A dynamic modeling and control method for patch clamp high-resistance sealing

Abstract

The present invention discloses a dynamic modeling and control method for high-resistance sealing of a patch clamp, which relates to the field of micromanipulation, and specifically includes the following steps: S1: performing force analysis on cells in the dynamic process of high-resistance sealing of the patch clamp, and estimating the resistance value variation curve over time; S2: performing experiments and recording the data of sealing resistance value variation over time under different holding pressures; S3: using an autoregressive model to perform model identification on the data of sealing resistance value variation over time under different holding pressures, and obtaining the sealing resistance value-holding pressure system transfer function; S4: in view of the problem that there are many undetermined interferences in the system, an anti-disturbance controller is designed to realize the anti-disturbance control of the holding pressure during the high-resistance sealing of the patch clamp. After the present invention is verified by experiments, the results show that the high-resistance sealing success rate in the method proposed by the present invention is improved by more than 35% compared with the method of fixing the holding pressure.

Description

Technical Field

[0001] The invention relates to the field of micromanipulation, in particular to a dynamic modeling and control method for high-resistance sealing of a patch clamp. Background Art

[0002] Patch clamp technology is the "gold standard" for current cell ion channel research. A key step is to absorb part of the cell membrane into the microtubule and form a GΩ-level resistance seal (high-resistance seal) between it and the inner wall of the microtubule, thereby shielding environmental electromagnetic noise and ensuring the accuracy of measuring the electrical signals of the cell membrane ion channel. High-resistance sealing is one of the key steps in patch clamp operation, and it is also a guarantee for the success rate and accuracy of cell signal detection. However, the formation of high-resistance sealing is a process of large nonlinear deformation of the cell membrane, and is accompanied by complex changes such as rearrangement of cell membrane ion channels and protein denaturation, which can easily cause large disturbances to the sealing resistance and affect the final success rate of high-resistance sealing.

[0003] During the current patch clamp sealing operation, the operator generally relies on experience to set a fixed negative pressure to suck in the cell membrane. During the sealing process, the suction pressure is generally not adjusted according to the change of the sealing resistance, but passively waits for the sealing resistance to reach the GΩ level; this is equivalent to open-loop control of the process, so the adhesion effect between the cell membrane and the inner wall of the microtubule is very random, and it is very easy to have insufficient final sealing resistance. Therefore, the development of dynamic modeling and control methods for the high-resistance sealing process of the patch clamp is of great significance to improve the success rate of high-resistance sealing, and then improve the success rate of signal detection of the patch clamp. Summary of the invention

[0004] In order to overcome the current control method that the holding pressure in the patch clamp high-resistance sealing process is determined by artificial experience and has no basis for the change of the sealing resistance value, and the problem that the cell membrane and the inner wall of the microtubule have a large randomness in the attachment effect, the present invention proposes a dynamic modeling and control method for patch clamp high-resistance sealing, which first analyzes the force of cells in the patch clamp high-resistance sealing process to obtain an estimated resistance change curve over time, and obtains the real resistance change curve over time through the patch clamp high-resistance sealing process experiment, and then uses an autoregressive model to identify the sealing resistance change curve over time to obtain a model discrete transfer function, and obtains a model transfer function after continuousization, and finally, in view of the problem that there are many undetermined interferences in the system, the holding pressure is adjusted by an auto-disturbance rejection controller, and the disturbance is estimated by an extended state observer for compensation, so as to realize the adaptive control of the holding pressure in the patch clamp high-resistance sealing process.

[0005] In order to achieve the above-mentioned invention object, the present invention adopts the following technical scheme:

[0006] The present invention provides a patch clamp high-resistance sealing dynamic modeling and control method, comprising the following steps:

[0007] S1: Perform force analysis on the dynamic process of patch clamp high-resistance sealing and estimate the curve of sealing resistance changing with time;

[0008] S2: During the experiment, the holding pressure is continuously changed, and the data of the sealing resistance and holding pressure changing with time are recorded to verify whether the measured curve is consistent with the estimated curve. If not, the experiment needs to be repeated.

[0009] S3: Use the autoregressive model to identify the curve of sealing resistance changing with time, and obtain the model transfer function after continuousization;

[0010] S4: In order to solve the problem of many uncertain interferences in the system, the self-anti-disturbance controller is used to adjust the holding pressure in real time to realize automatic control of the holding pressure during the patch clamp high-resistance sealing process.

[0011] In step S1, the formation of high-resistance seal is mainly affected by five forces: cell membrane stretching force caused by suction pressure, stress caused by cytoskeleton structure on sucked cell membrane (hereinafter referred to as cytoskeleton stress), electroosmotic force caused by voltage (if additional voltage is applied during sealing), adhesion force between cell membrane and inner wall of glass microtube (hereinafter referred to as cell membrane-tube wall adhesion force) and cell membrane viscosity. e It is the main driving force for pulling the cell membrane into the microtubule and continuously extending it during the entire high-resistance sealing process. c It can be divided into the tangential force F along the diaphragm ct and the component force F along the normal direction of the diaphragm cn The tangential force F ct The normal force F cn and cell membrane-tube wall adhesion force F a Together, they promote the adhesion of the membrane to the tube wall, increasing the adhesion area, while also hindering the cell membrane from entering the microtubule and extending by generating friction with the tube wall. v It hinders the dynamic deformation of the diaphragm.

[0012] According to the above mechanical analysis, the dynamic process of sealing can be divided into three parts: the initial stage, the middle stage and the final stage. In the initial stage of sealing, due to the small deformation of the cell membrane, the cell membrane and cytoskeleton entering the microtubule are small, and the cytoskeleton force and the cell membrane-tube wall adhesion force are very small. At this time, the suction pressure should be increased to allow the cell membrane to quickly enter the microtubule and form a preliminary adhesion with the inner wall, so that the resistance value increases rapidly, creating conditions for subsequent sealing. In the middle stage of sealing, as the deformation of the membrane increases in the microtubule, the suction pressure should be adjusted to achieve a certain balance between the cell membrane stretching force and the cytoskeleton force and the cell membrane-tube wall adhesion force, so that the cell membrane crawling, elongation and expansion of the attachment area proceed smoothly, which is manifested as a steady increase in the sealing resistance; when the sealing resistance reaches the GΩ level, it enters the final stage of sealing. At this time, the elongation of the membrane has reached its limit. In order to prevent the cell membrane from rupturing due to excessive stretching, the suction pressure and the stretching force generated by it should be slowly reduced at this stage. Only the normal force of the cytoskeleton and the adhesion force of the tube wall can further increase the adhesion between the membrane and the inner wall of the microtube and the sealing resistance, and finally stabilize to the ideal value.

[0013] Therefore, according to the above analysis, the sealing resistance in high-resistance sealing should rise smoothly and stabilize to the ideal value according to an "S"-shaped curve that first accelerates, then maintains a constant speed, and finally decelerates.

[0014] Furthermore, in step S2, the data of the change of sealing resistance over time under different holding pressures during the patch clamp high-resistance sealing process are experimentally recorded, and the actual data is verified based on the estimated curve. If the trend is different from the estimated curve, the test needs to be repeated; if the trend is the same as the estimated curve, the range of change of the sealing resistance during the patch clamp high-resistance sealing process is preliminarily obtained.

[0015] Furthermore, in step S3, an autoregressive model is used to identify the data of the change of sealing resistance over time when the holding pressure is changed during the patch clamp high-resistance sealing process by recursive least squares method (with holding pressure as input and sealing resistance as output), so as to obtain the transfer function.

[0016] Furthermore, in step S4, an ADRC self-disturbance rejection controller is designed according to the transfer function of sealing resistance-holding pressure determined in step S3. During the attachment process, the holding pressure is adjusted according to the real-time detection result of the sealing resistance so that the sealing resistance tracks the planned curve and stabilizes at an ideal value at the GΩ level, thereby realizing automatic control of the holding pressure during the attachment process.

[0017] Compared with the prior art, one or more of the above technical solutions have the following beneficial effects:

[0018] 1. The dynamic modeling and control method for high-resistance sealing of a patch clamp described in the present invention can realize real-time detection of the sealing resistance during the attachment process of the microtube electrode and the cell, and realize real-time control of the holding pressure according to the sealing resistance. Compared with the traditional method of giving the holding pressure based on artificial experience, it can complete the high-resistance sealing of the cell more accurately, and the success rate of high-resistance sealing is increased by more than 35%.

[0019] 2. The ADRC self-disturbance rejection controller designed in the present invention has a certain anti-interference effect on the disturbance term in the sealing resistance signal, and can reduce the influence of noise on the sealing process.

[0020] 3. The high-resistance sealing control method of the present invention allows users to realize automated cell high-resistance sealing operations simply by clicking the mouse on the display screen, which significantly reduces the technical threshold for high-resistance sealing and patch clamp operations and helps promote the popularization and application of this technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0022] Figure 1 It is the core content advancement diagram of the present invention;

[0023] Figure 2 is a cell force analysis diagram in an embodiment of the present invention;

[0024] Figure 3 It is a schematic diagram of positioning a microtube electrode on an operation interface and fine-tuning the position in an embodiment of the present invention;

[0025] Figure 4 is a schematic diagram of a curve of sealing resistance value varying with time measured in an experiment in an embodiment of the present invention;

[0026] Figure 5 is a flow chart of data processing before system identification and modeling of measured data in an embodiment of the present invention;

[0027] Figure 6 is a structural diagram of an active disturbance rejection control ADRC in an embodiment of the present invention;

[0028] Figure 7 It is a schematic diagram of realizing GΩ high-resistance sealing in the operation interface in an embodiment of the present invention. DETAILED DESCRIPTION

[0029] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0030] Embodiment 1:

[0031] like Figures 1 to 7 As shown, a dynamic modeling and control method for patch clamp high-resistance sealing specifically includes the following steps:

[0032] S1: Perform force analysis on the dynamic process of patch clamp high resistance sealing and estimate the curve of sealing resistance changing with time; the force analysis can be focused on Figure 2 , specifically:

[0033] The formation process of high-resistance sealing is mainly affected by five forces: cell membrane stretching force caused by suction pressure, stress caused by cytoskeleton structure on sucked cell membrane (hereinafter referred to as cytoskeleton stress), electroosmotic force caused by voltage (if additional voltage is applied during sealing), adhesion force between cell membrane and inner wall of glass microtube (hereinafter referred to as cell membrane-tube wall adhesion force) and cell membrane viscosity. e It is the main driving force for pulling the cell membrane into the microtubule and continuously extending it during the entire high-resistance sealing process. c It can be divided into the tangential force F along the diaphragm ct and the component force F along the normal direction of the diaphragm cn The tangential force F ct The normal force F cn and cell membrane-tube wall adhesion force F a Together, they promote the adhesion of the membrane to the tube wall, increasing the adhesion area, while also hindering the cell membrane from entering the microtubule and extending by generating friction with the tube wall. v It hinders the dynamic deformation of the diaphragm.

[0034] According to the above mechanical analysis, the dynamic process of sealing can be divided into three parts: the initial stage, the middle stage and the final stage. In the initial stage of sealing, due to the small deformation of the cell membrane, the cell membrane and cytoskeleton entering the microtubule are small, and the cytoskeleton force and the cell membrane-tube wall adhesion force are very small. At this time, the suction pressure should be increased to allow the cell membrane to quickly enter the microtubule and form a preliminary adhesion with the inner wall, so that the resistance value increases rapidly, creating conditions for subsequent sealing. In the middle stage of sealing, as the deformation of the membrane increases in the microtubule, the suction pressure should be adjusted to achieve a certain balance between the cell membrane stretching force and the cytoskeleton force and the cell membrane-tube wall adhesion force, so that the cell membrane crawling, elongation and expansion of the attachment area proceed smoothly, which is manifested as a steady increase in the sealing resistance; when the sealing resistance reaches the GΩ level, it enters the final stage of sealing. At this time, the elongation of the membrane has reached its limit. In order to prevent the cell membrane from rupturing due to excessive stretching, the suction pressure and the stretching force generated by it should be slowly reduced at this stage. Only the normal force of the cytoskeleton and the adhesion force of the tube wall can further increase the adhesion between the membrane and the inner wall of the microtube and the sealing resistance, and finally stabilize to the ideal value.

[0035] Therefore, according to the above analysis, the sealing resistance in high-resistance sealing should rise smoothly and stabilize to the ideal value according to an "S"-shaped curve that first accelerates, then maintains a constant speed, and finally decelerates.

[0036] S2: During the experiment, the holding pressure is continuously changed, and the data of the sealing resistance and holding pressure changing with time are recorded to verify whether the measured curve is consistent with the estimated curve. If not, the experiment needs to be repeated.

[0037] Specifically, in this Example 1, the cells operated were mouse brain slice cells, and the preparation process was as follows: healthy male SD mice were anesthetized with 1% sodium pentobarbital intraperitoneally, and then their heads were quickly cut off to remove the brains. 2 and 5% CO 2 ) in the slicing solution for a dozen seconds and then take it out. Use Precisionary Instruments VF-700 vibrating slicer to prepare brain slice specimens. Trim off the unnecessary parts along the coronal plane, stick it to the tissue holder of the vibrating slicer with 502 glue, fix it with agar, then quickly place it in the slicing slot and pour in 4°C slicing solution to immerse the brain tissue block. Adjust the distance between the blade and the tissue and start slicing. The slicer vibration frequency is 7, the forward speed is 3, and the thickness of the brain slice is 200μm. Every time a brain slice is cut out along the coronal plane, it is quickly transferred to the (33°C) incubation solution. The slicing solution and the incubation solution are continuously ventilated with mixed gas (95% O 2 and 5% CO 2 ). The cut brain slices can be observed and recorded after incubation for 1 hour.

[0038] In step S2 of this embodiment 1, after the mouse brain slice is prepared, the borosilicate glass electrode is stretched using a horizontal electrode puller (P-97, Sutter Instrument Co, Novato, CA). Turn on the power switch on the left side of the puller, enter the program number to select a set pulling program, and the program parameters used in the experiment are (P=500, HEAT=523, PULL=0, VEL=20, TIME=250). First, place the core microelectrode glass tube (GBF150-86-10HP) in the groove of the glass clamp on one side and fix it with a knob, then push the glass tube through the heating chamber and fix it with the glass tube clamp on the other side. Press the PULL button to pull the electrode. After hearing four "clicks", open the cover and carefully remove the drawn glass microtube electrode. Take a brain slice and place it in the bath and press it with the cover net. Before sealing, find the desired cells in the brain slice under a low-power (4x) objective lens, and then observe the cell body morphology and axonal course under a high-power (40x) water-immersion objective lens through image acquisition software. After selecting the appropriate target cells (adherent cells with high edge height, preferably elliptical or nearly triangular), mark the location on the computer screen and adjust the position of the micropipette electrode through the microelectrode manipulator, such as Figure 3 As shown, the electrode is placed in water with positive pressure and the water resistance is recorded; under a high-power water-immersion objective, the electrode is gradually guided to the cell through image sampling. Under visual conditions, combined with the electrode impedance graph provided by the recording software, the positive pressure is released in a timely manner, and negative pressure is lightly applied. By clamping the membrane potential level, a GΩ-level high-resistance seal is obtained. After the membrane is successfully broken, a whole-cell recording method can be formed. The membrane potential is clamped at a fixed level with a voltage clamp, and key data such as the cell holding pressure and corresponding resistance are recorded. The "S"-shaped sealing resistance change curve is optimized to make the final attachment area range and sealing resistance as large as possible.

[0039] Furthermore, the curve of the sealing resistance value obtained from the experiment changing with time is as follows: Figure 4 As shown, it is an "S"-shaped curve that first accelerates, then maintains a constant speed, and finally decelerates.

[0040] Furthermore, if Figure 5 As shown, in this Example 1, the data of the measured changes of the sealing resistance and the holding pressure over time during the patch clamp high-resistance sealing process are modeled as follows:

[0041] First, the measured data of the sealing resistance and holding pressure changing with time during the patch clamp high-resistance sealing process are tested to see whether the measured data curve is consistent with the results of the force analysis. If the test fails, the data needs to be re-measured; if the test passes, the recursive least squares method is used to select the autoregressive model to identify the model of the sealing resistance changing with time under different holding pressures, and the transfer function of the sealing resistance-holding pressure system model is obtained. Since the measured data is a time series, the obtained transfer function is a discrete system transfer function, which needs to be discretely inversely transformed to obtain the system transfer function.

[0042] The system discrete-time random linear model can be expressed as:

[0043] A(z -1 )y(k)=z -d B(z -1 )u(k)+v(k) (1)

[0044] Among them, y(k) is the output variable, u(k) is the input variable, v(k) is white noise, d is the time delay of the controlled object, A(z -1 ) and B(z -1 ) is the system parameter matrix, which can be expressed as:

[0045]

[0046]

[0047] Among them, n a is the order of the A(z) polynomial, n b is the order of the B(z) polynomial.

[0048] In order to form an identification model, we define the parameter vector θ and the data vector For later calculation, the corresponding identification model can be obtained:

[0049] θ=[a 1 ,a 1 ,…,a na ,b 0 ,b 1 ,…,b nb ] T (4)

[0050]

[0051] The controlled object model shown in formula (1) can be expressed as:

[0052]

[0053] Denote by N(N≥n a +nb +1) observations to estimate the unknown parameter θ is The criterion for minimizing the deviation between actual observations and model calculations can be expressed as:

[0054]

[0055] For the i-th observation, the deviation between the actual observation value and the value calculated by the estimated model, that is, the residual can be expressed as:

[0056]

[0057] Introduction notation:

[0058]

[0059] The controlled object model shown in formula (6) can be expressed as:

[0060] Y N =H N θ+V N (10)

[0061] The residual shown in formula (8) can be expressed as:

[0062]

[0063] The least squares objective function can be expressed as:

[0064]

[0065] The two ends of equation (12) are Find the partial derivative and we get:

[0066]

[0067] Let the objective function J in formula (13) be The partial derivative is 0, so:

[0068]

[0069] if is a non-singular matrix, then the least squares estimate of the parameter θ is for:

[0070]

[0071] Formula (13) Find the partial derivative and we get:

[0072]

[0073] It can be shown that when θ is the least squares estimate When , the objective function J shown in formula (12) takes a minimum value.

[0074] The least squares estimation method shown in formula (15) is a one-time algorithm that occupies a large amount of memory during calculation. The recursive least squares method can solve this problem.

[0075] Assume that up to time (k-1), the least squares estimate of the unknown parameters at time k for all observed data is for:

[0076]

[0077] When a new set of observation data is collected at the kth moment, the least squares estimate of the unknown parameters at the kth moment is obtained. for:

[0078]

[0079] Referring to formula (9), we have:

[0080]

[0081] Therefore, formula (17) can be expressed as:

[0082]

[0083] Import:

[0084] P(k)=(H k T H k ) -1 (twenty one)

[0085] From formula (19) and formula (21), we can get:

[0086]

[0087] Import:

[0088]

[0089] Since the matrix inversion formula satisfies:

[0090] (A+BCD) -1 =A -1 -A -1 B(C -1 +DA -1 B) -1 DA -1 (twenty four)

[0091] From equation (22), equation (23) and equation (24), we can get:

[0092]

[0093] Import:

[0094]

[0095] Therefore, formula (25) can be expressed as:

[0096]

[0097] Therefore, formula (20) can be expressed as:

[0098]

[0099] Formula (17) can also be expressed as:

[0100]

[0101] Then formula (28) can be expressed as:

[0102]

[0103] Formula (26) can be expressed as:

[0104]

[0105] From equation (30) and equation (31), we can get:

[0106]

[0107] In summary, the formula for estimating the parameter vector θ by recursive least squares method can be expressed as:

[0108]

[0109] The specific recursive steps are:

[0110] (1) Set the initial value and P(0);

[0111] (2) Constructing data vectors of sealing resistance under different holding pressures

[0112] (3) Apply u(k) to the cell to obtain the kth sealing resistance y(k);

[0113] (4) According to formula (33), K(k), and P(k);

[0114] (5) Recursively go to item k+1, return to step 2, construct a new data vector, and recalculate the parameters.

[0115] Furthermore, if Figure 6 As shown, in this example 1, based on the dynamic modeling results of the patch clamp high-resistance seal, an ADRC self-disturbance rejection controller is designed to detect the sealing resistance in real time and automatically control the holding pressure. The specific design is as follows:

[0116] First, the cell membrane-microtube electrode system in high-resistance sealing contains obvious elasticity, inertia and damping modules, so the cell membrane-microtube electrode in the sealing process should have the characteristics of a second-order or higher system. If a second-order system is used to design ADRC, the system state expression is as follows:

[0117]

[0118] Among them, x 1 is the sealing resistance, x 2 is the change rate of sealing resistance, x 3 is the expanded state (the number of state variables is equal to the system order, the second-order system x 3 becomes an expanded state), u is the ideal holding pressure, and f is the disturbance of the system.

[0119] Let ξ 1 =x, ξ 3 =f,ξ=[ξ 1 ξ 2 ξ 3 ], assuming f is differentiable, the system space can be defined as:

[0120]

[0121] in:

[0122] The extended state observer (ESO) of the system is:

[0123]

[0124] Where l = [l 1 l 2 l 3 ] is the gain matrix of the observer. The selection of gain should make the characteristic polynomial satisfy the Hurwitz criterion, that is:

[0125] λ 0 (s)=s 3 +l 1 s 2 +l 2 s+l 3 =(s+ω 0 ) 3 (37)

[0126] Among them, ω 0is the bandwidth of the system observer, and L is:

[0127] L=[3ω 0 3ω 0 3 ω 0 3 ] T (38)

[0128] Linear Error Feedback Control Law (NLSEF):

[0129]

[0130] When the output of the observer can track ξ 1 and 2 When 3 It can be accurately estimated, and the object can be simplified to a unity gain integrator, making the estimation error zero, that is:

[0131]

[0132] At this time u=u 0 +ξ 3 / b 0 , thus the control law of ADRC is determined as:

[0133]

[0134] Here r is the trajectory of setting the sealing resistance value, k 1 and k 2 Make s 2 +k 2 s+k 1 The controller gain k whose root satisfies the Hurwitz criterion 1 =ω c 2 , k 2 =2ω c , where ω c is the controller bandwidth.

[0135] Finally, the high-resistance sealing of the pyramidal neurons in the visual cortex of mice was performed based on the method of the present invention and compared with the high-resistance sealing results of the method with fixed suction and holding pressure. The fixed suction and holding pressure was selected to be 0.5psi based on the operator's experience. The results of the sealing of 20 neurons by the method of the present invention reached the GΩ level in 15 cases, with a success rate of 75%, while the success rate of the fixed air pressure method was only 55% (11 / 20). In comparison, the success rate of the method of the present invention was 1.36 times that of the fixed air pressure method, with an improvement percentage of more than 35%.

[0136] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A dynamic modeling and control method for patch clamp high-resistance sealing, characterized in that: The following steps are involved: S1: Perform force analysis on the dynamic process of patch clamp high-resistance sealing and estimate the curve of sealing resistance changing with time; S2: During the experiment, the holding pressure is continuously changed, and the data of the sealing resistance and holding pressure changing with time are recorded to verify whether the measured curve is consistent with the estimated curve. If not, the experiment needs to be repeated. S3: Use the autoregressive model to identify the curve of sealing resistance changing with time, and obtain the model transfer function after continuousization; S4: In view of the problem that there are many undetermined interferences in the system, the holding pressure is adjusted by using the anti-disturbance controller to realize the automatic control of the holding pressure in the patch clamp high-resistance sealing process; Among them, the data of the sealing resistance and holding pressure changing with time during the patch clamp high-resistance sealing process are tested to observe whether the measured data curve conforms to the results obtained by the force analysis. If the test fails, it is necessary to re-measure the data; if the test passes, the recursive least squares method is used, and the autoregressive model is selected to identify the model of the sealing resistance changing with time under different holding pressures, and the transfer function of the sealing resistance-holding pressure system model is obtained. Since the measured data is a time series, the obtained transfer function is a discrete system transfer function, which needs to be discretely inversely transformed to obtain the system transfer function; The system discrete-time random linear model can be expressed as: ; in, is the output variable, is the input variable, is white noise, is the time delay of the controlled object, and is the system parameter matrix.

2. The patch clamp high-resistance sealing dynamic modeling and control method according to claim 1, characterized in that: In the step S1, the estimated curve obtained by the force analysis is used to verify the sealing resistance variation curve obtained by the experiment.

3. The dynamic modeling and control method for patch clamp high-resistance sealing according to claim 1, characterized in that: In the step S2, the obtained data of the sealing resistance and the holding pressure changing with time includes a time series, a sealing resistance value and a holding pressure value.

4. The patch clamp high-resistance sealing dynamic modeling and control method according to claim 1, characterized in that: In step S3, the system with the holding pressure as input and the sealing resistance as output is identified by the recursive least square method based on the autoregressive model.

5. The patch clamp high-resistance sealing dynamic modeling and control method according to claim 4, characterized in that: In step S3, the model of the sealing resistance variation curve after identification is a discrete transfer function, and after continuous processing is performed on it, the transfer function of the continuous system is obtained.

6. The patch clamp high-resistance sealing dynamic modeling and control method according to claim 1, characterized in that: In the step S4, when the holding pressure is adjusted by using the active disturbance rejection controller, the expansion state observer is used to estimate the sealing resistance disturbance caused by the uncertainty in the step S4.

7. The patch clamp high-resistance sealing dynamic modeling and control method according to claim 6, characterized in that: In the step S4, the electrode-cell membrane in the adsorption process is taken as the controlled object, and the sealing resistance is stabilized at the GΩ level by adjusting the adsorption pressure.

Images