neural network may include a BP neural network or an RBF neural network.
The BP neural network algorithm imitates the signal transmission process between neurons of the living beings in nature, including activation of the neurons and transmission of signals through synapses, so that the BP neural network algorithm has the characteristics of strong learning ability and adaptation to the dynamic characteristics of an uncertainty system. In the embodiment of the present specification, the weight coefficient can be adjusted through self-learning of the neural network according to the corresponding characteristic of the control object, so as to adjust the PID parameter, and realize the optimized control performance.
The neural network-based controller of the embodiment of the present specification will be described below by taking a BP neural network as an example.
Fig. 3 shows a schematic structural diagram of a BP neural network applied to a fan controller of a medical ventilator according to an embodiment of the present disclosure.
As shown in fig. 3, the BP neural network structure of the embodiment of the present disclosure includes three-level network architecture of an input layer, an hidden layer, and an output layer. Wherein the input layer comprises M neurons, the hidden layer comprises Q neurons, and the output layer comprises N neurons. For example, the input layer may have 4 neurons, the hidden layer may have 5 neurons, and the output layer may have 3 neurons.
Specifically, the learning process of the BP neural network may include a forward propagation process of information and a backward propagation process of an error.
In the embodiment of the present specification, for the BP neural network algorithm forward propagation process, a rotation speed error may be determined based on the fan target rotation speed and the fan actual rotation speed; and inputting the target rotating speed of the fan, the actual rotating speed of the fan and the rotating speed error into a preset neural network to obtain the output fan rotating speed control parameters of the preset neural network.
Wherein optionally, the determining a rotational speed error based on the target rotational speed of the fan and the actual rotational speed of the fan may specifically include: determining a rotation speed difference value between the target rotation speed of the fan and the actual rotation speed of the fan; and determining the absolute value of the rotating speed difference value as the rotating speed error.
In the BP neural network used in the embodiments of the present specification, the input of the input layer neurons can be expressed as:the output of the input layer neurons can be expressed as: />
Wherein j may be a natural number selected from 1 to M;representing the jth electrical signal input to the input layer neuron;a j-th electrical signal representing the output of the input layer neuron; the values (1), (2) and (3) in the upper right brackets represent neurons of the input, hidden and output layers, respectively, in the neural network model.
Alternatively, the number of neurons of the input layer may be set to m=4, x 1 、x 2 、x 3 And x 4 Representing four different input signals, respectively.
In practical application, x 1 、x 2 、x 3 And x 4 Can be set to rin, yout, error and constant 1, respectively. Wherein, rin is the set target rotating speed of the fan; yout is the actual rotation speed of the fan; error is fan with actual rotation speed and set valueAn error value between the target rotational speeds; 1 is the error constant for nonlinear adjustment.
In the BP neural network used in the embodiment of the present specification, the input of the intermediate hidden layer neurons may be:the output of the intermediate hidden layer neurons may be: />
Wherein i may be a natural number selected from 1 to Q;an ith electrical signal representing an input to the intermediate hidden layer neuron; />An ith electrical signal representing the output of the intermediate hidden layer neuron; />Representing hidden layer weight values from a jth node of an input layer to an ith node of an intermediate hidden layer; f [. Cndot.]Representing the activation function of the intermediate hidden layer for converting the sum of the input signals into an output signal.
Alternatively, as shown in fig. 3, the number of neurons of the intermediate hidden layer may be set to q=5. In practical application, the number of neurons of the intermediate hidden layer can be adjusted according to the requirement.
Alternatively, the activation function f [. Cndot ] of the intermediate hidden layer may be specifically a tanh function, i.e., a hyperbolic tangent function. The specific expression of f is:
the derivative of the f [ cndot ] activation function is:
wherein e is a natural constant; e, e x And e -x X and-x in (2) are indices of a natural constant e, respectively.
In the BP neural network used in the embodiment of the present specification, the input of the output layer neuron may be:the output of the output layer neurons may be: />
Wherein l may be a natural number selected from 1 to N;a first electrical signal representing an input to the intermediate hidden layer neuron; />A first electrical signal representing the output of the intermediate hidden layer neuron; />Indicating the hidden layer weight value from the ith node of the intermediate hidden layer to the ith node of the output layer; g []Representing the activation function of the output layer for converting the sum of the input signals into an output signal.
Alternatively, as shown in fig. 3, the number of neurons of the output layer may be set to n=3.
In practical application, three adjustable parameters, k respectively, are provided in the PID regulator p 、k i 、k d Can be connected with three output nodes of the output layerAnd->Corresponding toThe specific correspondence may be:
alternatively, the activation function g [ cndot ] of the output layer may be specifically a function based on tanh. The specific expression of g [. Cndot. ] can be:
the derivative of the g [. Cndot ] activation function is:
wherein e is a natural constant; e, e x And e -x X and-x in (2) are indices of a natural constant e, respectively.
In practical application, the process of the BP neural network for processing the target rotation speed of the fan, the actual rotation speed of the fan and the rotation speed error may include: inputting the target rotating speed of the fan, the actual rotating speed of the fan, the rotating speed error and a preset error constant as input values to the input layer; the input value is transmitted to the output layer through the hidden layer, so that output data of the output layer are obtained; if the output data is inconsistent with expected data corresponding to the working condition at the current moment, determining an incoming error based on the output data and the expected data; reversely transmitting the input error from the output layer to the input layer through the hidden layer, and correcting network parameters of the BP neural network until a preset training ending condition is reached; the preset training ending conditions comprise: and calculating a performance index value based on the output data of the output layer and the expected data to be smaller than a preset index threshold, or enabling the processing times of the BP neural network to the input value to reach a preset time threshold.
Wherein optionally determining an incoming error based on the output data and the desired data may comprise: respectively determining the data difference between each output data and the expected data; squaring each data difference to obtain a square value; summing all the square values to obtain a summation result; determining a result of averaging the summed result as the incoming error.
In the embodiment of the present specification, for the back propagation process of the BP neural network algorithm, the square of the error between the actual rotation speed of the fan and the target rotation speed of the fan may be used as an index for system evaluation.
Specifically, the evaluation index of the BP neural network may be set as:
E(k)＝0.5(rin(k)-yout(k)) 2 ；
wherein k represents any time or cycle during the adjustment process; rin (k) represents a preset target fan speed; yout (k) represents the actual rotational speed of the fan at the current time (i.e., time k).
In practical application, through forward propagation, an input signal is transmitted from an input layer to an output layer through an intermediate hidden layer, if a value of a performance index E (k) meets a desired condition, for example, a set threshold is reached, it may be indicated that an optimal solution of the BP neural network algorithm is reached, and the learning algorithm may be ended; otherwise, a back propagation algorithm is performed, i.e. an error signal is returned to the path, by modifying the connection weights of the neurons of each layer such that the error between the final actual output value and the desired output value is minimized.
Alternatively, in practical applications, the input signal is transmitted from the input layer to the output layer via the intermediate hidden layer by forward propagation, and if the number of iterations reaches a preset condition, for example, a set number of times threshold is reached, the learning algorithm may end.
In the embodiment of the present specification, the gradient descent method is adopted to correct the weight coefficient of the BP neural network in the back propagation process, that is, the negative gradient direction of the weight coefficient is subjected to search adjustment according to E (k); optionally, an inertia term that allows the search to converge quickly may be added to achieve the goal of a quick search.
Alternatively, the weight coefficient of the output layer may be updated according to the following formula (1):
wherein n is the learning rate; a is an inertia coefficient, also known as a momentum factor. In practical application, n and a may both be fixed values. The values of n and a may be set as desired.
In the formula (1), according to the chain law,
in the formula (2), the amino acid sequence of the formula (2),
in the formula (2), forCan be approximated by a sign function->Instead, the resulting imprecise effect can be compensated by adjusting the learning rate n;
in the formula (2), forOwing to the following formula (5)/(x)>Can be calculated as follows:
in the formula (2), the amino acid sequence of the formula (2),activating the derivative of the function for the output layer, i.e. < ->
In the formula (2), the amino acid sequence of the formula (2),i.e. the output of the intermediate hidden layer.
Thus, the weight coefficient update algorithm of the output layer can be converted from equation (1) to equation (3) below:
in the formula (3), the amino acid sequence of the compound,
similarly, the weight coefficient updating algorithm of the intermediate hidden layer can be obtained as the following formula (4):
in the formula (4), the amino acid sequence of the compound,
step 206: and inputting the fan rotating speed control parameters to an incremental PID controller to obtain fan voltage control parameters at the next moment.
The incremental PID controller is a controller that performs PID control on the increment of the control amount. Specifically, the incremental PID controller makes a difference between the control amount at the present time and the control amount at the previous time, and uses the difference as a new control amount.
The conventional digital incremental PID control algorithm directly performs closed-loop control on the object, and the control effect is represented by a proportionality coefficient k p Integral coefficient k i And differential coefficient k d Three parameters are determined.
The PID control is to calculate the proportion, integral and differential according to the deviation error (k) between the target value and the actual output value, and add the results to obtain the control output u (t), and the expression of the PID algorithm in the continuous time domain is as follows:
wherein k is p Representing the proportionality coefficient, T I Represent the integration time constant, T D Representing the differential time constant.
Discretizing the PID algorithm expression, wherein a row of time points k of sampling time represent continuous time, a sum formula replaces integration, and an increment replaces differentiation, so that the PID algorithm expression is obtained as follows:
wherein k is p Represents the proportionality coefficient, k i Represents the integral coefficient, k d Representing the differential coefficient, T representing the sampling period.
The incremental numerical PID control algorithm is represented by equation (5) and equation (6):
u (k) =u (k-1) +Δu (k) formula (5);
in the formula (5), k represents a sampling number, and u (k) represents the current output of the PID controller; u (k-1) represents the last output of the PID controller; Δu (k) represents the current output change amount of the PID controller.
Δu(k)＝k p (error(k)-error(k-1))+k i error(k)+k d (error(k)-2error(k-1)+error(k-2))
Formula (6);
in formula (6), k p Representing a scaling factor; k (k) i Representing an integral coefficient; k (k) d Representing the differential coefficient; error (k) represents the error amount of the actual value (e.g., the actual fan speed) and the target value (e.g., the target fan speed) at the present (i.e., kth) sampling time; error (k-1) represents the error amount between the actual value (e.g., the actual fan speed) and the target value (e.g., the target fan speed) at the last (i.e., the k-1 th) sampling time; error (k-2) represents the amount of error between the actual value (e.g., the actual fan speed) and the target value (e.g., the target fan speed) at the last (i.e., the k-2) th sampling time.
In the embodiments of the present description, a neural network-based PID controller is used to adjust PID control parameters in conjunction with a neural network on the basis of a conventional digital incremental PID control algorithm.
Fig. 4 is a schematic diagram showing a control principle of a PID controller based on a BP neural network applied to a blower of a medical ventilator according to an embodiment of the present specification.
As shown in fig. 4, in step 206, the fan speed control parameter is input to an incremental PID controller to obtain a fan voltage control parameter at the next moment, which may specifically include: and inputting the fan rotating speed control parameter and the rotating speed error into an incremental PID controller to obtain the fan voltage control parameter at the next moment.
In an alternative embodiment, the error efficiency may also be determined based on the rotational speed error. For example, based on the error, the efficiency of the change of the error with time, i.e., the error efficiency, is calculated and denoted as de/dt.
Thus, as shown in fig. 4, in step 206, the fan speed control parameter is input to an incremental PID controller to obtain a fan voltage control parameter at the next moment, which may specifically include: and inputting the fan rotating speed control parameter, the rotating speed error and the error efficiency into an incremental PID controller to obtain a fan voltage control parameter at the next moment.
Step 208: and adjusting the rotating speed of the fan of the medical ventilation equipment according to the fan voltage control parameter.
In practical applications, in step 206, obtaining the fan voltage control parameter at the next moment may specifically include obtaining the PWM duty cycle of the fan driving circuit at the next moment. Accordingly, in step 208, the fan rotation speed of the medical ventilator is adjusted according to the fan voltage control parameter, which may specifically be that the fan rotation speed of the medical ventilator is adjusted according to the PWM duty cycle of the fan driving circuit.
The PWM (pulse width modulation) (Pulse width modulation) is an analog control mode, and the bias of the base electrode or the grid electrode of the transistor is modulated according to the corresponding load change to change the on time of the transistor or the MOS transistor, so that the output of the switching regulated power supply is changed. This way, the output voltage of the power supply can be kept constant when the operating conditions change, and is a very effective technique for controlling the analog circuit by means of the digital signal of the microprocessor.
Duty cycle refers to the ratio of the excitation time to the total cycle time in a discontinuous, continuous or short-term operating regime. PWM duty cycle is a typical application of duty cycle, which refers to the ratio of the time the high level is maintained to the modulation period of the PWM in the output PWM waveform, such as: the frequency of the PWM is 1000Hz, the corresponding modulation period is 1ms, if the time of the high level is 200us, the time of the low level is 800us, the duty ratio is 200:1000, and the duty ratio of the PWM is 1/5.
Based on one or more embodiments in the specification, obtaining a preset fan target rotating speed and a fan actual rotating speed at the current moment, and obtaining a fan rotating speed control parameter of the medical ventilation equipment through a preset neural network based on the fan target rotating speed and the fan actual rotating speed; then inputting the fan rotating speed control parameters to an incremental PID controller to obtain fan voltage control parameters at the next moment; and then the fan rotating speed of the medical ventilation equipment is adjusted according to the fan voltage control parameter, so that the regulation of the PID regulator can be better realized by using the BP neural network, the defects of larger inertia and delay of the conventional PID regulator are overcome, the fan system can realize stable, accurate and efficient control of the fan rotating speed, and further realize stable, accurate and efficient control of the fan ventilation flow applied to the medical ventilation equipment.
According to the above description, the embodiment of the present disclosure provides a schematic execution flow chart of a PID control algorithm based on a BP neural network applied to a medical ventilator in a practical application scenario, as shown in fig. 5.
In fig. 5, step 501: initializing parameters of a control system and BP neural network.
Specifically, the weight coefficient of the initial BP neural network may be set.
Specifically, an initial input value and an initial output value may be set. For example, the initial yout value and error value may be set to zero.
Specifically, the values of the learning rate n and the momentum factor coefficient/inertia coefficient a in correcting the weight coefficient by the gradient descent method may also be set.
Step 502: forward propagation computation.
Specifically, the output values of nodes of an implicit layer and an output layer of the BP neural network can be calculated to obtain PID control parameters.
Step 503: and calculating the deviation between the actual value and the target value.
Specifically, the PID control parameters can be input to an incremental PID controller, and the control output is calculated; further, a controlled object (for example, a fan) is controlled based on the control output, and an actual output value of the controlled object is obtained. This can obtain a deviation between the actual value and the target value.
Step 504: back propagation computation.
Specifically, the inverse error may be calculated, thereby correcting the weight coefficient of the output layer and the weight coefficient of the hidden layer.
Step 505: judging whether to end the learning of the neural network parameters according to the preset training ending conditions, if yes, ending the flow of fig. 5, and if not, repeatedly executing the steps 502 to 504.
In the prior art
Based on a similar concept as the above-described embodiments, embodiments of the present specification also provide a fan speed control apparatus corresponding to the fan speed control method of the medical ventilator of fig. 2, as shown in fig. 6.
In fig. 6, a fan rotation speed control apparatus of a medical ventilator may include:
the data acquisition module 602 is configured to acquire a preset target rotation speed of the fan and an actual rotation speed of the fan at a current moment;
the neural network module 604 is configured to obtain, based on the target rotation speed of the fan and the actual rotation speed of the fan, a fan rotation speed control parameter of the medical ventilation device through a preset neural network;
the PID control module 606 is configured to input the fan rotation speed control parameter to an incremental PID controller, to obtain a fan voltage control parameter at a next moment;
the rotation speed adjustment module 608 is configured to adjust a rotation speed of the fan of the medical ventilation device according to the fan voltage control parameter.
Optionally, the neural network module 604 may specifically be configured to: determining a rotational speed error based on the target rotational speed of the fan and the actual rotational speed of the fan; and inputting the target rotating speed of the fan, the actual rotating speed of the fan and the rotating speed error into a preset neural network to obtain the output fan rotating speed control parameters of the preset neural network.
Alternatively, the preset neural network may include a BP neural network or an RBF neural network.
Optionally, the BP neural network may include an input layer, an hidden layer, and an output layer. And, the process of the BP neural network for processing the target rotation speed of the fan, the actual rotation speed of the fan and the rotation speed error may specifically include: inputting the target rotating speed of the fan, the actual rotating speed of the fan, the rotating speed error and a preset error constant as input values to the input layer; the input value is transmitted to the output layer through the hidden layer, so that output data of the output layer are obtained; if the output data is inconsistent with expected data corresponding to the working condition at the current moment, determining an incoming error based on the output data and the expected data; reversely transmitting the input error from the output layer to the input layer through the hidden layer, and correcting network parameters of the BP neural network until a preset training ending condition is reached; the preset training ending conditions comprise: and calculating a performance index value based on the output data of the output layer and the expected data to be smaller than a preset index threshold, or enabling the processing times of the BP neural network to the input value to reach a preset time threshold.
Optionally, the input layer has 4 neurons, the hidden layer has 5 neurons, and the output layer has 3 neurons.
Optionally, the fan rotation speed control device may be further configured to: based on the rotational speed error, an error efficiency is determined. Accordingly, the PID control module 606 may be specifically configured to: and inputting the fan rotating speed control parameter, the rotating speed error and the error efficiency into an incremental PID controller to obtain a fan voltage control parameter at the next moment.
Alternatively, the fan voltage control parameter may specifically include a PWM duty cycle of the fan driving circuit.
It will be appreciated that each of the modules described above refers to a computer program or program segment for performing one or more particular functions. Furthermore, the distinction of the above-described modules does not represent that the actual program code must also be separate.
Based on the same thought, the embodiment of the specification also provides a fan rotating speed control device of the medical ventilation device, which corresponds to the method.
In practical application, the fan rotation speed control device of the medical ventilation device may include:
at least one processor; the method comprises the steps of,
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to:
acquiring a preset target rotating speed of a fan and an actual rotating speed of the fan at the current moment;
based on the target rotating speed of the fan and the actual rotating speed of the fan, obtaining fan rotating speed control parameters of the medical ventilation equipment through a preset neural network;
inputting the fan rotating speed control parameters to an incremental PID controller to obtain fan voltage control parameters at the next moment;
and adjusting the rotating speed of the fan of the medical ventilation equipment according to the fan voltage control parameter.
Based on the same thought, the embodiment of the specification also provides a fan system applied to the medical ventilation equipment, which corresponds to the method.
In practical applications, the fan system applied to the medical ventilation device may include: the sensing measurement module, the controller and the motor module;
the sensing measurement module is used for collecting the actual rotating speed of the fan;
the controller comprises a neural network unit and a PID control unit; the fan speed control method is used for receiving the actual speed of the fan acquired by the sensor module and executing the medical ventilation equipment in one or more embodiments;
the motor module is used for responding to the controller to control the fan and adjusting the rotating speed of the fan.
The foregoing describes specific embodiments of the present disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results.
The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and variations of the present application will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the application are to be included in the scope of the claims of the present application.
Claims (10)
1. A method for controlling the rotational speed of a fan of a medical ventilation device, comprising:
acquiring a preset target rotating speed of a fan and an actual rotating speed of the fan at the current moment;
based on the target rotating speed of the fan and the actual rotating speed of the fan, obtaining fan rotating speed control parameters of the medical ventilation equipment through a preset neural network;
inputting the fan rotating speed control parameters to an incremental PID controller to obtain fan voltage control parameters at the next moment;
and adjusting the rotating speed of the fan of the medical ventilation equipment according to the fan voltage control parameter.
2. The method of claim 1, wherein the obtaining, based on the target rotational speed of the blower and the actual rotational speed of the blower, a blower rotational speed control parameter of the medical ventilator through a preset neural network specifically includes:
determining a rotational speed error based on the target rotational speed of the fan and the actual rotational speed of the fan;
and inputting the target rotating speed of the fan, the actual rotating speed of the fan and the rotating speed error into a preset neural network to obtain the output fan rotating speed control parameters of the preset neural network.
3. The method of claim 2, wherein the preset neural network comprises a BP neural network or an RBF neural network.
4. The method of claim 3, wherein the BP neural network comprises an input layer, an hidden layer, and an output layer; the input layer has 4 neurons, the hidden layer has 5 neurons, and the output layer has 3 neurons.
5. The method of claim 4, wherein the processing of the fan target speed, the fan actual speed, and the speed error by the BP neural network comprises:
inputting the target rotating speed of the fan, the actual rotating speed of the fan, the rotating speed error and a preset error constant as input values to the input layer;
the input value is transmitted to the output layer through the hidden layer, so that output data of the output layer are obtained;
if the output data is inconsistent with expected data corresponding to the working condition at the current moment, determining an incoming error based on the output data and the expected data;
reversely transmitting the input error from the output layer to the input layer through the hidden layer, and correcting network parameters of the BP neural network until a preset training ending condition is reached; the preset training ending conditions comprise: and calculating a performance index value based on the output data of the output layer and the expected data to be smaller than a preset index threshold, or enabling the processing times of the BP neural network to the input value to reach a preset time threshold.
6. The method of claim 2, wherein the inputting the fan speed control parameter to the incremental PID controller, before obtaining the fan voltage control parameter for the next time, further comprises:
based on the rotational speed error, an error efficiency is determined.
7. The method of claim 6, wherein the inputting the fan speed control parameter to an incremental PID controller obtains a fan voltage control parameter at a next time, specifically comprising:
and inputting the fan rotating speed control parameter, the rotating speed error and the error efficiency into an incremental PID controller to obtain a fan voltage control parameter at the next moment.
8. The method of claim 1, wherein the fan voltage control parameter specifically comprises a PWM duty cycle of a fan drive circuit.
9. A fan rotational speed control apparatus for a medical ventilator, comprising:
the data acquisition module is used for acquiring a preset target rotating speed of the fan and an actual rotating speed of the fan at the current moment;
the neural network module is used for obtaining fan rotating speed control parameters of the medical ventilation equipment through a preset neural network based on the fan target rotating speed and the fan actual rotating speed;
the PID control module is used for inputting the fan rotating speed control parameters to the incremental PID controller to obtain fan voltage control parameters at the next moment;
and the rotating speed adjusting module is used for adjusting the rotating speed of the fan of the medical ventilation equipment according to the fan voltage control parameter.
10. A fan system applied to medical ventilation equipment is characterized by comprising a sensing measurement module, a controller and a motor module;
the sensing measurement module is used for collecting the actual rotating speed of the fan;
the controller comprises a neural network unit and a PID control unit; the fan speed control method is used for receiving the actual rotating speed of the fan acquired by the sensor module and executing any one of claims 1 to 8;
the motor module is used for responding to the controller to control the fan and adjusting the rotating speed of the fan.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311112343.2A CN117055638A (en) | 2023-08-30 | 2023-08-30 | Fan rotating speed control method and device of medical ventilation equipment and fan system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311112343.2A CN117055638A (en) | 2023-08-30 | 2023-08-30 | Fan rotating speed control method and device of medical ventilation equipment and fan system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117055638A true CN117055638A (en) | 2023-11-14 |
Family
ID=88664372
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311112343.2A Pending CN117055638A (en) | 2023-08-30 | 2023-08-30 | Fan rotating speed control method and device of medical ventilation equipment and fan system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117055638A (en) |
-
2023
- 2023-08-30 CN CN202311112343.2A patent/CN117055638A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108284442B (en) | Mechanical arm flexible joint control method based on fuzzy neural network | |
| CN108828960B (en) | Pneumatic muscle model-free high-order iterative learning control method | |
| CN102497156A (en) | Neural-network self-correcting control method of permanent magnet synchronous motor speed loop | |
| Milovanović et al. | Adaptive PID control based on orthogonal endocrine neural networks | |
| CN108696210A (en) | Direct current generator current loop controller methods of self-tuning based on parameter identification | |
| CN111856942B (en) | Time lag and error improvement method for remote intervention operation control system | |
| CN107263543A (en) | The calibration method and computing device of a kind of joint of robot | |
| CN104122795A (en) | Novel extremal function index based intelligent self-turning PID (Proportion Integration Differentiation) indoor temperature control algorithm | |
| KR20200026050A (en) | Control device and control method | |
| CN113425961B (en) | Control method and control device of turbine motor and breathing machine | |
| CN115356919B (en) | Self-adaptive adjusting method for PID controller of chlorine dioxide sterilizer | |
| CN105911862A (en) | Electric heating furnace temperature control method | |
| CN109743002B (en) | Servo system controller, feedforward control signal determination method and inertia identification method | |
| CN117055638A (en) | Fan rotating speed control method and device of medical ventilation equipment and fan system | |
| CN112704791B (en) | CPAP mode gas transmission control method of respiratory support equipment and respiratory support equipment | |
| CN111800053A (en) | Motor driving system and method | |
| CN110932629B (en) | Inertia estimation method based on single neuron network and self-adaptive adjustment strategy thereof | |
| CN116893701A (en) | Ventilation humidity control method and device for medical ventilation equipment and ventilation humidity control system | |
| CN115327888B (en) | PID algorithm of circulating water pump flow, direct-current circulating water pump and air energy heat pump | |
| CN113325720B (en) | Self-adaptive tracking control method for rehabilitation training robot with movement speed decision | |
| CN112947606A (en) | Boiler liquid level control system and method based on BP neural network PID predictive control | |
| CN113377115B (en) | Stable control method for service robot to autonomously learn transient movement time | |
| CN112072961A (en) | Brushless DC motor speed control system based on ANFIS | |
| CN219041655U (en) | Crawler speed regulating device of ladder climbing machine | |
| CN213693530U (en) | Motor drive system |
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 |