Robust output adjusting method based on normal blood glucose concentration range of human body

Abstract

The invention discloses a robust output regulation method based on a normal blood glucose concentration range of a human body, which comprises a tracking differentiator for blood glucose regulation transition, a regulator, an insulin pump, an expansion state observer and a sensor, the regulator is respectively connected with the tracking differentiator, the insulin pump and the expansion state observer, the insulin pump is used for injection, and the expansion state observer is connected with the sensor. And the sensor is connected with an object injected by the insulin pump. According to the invention, the blood sugar is regulated to a normal blood sugar concentration range of a human body, and specifically, the blood sugar of a patient can be effectively regulated to a normal blood sugar range of 70-110mg / dl of the human body.

Description

Technical Field

[0001] The invention relates to the technical field of insulin pump blood sugar regulation, in particular to a robust output regulation method based on a normal blood sugar concentration range of a human body. Background Art

[0002] Diabetes is a metabolic disease characterized by chronic hyperglycemia caused by defects in insulin secretion and / or action. According to data from the International Diabetes Federation (IDF), as of 2019, China has the largest number of diabetic patients (20-79 years old) in the world, reaching 116.4 million, and 35.5 million diabetic patients over the age of 65 (https: / / diabetesatlas.org / en / ). According to the diagnostic criteria of the American Diabetes Association (ADA), the prevalence of diabetes in my country is as high as 12.8%, the proportion of prediabetes in Chinese adults is 35.2%, and the estimated number of diabetic patients is 129.8 million.

[0003] Insulin therapy is one of the important means of treating diabetes. As one of the intensive insulin treatment options, continuous subcutaneous insulin infusion (CSII) can simulate the physiological secretion pattern of human insulin to the greatest extent, which is conducive to achieving blood sugar control in diabetic patients. Nathan DM pointed out in the article "Intensive diabetes treatment and cardiovascular disease in patients with type 1 diabetes" that the results of the Diabetes Control and Complications Trial showed that intensive insulin therapy including CSII can significantly reduce glycated hemoglobin levels, reduce complications, and prolong the life of patients. This study provides solid clinical evidence for the clinical application of CSII. Murphy Helen's research results in the article "Safety and Efficacy of 24-h Closed-Loop Insulin Delivery in Well-Controlled Pregnant Women With Type 1 Diabetes" showed that in a study of pregnant women with type 1 diabetes, it was found that the efficacy of closed-loop insulin pump devices was comparable to that of open-loop insulin pump devices, but it could significantly reduce the occurrence of hypoglycemia. The research results of Zoe A. Stewart in the article "Closed-Loop Insulin Delivery during Pregnancy in Women with Type 1 Diabetes" further show that whether before or after delivery, closed-loop insulin pump therapy can more effectively control blood sugar than sensor-enhanced insulin pumps.

[0004] According to the "Guidelines for Insulin Pump Therapy in China (2021 Edition)", insulin pumps have been in the Chinese market for more than 20 years. To date, about 3,000 hospitals have carried out insulin pump therapy. It is estimated that the average number of patients receiving short-term intensive insulin pump therapy each year has exceeded one million, and there are currently nearly 60,000 patients using insulin pumps for a long time. Currently, the new closed-loop insulin pump that integrates real-time continuous glucose monitoring (CGM) and CSII is a hot spot for insulin pump hypoglycemic therapy.

[0005] The closed-loop insulin pump device includes a blood glucose sensor buried under the skin and an insulin pump implanted in the abdomen. The blood glucose sensor can measure the patient's blood glucose concentration and send the information to the controller, which calculates the insulin release rate based on the difference between the blood glucose baseline concentration and the feedback blood glucose concentration. The insulin release rate is sent as a control signal to the insulin pump, and the insulin pump releases insulin through a catheter placed under the patient's skin. The schematic diagram of the closed-loop control system for treating diabetes using a closed-loop insulin pump can be used Figure 1 express.

[0006] right Figure 1 For the closed-loop system shown in the figure, how to design a controller to adjust the patient's blood sugar to the baseline value is the core issue. Although there are a lot of studies on the establishment of insulin blood sugar regulation models, controllers designed using theoretical mathematical models and nominal parameters are difficult to use in actual treatment. The physical signs of each patient are different, the corresponding model parameters are different, and the external interference to the model is also different. It is important to note that slight changes in some parameters may have a significant impact on the performance of the closed-loop system and may even lead to the death of the patient. Therefore, how to design a controller that is highly robust to various parameter changes and external interference is a very important issue.

[0007] At present, blood glucose closed-loop controllers include PID controllers, H∞ controllers, neural network controllers, fuzzy controllers, model predictive controllers, and active disturbance rejection control (ADRC). According to the current literature survey, in 2021, the paper "Design of Active Disturbance Rejection Controller for Diabetes Blood Glucose Regulation" first introduced the ADRC method to solve the insulin-blood glucose regulation problem of diabetes. The paper used singular perturbation theory to strictly analyze the boundedness and ultimate boundedness of the closed-loop system based on ADRC. The simulation results show that for multiple patients with different vital signs and parameters, the designed active disturbance rejection controller can smoothly adjust the patient's blood glucose to the baseline value.

[0008] However, the ADRC method proposed in the article "Design of Diabetes Blood Glucose Regulation Auto-Disturbance Rejection Controller" has the following problems: the normal blood glucose concentration range of the human body is not considered in the controller design process, but the blood glucose reference value is set to the lower limit of the human blood glucose range of 70mg / dl, and there is a risk of hypoglycemia in the control process. How to consider the normal blood glucose concentration range of the human body in the controller design process and design a reasonable robust controller is a very necessary and meaningful thing. Summary of the invention

[0009] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a robust output regulator and method based on the normal blood sugar concentration range of the human body.

[0010] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: a robust output regulation method based on the normal blood glucose concentration range of the human body, comprising: a tracking differentiator, a regulator, an insulin pump, an expanded state observer and a sensor for blood glucose regulation transition, the regulator is respectively connected to the tracking differentiator, the insulin pump and the expanded state observer, the insulin pump is used for injection, and the sensor is connected to the object injected by the insulin pump.

[0011] A robust output regulation method based on the normal blood glucose concentration range of the human body is designed with the following steps:

[0012] (1) Construct control variables;

[0013] (2) Design a tracking differentiator to arrange the transition process of blood glucose regulation;

[0014] (3) construct tracking error and transformation error;

[0015] (4) Construct sliding mode variables and derived variables h;

[0016] (5) Design an extended state observer;

[0017] (6) Design the control input u(t) using the estimated value of the extended state observer.

[0018] As a further description of the above technical solution: the step of constructing the control variable includes defining the adjustment target r1 as the middle value of the normal blood glucose concentration range of the human body, 90 mg / dl, and the variable ρ as the adjustment range of r1, 20 mg / dl.

[0019] As a further description of the above technical solution: the design tracking differentiator arranges the transition process of blood sugar regulation as follows:

[0020]

[0021] Among them, the polynomial is a Hurwitz polynomial. Definition And x * (0) = x(0).

[0022] As a further description of the above technical solution: In the steps of constructing the tracking error and transforming the error, the tracking error is defined as It is required that for all t≥0, |e(t)|<ρ; construct the transformation error z(t) to satisfy

[0023] e(t)=ρS(z(t)),e(0)=z(0)=0,

[0024] in, e(·) is an exponential function,

[0025] The transformation error is

[0026] As a further description of the above technical solution: In the step of constructing the sliding mode variable and the derived variable h,

[0027]

[0028] in,

[0029]

[0030] g = p3G.

[0031] As a further description of the above technical solution: In the step of designing an extended state observer, define x = [x1, x2, x3] T ,x1=G, The extended state observer is as follows

[0032]

[0033] in, and are the estimated values ​​of the system states x1(t), x2(t), x3(t) and the system uncertainty σ(t), is a small positive number, g0 is the nominal value of the control input gain g, and the parameter α i , i=1,2,3,4 satisfies the polynomial s 4 +α1s 3 +α2s 2 +α3s+α4 is Hurwitz's.

[0034] As a further description of the above technical solution: In the step of designing the control input u(t) using the estimated value of the extended state observer,

[0035]

[0036] in,

[0037]

[0038] and represents the variables h(·) and s estimated using the extended state observer, the saturation function

[0039] Defined as

[0040]

[0041] At the same time, the saturation boundary M is used to obtain a reasonable value through repeated trial and error of simulation and experiments.

[0042] Due to the adoption of the above technical scheme, the beneficial effect achieved by the present invention is: blood sugar is adjusted to the normal blood sugar concentration range of the human body, specifically, the patient's blood sugar can be effectively adjusted to the normal blood sugar range of 70-110 mg / dl. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 The invention proposes a robust output regulator based on the normal blood glucose concentration range of the human body. The working principle of the closed-loop insulin pump device in the background art;

[0044] Figure 2 It is a schematic diagram of a robust output regulator based on a normal blood sugar concentration range of a human body proposed by the present invention;

[0045] Figure 3 It is a schematic diagram of parameter selection of a robust output regulator based on a normal blood glucose concentration range of a human body proposed by the present invention;

[0046] Figure 4 These are the model parameters corresponding to five diabetic patients of a robust output regulator based on the normal blood glucose concentration range of the human body proposed by the present invention;

[0047] Figure 5 It is the parameter value of D(t) in the simulation of a robust output regulator based on the normal blood glucose concentration range of the human body proposed by the present invention;

[0048] Figure 6 The present invention proposes a robust output regulator based on the normal blood glucose concentration range of the human body, and the simulation results of the blood glucose concentration of five diabetic patients within four days are shown. DETAILED DESCRIPTION

[0049] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] The present invention provides the following embodiments to describe the present invention in detail:

[0051] Example

[0052] Reference Figure 2-6The present invention patent proposes a new robust output regulator for regulating blood sugar by an insulin pump, which can regulate blood sugar to the normal blood sugar concentration range of the human body.

[0053] Considering the Bergman Minimum Model of the Human Body's Insulin-Glucose Regulation System

[0054]

[0055] Among them, t represents time, + represents the positive reflection of glucose intake, and the physical meanings of the variables and constants in the equation are as follows:

[0056] G(t) Blood glucose concentration (mg / dl); X(t) Insulin concentration in the distal compartment (l / min); I(t) Insulin concentration in plasma (μU / ml); <![CDATA[G b ]]> Blood glucose baseline value (mg / dl); <![CDATA[I b ]]> Insulin baseline value (μU / ml); <![CDATA[p1]]> Glucose uptake rate in muscle and liver, normal value independent of insulin (l / min); <![CDATA[p2]]> The rate of decline of the glucose absorption capacity of the tissue (l / min); <![CDATA[p3]]> <![CDATA[Growth rate of the glucose absorption capacity of the tissue ((μU / ml)min -2 );]]> n First-order decay rate of insulin in blood (l / min); h Insulin release threshold (mg / dl); γ <![CDATA[When the blood glucose concentration is greater than the threshold h, the release rate of insulin ((μU / ml)min -2 (mg / dl) -1 );]]> .

[0057] y(t) represents the measurable output, and the term D(t) represents the rate at which blood absorbs glucose from the intestine after a meal (mg / dl / min). Since the normal insulin blood sugar regulation system cannot work properly in diabetic patients, this term is considered as a disturbance to the system. The control input u(t) represents the injection rate of insulin, the purpose of which is to achieve normal blood sugar regulation in diabetic patients. The third equation of the Bergman minimum model described by equation (1) is γ[G(t)-h] + The t term represents the insulin release mechanism in a normal human body to regulate blood sugar, but in diabetics, this term can be considered non-existent (γ≈0). In addition, for diabetics, the value of p1 decreases significantly and can be approximated to zero (p1≈0). It should be noted that all parameter values ​​in the Bergman model will vary with different patients and are not constant values. The nominal values ​​given are the measurement results of people with standard weight.

[0058] The robust output regulator proposed in this invention is designed to control the input u(t) so that the blood glucose concentration is stabilized at 70-110 mg / dl. The design steps include the following six steps:

[0059] 1. Construct control variables. Define the adjustment target r1 as the middle value of the normal blood glucose concentration range of the human body, 90 mg / dl, and the variable ρ as the adjustment range of r1, 20 mg / dl.

[0060] 2. Design a tracking differentiator to arrange the transition process of blood sugar regulation.

[0061]

[0062] in, is stable. Definition And x * (0) = x(0).

[0063] 3. Construct tracking error and transformation error. Tracking error is defined as It is required that for all t≥0, |e(t)|<ρ. Construct the transformation error z(t) to satisfy

[0064] e(t)=ρS(z(t)),e(0)=z(0)=0, (3)

[0066] in, e (·) is an exponential function. Further, we can get the transformation error

[0067]

[0068] 4. Construct sliding mode variables and derived variables h. As follows

[0069]

[0070] in,

[0071]

[0072] g=p3G.

[0073] 5. Design an extended state observer.

[0074] Define x = [x1, x2, x3] T ,x1=G, The designed extended state observer is as follows

[0075]

[0076] in, and are the estimated values ​​of the system states x1(t), x2(t), x3(t) and the system uncertainty σ(t), is a small positive number, g0 is the nominal value of the control input gain g. i , i=1,2,3,4 satisfies the polynomial s 4 +α1s 3 +α2s 2 +α3s+α4 is Hurwitz's.

[0077] 6. Design the control input u(t) using the estimated value of the extended state observer.

[0078]

[0079] in,

[0080] and represents the variables h(·) and s using the estimated values ​​of the extended state observer. Saturation function The purpose of introduction is to protect the system from the peak phenomenon of the transition process, which is defined as

[0081]

[0082] At the same time, the saturation boundary M can be reasonably determined through repeated trial and error of simulation and experiment.

[0083] In the above embodiment, the feasibility of the robust output regulator proposed by the present invention is verified by simulation examples. The parameters of the robust output regulator are selected as follows: Figure 3 As shown, the model parameters corresponding to the five diabetic patients are as follows Figure 4 shown.

[0084] In the simulation, D(t) is expressed as follows

[0085]

[0086] Where B>0, t meal Indicates meal time. The values ​​of various parameters in the simulation are as follows Figure 5 shown.

[0087] The above parameters are used to simulate the insulin-blood sugar closed-loop regulation system. The results are as follows: Figure 6 As shown. Figure 6 From the simulation results, it can be seen that although the initial blood glucose concentrations of the five patients all exceeded 110 mg / dl, under the action of the robust output regulator proposed in the patent of this invention, the blood glucose concentrations can be adjusted to the normal range within 136-150 minutes, and in the next 4 days, even if the five patients have meals, the blood glucose concentrations will not deviate from the safe range of human blood glucose.

[0088] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, changes, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A robust output regulation method based on the normal blood glucose concentration range of the human body, characterized in that: include: A tracking differentiator, a regulator, an insulin pump, an expanded state observer and a sensor are used for blood sugar regulation transition. The regulator is connected to the tracking differentiator, the insulin pump and the expanded state observer respectively. The insulin pump is used for injection, and the sensor is connected to the object injected by the insulin pump.

2. The robust output regulation method based on the normal blood glucose concentration range of the human body according to claim 1, characterized in that: The design steps are as follows: (1) Construct control variables; (2) Design a tracking differentiator to arrange the transition process of blood glucose regulation; (3) construct tracking error and transformation error; (4) Construct sliding mode variables and derived variables h; (5) Design an extended state observer; (6) Design the control input u(t) using the estimated value of the extended state observer.

3. The robust output regulation method based on the normal blood glucose concentration range of the human body according to claim 2, characterized in that: The step of constructing the control variable includes defining the adjustment target r1 as the middle value of the normal blood glucose concentration range of the human body, 90 mg / dl, and the variable ρ as the adjustment range of r1, 20 mg / dl.

4. The robust output regulation method based on the normal blood glucose concentration range of the human body according to claim 2, characterized in that: The design tracking differentiator arranges the transition process of blood glucose regulation as follows: Among them, the polynomial is the Hurwitz polynomial, defined as And x * (0) = x(0).

5. The robust output regulation method based on the normal blood glucose concentration range of the human body according to claim 2, characterized in that: In the steps of constructing the tracking error and transforming the error, the tracking error is defined as It is required that for all t≥0, |e(t)|<ρ. Construct the transformation error z(t), satisfying e(t)=ρS(z(t)), e(0)=z(0)=0, where e (·) is an exponential function, the transformation error is 6. The robust output regulation method based on the normal blood glucose concentration range of the human body according to claim 2, characterized in that: In the step of constructing sliding mode variables and derived variables h, in, g = p3G.

7. The robust output regulation method based on the normal blood glucose concentration range of the human body according to claim 2, characterized in that: In the step of designing the extended state observer, define The extended state observer is as follows in, and are the estimated values ​​of the system states x1(t), x2(t), x3(t) and the system uncertainty σ(t), is a small positive number, g0 is the nominal value of the control input gain g, and the parameter α i , i=1,2,3,4 satisfies the polynomial s 4 +α1s 3 +α2s 2 +α3s+α4 is Hurwitz's.

8. The robust output regulation method based on the normal blood glucose concentration range of the human body according to claim 2, characterized in that: In the step of designing the control input u(t) using the estimated value of the extended state observer, in, and represents the variables h(·) and s estimated using the extended state observer, the saturation function Defined as At the same time, the saturation boundary M is used to obtain a reasonable value through repeated trial and error of simulation and experiment.

Images