Measurement impedance correction method and device based on circuit distribution parameter adaptive identification

Abstract

The invention relates to a measured impedance correction method and device based on circuit distribution parameter self-adaptive identification. The method comprises the following steps that two amplitudes Z1 and Z2 obtained by conducting impedance measurement on two precision resistors R1 and R2 with known resistance values under the set excitation frequency through a measurement circuit are obtained; based on the amplitude Z1, adopting a simplified circuit model to obtain the value of the distributed capacitance C0, and based on the amplitude Z2, adopting a global probability search algorithm to obtain the values of the distributed capacitance C1 and C2; constructing a measurement circuit model corresponding to the measurement circuit based on the distributed capacitors C0, C1 and C2; and correcting the directly obtained measurement impedance by using a global optimization algorithm according to the measurement circuit model. Compared with the prior art, the method has the advantages of effectively improving the measurement performance and the like.

Description

Technical Field

[0001] The invention belongs to electrical impedance tomography technology, relates to an impedance measurement method, and in particular to a measurement impedance correction method and device based on adaptive identification of circuit distributed parameters. Background Art

[0002] Electrical impedance tomography is a radiation-free, non-invasive imaging technology suitable for long-term dynamic monitoring. It is widely used in medical applications such as lung detection, brain imaging and breast cancer screening. By measuring the impedance of biological tissues at different frequencies, its electrical characteristics information, such as conductivity, can be obtained, which can be used for disease diagnosis and biological tissue identification. At excitation frequencies above 200kHz, due to the weakening of the capacitance effect of the cell membrane, the excitation current can pass through the cell membrane into the cell, and the measured electrical characteristics information contains more biological tissue characteristics. However, when the excitation frequency exceeds 200kHz, the parasitic parameters of the measurement circuit, such as the chip pin-to-ground capacitance, the multiplexer on-resistance, and the transmission line-to-ground capacitance, exist, and the measured impedance value contains parasitic parameters, resulting in an increase in the error of the measurement result. When the frequency remains unchanged, the larger the impedance, the greater the influence of the parasitic parameters on the measurement result. The accuracy and stability of the impedance directly affect the quality of imaging.

[0003] Traditional methods usually use fixed circuit model parameters, which may lead to measurement errors and imaging artifacts, and systems using fixed parameters are very sensitive to noise and model mismatch, which reduces the robustness and accuracy of the measurement. Therefore, it is necessary to update the circuit model parameters according to the current measurement state to improve the measurement accuracy. Patent document CN105976411A (application number: CN201610332979.1) discloses a method of using the ratio of the average value of the voltage amplitude measured multiple times on the electrode under test to the sum of the average values ​​of the voltage amplitude on all electrodes as the normalization correction parameter of the transmission characteristic of the corresponding electrode measurement module. This method simplifies the nonlinear measurement equivalent circuit transfer function into a linear relationship, without considering the parasitic parameters on the measurement channel, such as the chip pin-to-ground capacitance, switch on-resistance, etc., and cannot cope with the situation of changes in the measurement state, and the corrected value deviates from the theoretical value.

[0004] Therefore, in order to improve the measurement accuracy and stability of the measurement system, a circuit distributed parameter self-identification method that can be updated in real time is needed. Summary of the invention

[0005] The purpose of the present invention is to provide a measurement impedance correction method and device based on circuit distributed parameter adaptive identification, which can effectively improve measurement performance.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] A measurement impedance correction method based on adaptive identification of circuit distributed parameters comprises the following steps:

[0008] Obtain two amplitudes Z1 and Z2 obtained by measuring the impedance of two precision resistors R1 and R2 of known resistance values ​​at a set excitation frequency by a measuring circuit;

[0009] Based on the amplitude Z1, a simplified circuit model is used to obtain the value of the distributed capacitance C0, and based on the amplitude Z2, a global probability search algorithm is used to obtain the values ​​of the distributed capacitances C1 and C2;

[0010] Constructing a measurement circuit model corresponding to the measurement circuit based on the distributed capacitances C0, C1 and C2;

[0011] According to the measurement circuit model, the directly obtained measured impedance is corrected by using a global optimization algorithm.

[0012] Furthermore, the resistance of the precision resistor R1 is less than 50Ω.

[0013] Furthermore, the excitation frequency is set to 1 MHz.

[0014] Furthermore, the value of the distributed capacitance C0 is obtained based on the amplitude Z1 and the measured impedance amplitude obtained according to the channel measurement voltage difference and the sampled current.

[0015] Furthermore, the global probability search algorithm is a simulated annealing algorithm.

[0016] Furthermore, when the simulated annealing algorithm is used to estimate the values ​​of the distributed capacitances C1 and C2, the objective function expression of the design is:

[0017] minimize E(S)=||Z t_s (C1,C2)|-|Z2|| 2 Among them, Z t_s (C1, C2) is a simulation value of the circuit model calculated based on the solution S = (C1, C2), and represents the impedance value obtained by directly measuring the voltage difference and dividing the excitation current value of the measurement circuit.

[0018] Furthermore, the global optimization algorithm is a differential evolution algorithm.

[0019] Furthermore, the process of using the differential evolution algorithm for correction includes:

[0020] Initialize the algorithm parameters, randomly generate a population, split the corrected measured impedance into resistance and capacitance, and use the corrected resistance and capacitance as individuals in the population;

[0021] Calculate the fitness of each individual in the population;

[0022] The individuals in the population are subjected to mutation, crossover and selection operations, and individuals with better fitness are selected to enter the next generation. The iteration continues until the termination condition is met, and the individual with the best fitness in the current population is output as the optimal solution. The corrected measured impedance is obtained based on the optimal solution.

[0023] The present invention also provides a computer-readable storage medium, characterized in that it includes one or more programs for execution by one or more processors of an electronic device, and the one or more programs include instructions for executing the measurement impedance correction method based on adaptive identification of circuit distributed parameters as described above.

[0024] The present invention also provides an electronic device, comprising one or more processors, a memory and one or more programs stored in the memory, wherein the one or more programs include instructions for executing the measurement impedance correction method based on adaptive identification of circuit distributed parameters as described above.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] 1. The present invention constructs a correction model between the measured impedance and the impedance value to be measured regarding the excitation frequency, including capacitance, inductance and resistance. The correction coefficient has a nonlinear relationship with the excitation frequency, thereby constructing a more practical nonlinear measurement model and improving the recognition performance.

[0027] 2. The present invention can realize the adaptive identification of the circuit distributed parameters at any time when recalibration is required without disassembling the circuit, and can effectively cope with the situation of measurement state changes, with good adaptability. The present invention realizes the self-identification of the measurement circuit distributed parameters, can identify the parameters on the board during the measurement process, reduces the interference of human intervention and off-board factors, and improves the automation of the system.

[0028] 3. The method of the present invention can realize self-identification of the distributed parameters of the circuit in the frequency range of more than 200kHz to 1MHz, and has low computational complexity. It can quickly and accurately correct the measured impedance according to the circuit model, realize accurate measurement of impedance, and thus improve the accuracy of bioelectrical impedance data reconstruction. The present invention realizes the correction of measured impedance based on the circuit model of the identified distributed parameters, and increases the excitation frequency for accurate impedance measurement to 1MHz, which is suitable for a wider range of high-frequency application scenarios and is conducive to the accurate acquisition and analysis of the electrical characteristics of biological tissues above 200kHz.

[0029] 4. The present invention has a better correction effect on larger impedances, and there is no need to change the existing measurement channel circuit design, which is more conducive to integration into the measurement system. It has important application value and good application prospects in image reconstruction such as electrical impedance tomography. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 is a flow chart of a method according to an embodiment of the present invention;

[0031] Figure 2 Flow chart of a method for obtaining C1 and C2 using a global probability search algorithm according to Embodiment 1 of the present invention;

[0032] Figure 3 This is a flow chart of a method for calibrating and measuring impedance according to Embodiment 1 of the present invention;

[0033] Figure 4 An equivalent model of a measuring circuit including distributed circuit parameters established when measuring impedance using the method of the present invention;

[0034] Figure 5 A circuit equivalent model for calculating the distributed capacitance C0 using the method of the present invention;

[0035] Figure 6 The bar graphs are of the relative residuals between the simulated values ​​and the measured values ​​of the equivalent circuit model obtained by the method of the present invention at 500kHz, 800kHz and 1MHz, wherein (6a) is 500kHz, (6b) is 800kHz, and (6c) is 1MHz;

[0036] Figure 7 The relative error box plots of the measured impedance resistance parts and the theoretical values ​​before and after correction of Example 2 and Example 3 obtained by the method of the present invention, wherein (7a) is Example 2, and (7b) is Example 3;

[0037] Figure 8 The box plot of the capacitance value of Example 3 after correction obtained by the method of the present invention versus frequency is shown. DETAILED DESCRIPTION

[0038] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0039] Example 1

[0040] like Figure 1 As shown, this embodiment proposes a method for adaptively identifying circuit distributed parameters to improve measurement performance, including the following steps:

[0041] S1. Obtain two amplitudes Z1 and Z2 obtained by respectively performing impedance measurement of two precision resistors R1 and R2 of known resistance values ​​on a measurement channel k by a measurement circuit at a set excitation frequency.

[0042] Impedance measurement value Z of channel k LIt is obtained based on the ratio of the measured voltage and the excitation current. The measured impedance Z L It can be expressed as:

[0043]

[0044] Where L is the relay path inductance, ω is the angular frequency of the excitation signal, C1 is the parasitic capacitance of the connection line from the multiplexer output to the relay input, C2 is the parasitic capacitance of the cable connecting the data acquisition system and the electrode array, and Z LX is the actual impedance of the object being measured.

[0045] When the measured object is resistors R1 and R2, the corresponding impedance measurement values ​​are expressed as Z1 and Z2. When the measured object is precision resistor R1, the actual impedance Z of the measured object is L1 =R1, when the object to be measured is a precision resistor R2, the actual impedance of the object to be measured is Z L2 =R2.

[0046] S2. Based on the amplitude Z1, a simplified circuit model is used to obtain the value of the distributed capacitance C0, where C0 is the parasitic capacitance between the input pins of the adjacent channel multiplexing switches.

[0047] Figure 4 The equivalent model of the measuring circuit containing distributed circuit parameters established when the impedance is measured by the method of the present invention, wherein R S is the current sampling resistor, by measuring R S The voltage value on the excitation current value I S , R SWi and R SWj represent the on-resistance of the multiplexer switch on the selected measurement channel i and channel j, respectively, L i and L j Represents the relay conduction inductance on channel i and channel j respectively, and the impedance to be measured is Z LX Equivalent to resistance R X With C X If R1 is small enough, the impedance Z after R1 and C2 are connected in parallel is LX_C2 satisfy:

[0048]

[0049] Where, ω=2πf, and f=1MHz, and R1<50Ω, then (ωR1C2) 2 <9.87×10 -4 <<1, so |Z LX_C2 |≈R1.

[0050] At this time, measure the voltage V i and V jThe corresponding measured impedance Z′ L can be simplified to:

[0051]

[0052] Among them, the L value obtained by the impedance analyzer is 1.6pH, then 2ω 2 LC1<1.6×10 -2 <<1 and (ωR1C1) 2 <9.87×10 -4 <<1, so |Z′ L |≈|R1+2jωL|.

[0053] The above steps are expressed as Figure 4 The equivalent model of the measurement circuit shown is simplified to Figure 5 Equivalent model of the circuit shown.

[0054] When the R1 value is small enough to meet the circuit simplification requirements, Z′ L The parasitic capacitance C0 between the input pins of the adjacent channel multiplexing switch satisfies the following formula:

[0055]

[0056] Among them, R SW is the on-resistance of the multiplexer switch, Z C0 is the impedance of C0, I ij is the actual current flowing through the measurement object, Z t To measure the actual voltage difference V according to the channel ij =|V i -V j | and the measured sampling current |I S |The measured impedance amplitude is obtained. When the measured object is R1, Z t =Z1. R SW and L are intrinsic properties of the device and their values ​​are determined, so they can be obtained by Z t and Z′ L The relationship between and is used to find the value of C0.

[0057] At this time, according to Figure 5 The equivalent circuit of Z can be obtained according to the following equations C0 The value of C0 is obtained as follows:

[0058]

[0059] S3. Obtain the values ​​of distributed capacitances C1 and C2 using a global probability search algorithm. There are many methods for global probability search algorithms. This embodiment uses a simulated annealing algorithm to estimate the distributed capacitance values.

[0060] refer to Figure 2 As shown, in this embodiment, step S3 specifically includes:

[0061] S301, initialize the problem parameters, set the initial solution S = (C1, C2), and input the measured value Z2 of the precision resistor R2 at 1MHz, and calculate the objective function value of the initial solution, where the objective function is:

[0062] minimizeE(S)=||Z t_s (C1, C2)|-|Z2|| 2

[0063] Where Z t_s The simulation value calculated by substituting the solution S into the circuit model represents the impedance value obtained by directly measuring the voltage difference divided by the excitation current value of the measurement circuit.

[0064] S302, generating a new solution S' by performing a small random perturbation on the solution S, and calculating the objective function value E(S) of the new solution;

[0065] S303, if the objective function E(S') of the new solution is smaller than the current solution E(S), the new solution S' is accepted;

[0066] S304, if E(S') is greater than the current solution E(S), then the acceptance probability exp((E(S)-E(S')) / T) is used to decide whether to accept the new solution, where T is the current temperature. If the random number between (0, 1) is less than the acceptance probability, then the new solution is accepted, otherwise the new solution is not accepted;

[0067] S305, lower the temperature, T = α*T, continue iterating until the iteration is completed, and output the optimized solution S = (C1, C2).

[0068] S4, obtaining a measurement circuit model based on distributed circuit parameters, where the distributed circuit parameters include the distributed capacitances C0, C1 and C2 obtained by adaptive identification in steps S2 and S3 and the device intrinsic property R SW and L.

[0069] S5. Correct the measured impedance. According to the measurement circuit model, the directly obtained measured impedance Z L The resistance value R is obtained by global optimization algorithm correction t And capacitance C t , the impedance value directly measured by the measurement object at different frequencies, especially at an excitation frequency of 500kHz to 1MHz, can be corrected. There are many global optimization algorithms, and this embodiment uses a differential evolution algorithm to correct the measured impedance.

[0070] refer to Figure 3 As shown, in this embodiment, step S5 specifically includes:

[0071] S501, initialize algorithm parameters and input measured impedance Z L The measured value Z at the current frequency f L [f], whose value is the voltage difference measurement value V at the excitation frequency f ij [f] and the excitation current measurement value I S ratio of [f];

[0072] S502, randomly generate a population {xi=[R E , C E ]};

[0073] S503, evaluate the fitness, calculate the fitness of each individual in the population, that is, the objective function value is:

[0074]

[0075] Where n is the number of individuals in the population, Z t_s (xi) is the simulation value calculated in the circuit model under different individuals;

[0076] S504, mutation operation, for each individual xi in the population, select three different individuals X a , X b , X c , generate mutation vector V i :

[0077] V i =X a +F×(X b -X c )

[0078] Where F is the variation factor;

[0079] S505, crossover operation, for each individual xi and its mutation vector V i Perform crossover to generate the test vector U i , for the jth component:

[0080]

[0081] where j rand is a randomly selected index, ensuring that at least one component comes from the mutation vector, and CR is the crossover probability;

[0082] S506, select operation, compare test vector U i The fitness of the target vector xi is calculated, and individuals with better fitness are selected to enter the next generation. The iteration continues until the termination condition is met, and the individual with the best fitness in the current population is output as the optimal solution.

[0083] The corrected impedance value is separated into resistance value R EAnd capacitance C E , the corrected impedance Z is obtained according to the following formula E The real and imaginary parts of :

[0084]

[0085] It can be seen from the above implementation scheme that the present invention measures different precision resistors R1 and R2 at an excitation frequency of 1MHz to obtain corresponding impedance values, combines the impedance analyzer and the measured impedance value of R1 to obtain capacitance C0, and then obtains capacitances C1 and C2 based on the measured impedance value of R2 through a global probability search algorithm, thereby identifying the distributed parameter model of the measurement circuit. Based on the circuit model at this time, the different measured impedances at different frequencies are corrected according to the global optimization algorithm to obtain the resistor R constituting the measured impedance. E and capacitor C E .

[0086] The method of the present invention flexibly combines the measured impedance and the global probability search algorithm, adaptively adjusts the circuit distribution parameters, and maintains high precision and stability of the measurement in a changing environment. At the same time, the actual impedance value can be analyzed from the impedance measurement value of the tested object to eliminate the influence of the distributed circuit parameters and ensure the accuracy of the measured impedance value after correction, especially in the range of 500kHz to 1MHz excitation frequency.

[0087] If the above method is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0088] Embodiment 2:

[0089] Embodiment 2 is a preferred embodiment of Embodiment 1, and is used to illustrate the present invention in more detail.

[0090] Impedance to be measured Z L The impedance amplitude {|Z is formed by connecting a 499Ω precision resistor and a 0pF precision capacitor in parallel to measure the impedance amplitude {|Z L(k)|, k=1,2,3}, where k is the frequency index, corresponding to 500kHz, 800kHz and 1MHz respectively.

[0091] according to Figure 1 and Figure 2 The steps shown in the figure can obtain the distributed circuit parameters: R SW =126Ω, L=1.6μH, C0=16pF, C1=37pF, C2=59.7pF.

[0092] refer to Figure 6 As shown, Figures (a), (b), and (c) correspond to bar graphs of the relative residuals of the simulated values ​​and the measured values ​​of the equivalent circuit model obtained by the method of the present invention at 500kHz, 800kHz, and 1MHz, respectively, when R2=499Ω, wherein the relative residual distributions are all less than 0.5%, and most of them are distributed on the left side of the average value, which means that the equivalent circuit model has a high goodness of fit;

[0093] according to Figure 3 The steps shown are for {|Z L (k)|, k=1,2,3} The resistance value {|R E (k)|, k = 1, 2, 3} and capacitance {|C E (k)|,k=1,2,3}, thus we get Figure 7 (a) Box plot of the relative error between the resistance part of the measured impedance and the theoretical value before and after correction. It can be clearly seen that the overall relative error after correction is significantly reduced compared to before correction, especially at 1MHz, the relative error is reduced from 5.8% before correction to less than 0.5%.

[0094] Embodiment 3:

[0095] Embodiment 3 is a preferred embodiment of Embodiment 1, and is used to illustrate the present invention in more detail.

[0096] Impedance to be measured Z L The impedance amplitude {|Z is formed by connecting a 499Ω precision resistor and a 56pF precision capacitor in parallel to measure the impedance amplitude at different excitation frequencies. L (k)|, k=1,2,3}, where k is the frequency index, corresponding to 500kHz, 800kHz and 1MHz respectively.

[0097] according to Figure 3 The steps shown are for {|Z L (k)|, k=1,2,3} The resistance value {|R E (k)|, k = 1, 2, 3} and capacitance {|C E (k)|,k=1,2,3}, thus we get Figure 7(b) The box plot of the relative error between the resistance part of the measured impedance before and after correction and the theoretical value shows that the relative error of the resistance part after correction is greatly reduced as a whole compared with that before correction, especially at 1MHz, the relative error is reduced from 12.5% ​​before correction to less than 0.5%. Figure 8 is the corrected capacitance value. It can be seen that the corrected capacitance gradually approaches 56pF as the frequency increases.

Claims

1. A measurement impedance correction method based on adaptive identification of circuit distributed parameters, characterized in that: The following steps are involved: Obtain two amplitudes Z1 and Z2 obtained by measuring the impedance of two precision resistors R1 and R2 of known resistance values ​​at a set excitation frequency by a measuring circuit; Based on the amplitude Z1, a simplified circuit model is used to obtain the value of the distributed capacitance C0, and based on the amplitude Z2, a global probability search algorithm is used to obtain the values ​​of the distributed capacitances C1 and C2; Constructing a measurement circuit model corresponding to the measurement circuit based on the distributed capacitances C0, C1 and C2; According to the measurement circuit model, the directly obtained measured impedance is corrected by using a global optimization algorithm.

2. The measurement impedance correction method based on circuit distributed parameter adaptive identification according to claim 1, characterized in that: The resistance of the precision resistor R1 is less than 50Ω.

3. The measurement impedance correction method based on circuit distributed parameter adaptive identification according to claim 1, characterized in that: The excitation frequency is set to 1 MHz.

4. The measurement impedance correction method based on circuit distributed parameter adaptive identification according to claim 1, characterized in that: The value of the distributed capacitance C0 is obtained based on the amplitude Z1 and the measured impedance amplitude obtained according to the channel measurement voltage difference and the sampled current.

5. The measurement impedance correction method based on circuit distributed parameter adaptive identification according to claim 1, characterized in that: The global probability search algorithm is a simulated annealing algorithm.

6. The measurement impedance correction method based on circuit distributed parameter adaptive identification according to claim 5, characterized in that: When the simulated annealing algorithm is used to estimate the values ​​of the distributed capacitances C1 and C2, the objective function expression of the design is: minimizeE(S)=||Z t_s (C1,C2)|-|Z2|| 2 Among them, Z t_s (C1, C2) is a simulation value of the circuit model calculated based on the solution S = (C1, C2), and represents the impedance value obtained by directly measuring the voltage difference and dividing the excitation current value of the measurement circuit.

7. The measurement impedance correction method based on circuit distributed parameter adaptive identification according to claim 1, characterized in that: The global optimization algorithm is a differential evolution algorithm.

8. The measurement impedance correction method based on circuit distributed parameter adaptive identification according to claim 7, characterized in that: The process of using the differential evolution algorithm for correction includes: Initialize the algorithm parameters, randomly generate a population, split the corrected measured impedance into resistance and capacitance, and use the corrected resistance and capacitance as individuals in the population; Calculate the fitness of each individual in the population; The individuals in the population are subjected to mutation, crossover and selection operations, and individuals with better fitness are selected to enter the next generation. The iteration continues until the termination condition is met, and the individual with the best fitness in the current population is output as the optimal solution. The corrected measured impedance is obtained based on the optimal solution.

9. A computer-readable storage medium, characterized in that: It comprises one or more programs for execution by one or more processors of an electronic device, wherein the one or more programs comprise instructions for executing the measurement impedance correction method based on adaptive identification of circuit distributed parameters as claimed in any one of claims 1 to 8.

10. An electronic device, characterized in that: The system comprises one or more processors, a memory and one or more programs stored in the memory, wherein the one or more programs include instructions for executing the measurement impedance correction method based on adaptive identification of circuit distributed parameters as claimed in any one of claims 1 to 8.

Images