Intelligent generation method and system of field effect transistor device model

Abstract

The invention discloses an intelligent generation method and system for a field-effect transistor device model, and relates to the technical field of semiconductor device modeling, and the method comprises the steps: collecting FET related data in a working process of a field-effect transistor device in real time, carrying out the preprocessing, and rapidly recognizing the potential deviation and coupling change according to the preprocessed FET related data, thereby achieving the intelligent generation of the field-effect transistor device model. The method comprises the following steps: accurately modeling the frequency domain response characteristics of the field effect transistor device, constructing an FET frequency domain equivalent current model, and automatically sending a calibration instruction and controlling the field effect transistor device to calibrate when detecting that the working state of the field effect transistor device is abnormal, so as to realize the timely identification and processing of the abnormality of the field effect transistor device. The stability and the output consistency of the field effect transistor device in the signal amplification process are ensured, a feedback mechanism with intelligent evaluation and adaptive calibration capabilities is constructed, and the long-term operation reliability and the signal processing precision of the field effect transistor device in a complex application scene are effectively improved.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of semiconductor device modeling, in particular to an intelligent generation method and system of a field effect transistor (FET) device model. BACKGROUND

[0002] The FET is a kind of semiconductor device based on electric field regulation of conductive channel, and its structure mainly includes a source, a drain, a gate and a semiconductor channel. The FET device adjusts the carrier concentration in the channel through the gate voltage, thereby controlling the current between the source and the drain. The FET has the characteristics of high input impedance, low power consumption and fast response, and is widely used in high-sensitivity electronic systems such as signal amplification, analog modulation, charge detection and biological signal acquisition.

[0003] In specific application scenarios such as flexible wearable devices, brain-computer interface systems and biochemical sensing platforms, the FET device often needs to be physically coupled with non-traditional contact media such as human skin, electrolyte or microneedle array. Such non-ideal contact conditions can cause significant drift in the gate potential, capacitance structure and transconductance characteristics of the FET, thereby affecting the current response stability of the FET in the frequency domain. For example, in a flexible wearable device, the contact impedance between the electrode and the skin can fluctuate significantly with changes in user skin humidity, motion state and micro-sweat composition, resulting in a deviation between the model prediction and the actual output. This deviation can cause the FET field effect transistor device to perform poorly in low signal-to-noise ratio scenarios, making it difficult to restore weak-amplitude and high-frequency neural signals, resulting in amplitude distortion or response attenuation in the FET device output current, and seriously affecting the performance reliability of the FET device in key signal sensing tasks. SUMMARY

[0004] To overcome the deficiencies of the prior art, the application provides an intelligent generation method and system of a field effect transistor device model. The FET-related data in the working process of the field effect transistor device is collected in real time, an FET frequency domain equivalent current model is constructed to evaluate the working state of the field effect transistor device, and the field effect transistor device is calibrated to realize timely identification and processing of field effect transistor device abnormalities, thereby ensuring the stability and output consistency of the field effect transistor device in the signal amplification process. A feedback mechanism with intelligent evaluation and adaptive calibration capability is constructed to effectively improve the long-term operation reliability and signal processing accuracy of the field effect transistor device in complex application scenarios, and the problems mentioned in the background art are solved.

[0005] To achieve the above purpose, the application is implemented by the following technical scheme: comprising the following steps:

[0006] S1, collecting FET-related data in the working process of the field effect transistor device in real time, preprocessing, and constructing an FET data set S according to the preprocessed FET-related data;

[0007] S2, performing potential shift condition analysis and capacitance series analysis based on the FET data set S to obtain an equivalent gate potential shift amount ΔPy and an equivalent total coupling capacitance Ccoup;

[0008] S3, based on the equivalent gate potential shift amount ΔPy and the equivalent total coupling capacitance Ccoup, calculating and obtaining a corrected threshold voltage Vthref, an effective input voltage Vin(ω) and a corrected transconductance gmeff(ω) to build a FET frequency domain equivalent current model;

[0009] S4, according to the FET frequency domain equivalent current model, obtaining an amplitude shift rate ΔAMP(ω), and setting an amplitude shift threshold ф, comparing the amplitude shift rate ΔAMP(ω) with the amplitude shift threshold ф, evaluating the field effect transistor device working state, and marking the field effect transistor device as an abnormal device and generating a device calibration instruction when the field effect transistor device working state is abnormal;

[0010] S5, according to the field effect transistor device working state evaluation result, when the field effect transistor device is an abnormal device, calibrating the field effect transistor device according to the device calibration instruction.

[0011] Preferably, S1 includes the following specific steps:

[0012] S11, embedding an embedded wearable sensor array in the field effect transistor device to monitor the field effect transistor device in real time using the embedded wearable sensor array, obtaining FET related data at the gate-skin contact interface during the working process of the field effect transistor device, and preprocessing the obtained FET related data, and constructing a FET data set S according to the preprocessed FET related data, wherein the preprocessing refers to signal denoising, baseline calibration, time domain and frequency domain normalization and outlier rejection;

[0013] The FET related data includes voltage fluctuation signal V, electrochemical impedance Zelec, interfacial chloride ion activity aCI, double-layer capacitance density CdL, interfacial contact area Ac and interfacial contact temperature Tc.

[0014] The embedded wearable sensor array includes a voltage sensor, a micro impedance spectrum sensor, an ion concentration sensor, a capacitive contact sensor and a contact type thermistor.

[0015] Preferably, S2 includes the following specific steps:

[0016] S21, according to the FET data set S, analyzing the potential shift condition caused by the change of chloride ion activity under the condition that the gate of the field effect transistor device is in contact with the skin to obtain an equivalent gate potential shift amount ΔPy, wherein the equivalent gate potential shift amount ΔPy is obtained in the following manner:

[0017]

[0018] wherein, γCI represents the response sensitivity coefficient of chloride ion, k represents the Boltzmann constant, Tc represents the interface contact temperature, q represents the elementary charge, aCI represents the interface chloride ion activity, aCIref represents the standard interface chloride ion activity, and ln represents the natural logarithm function.

[0019] Preferably, the S2 further comprises the following specific steps:

[0020] S22, multiplying the double-layer capacitance density CdLand the interface contact area Acunder the physiological contact state to obtain the interface polarization layer capacitance C1according to the FET data set S;

[0021] S23, obtaining the gate oxide capacitance Coxof the field effect transistor device through the field effect transistor device process database, and analyzing the interface polarization layer capacitance C1and the gate oxide capacitance Coxin series according to the capacitance series principle to obtain the equivalent total coupling capacitance Ccoupto be formed by the gate dielectric structure of the field effect transistor device under the biological contact condition, wherein the equivalent total coupling capacitance Ccoupto be obtained in the following manner:

[0022] Preferably, the S3 comprises the following specific steps:

[0023] S31, obtaining the calibrated threshold voltage Vthref of the field effect transistor device according to the field effect transistor device process database, and calculating the difference between the calibrated threshold voltage Vthref and the equivalent gate potential offset ΔPy to correct the threshold voltage and obtain the corrected threshold voltage Vtheff under the physiological contact state;

[0024] S32, performing Fourier transform on the voltage fluctuation signal V according to the FET data set S to identify the main frequency point f of the voltage fluctuation signal V and convert to obtain the angular frequency ω of the main frequency point f, wherein ω = 2πf, and performing coupling analysis on the equivalent total coupling capacitance Ccoupto and the angular frequency ω of the main frequency point f, using the capacitive impedance calculation formula to perform complex domain division calculation to obtain the gate input impedance value Zin(ω) of the field effect transistor device at the angular frequency ω, wherein the calculation formula of the gate input impedance value Zin(ω) is wherein, j represents the complex unit.

[0025] Preferably, the S3 further comprises the following specific steps:

[0026] S33, calculating the signal coupling coefficient η based on the gate input impedance value Zin(ω) and the electrochemical impedance Zelec(ω), wherein the calculation formula of the signal coupling coefficient η is

[0027] S34, according to the field effect tube device process database, and combined with the signal coupling coefficient η, the corrected transconductance gmeff(ω) of the field effect tube device under the physiological contact state is calculated, wherein the corrected transconductance gmeff(ω) is specifically obtained in the following manner:

[0028]

[0029] In the formula, gmref represents the transconductance of the field effect tube device under the standard temperature state, η represents the signal coupling coefficient, Tref represents the reference temperature, Tc represents the interface contact temperature, and m represents the exponential factor.

[0030] Preferably, the specific steps of S3 further include:

[0031] S35, the signal coupling coefficient η is multiplied by the original input voltage Vinraw(ω) to obtain the effective input voltage Vin(ω), and the corrected transconductance gmeff(ω) and the corrected threshold voltage Vtheff are coupled to build a FET frequency domain equivalent current model, wherein the FET frequency domain equivalent current model has the following specific form:

[0032] Id(ω)=gmeff(ω)·[Vin(ω)-Vtheff];

[0033] In the formula, Id(ω) represents the drain output current of the field effect tube device at the angular frequency ω.

[0034] Preferably, the specific steps of S4 include:

[0035] S41, according to the FET frequency domain equivalent current model, the drain output current Id(ω) of the field effect tube device at the angular frequency ω is obtained, and the difference between the theoretical drain output current Idref(ω) of the field effect tube device under the standard reference state is calculated to obtain the current offset of the field effect tube device at the angular frequency ω, and the current offset is divided by the theoretical drain output current Idref(ω) to perform normalization processing to obtain the amplitude shift rate ΔAMP(ω);

[0036] S42, collect the historical FET related data of the field effect tube device under the standard state as samples, perform statistical analysis, and select the amplitude shift rate ΔAMP(ω) of the upper limit value of the 95% confidence interval as the amplitude shift threshold ф, and compare the amplitude shift threshold ф with the amplitude shift rate ΔAMP(ω) to evaluate the working state of the field effect tube device, and the specific evaluation content is as follows:

[0037] If the amplitude shift rate ΔAMP(ω) is greater than or equal to the amplitude shift threshold ф, it is determined that the working state of the field effect tube device is abnormal, and the field effect tube device is marked as an abnormal device, and a device calibration instruction is generated;

[0038] If the amplitude shift rate ΔAMP(ω) < amplitude shift threshold ф, it is determined that the field effect transistor device is in a normal working state, and no processing is required.

[0039] Preferably, the specific steps of S5 include:

[0040] S51, according to the field effect transistor device working state evaluation result, when the field effect transistor device is an abnormal device, according to the equipment calibration instruction to calibrate the field effect transistor device, wherein the equipment calibration instruction includes electrical compensation adjustment instruction and mechanical contact compensation adjustment instruction;

[0041] The electrical compensation adjustment instruction refers to using a digital controller to control the gate input voltage in the interval {0.38V, 0.42V} and using a programmable band-pass filter to control the signal bandwidth of the field effect transistor device input port in the interval {0.5Hz, 30Hz};

[0042] The mechanical contact compensation adjustment instruction refers to using a micro liquid pump to control the electrolyte liquid, and controlling the liquid film thickness in the interval {0.5Hz, 30Hz}.

[0043] An intelligent generation system of a field effect transistor device model, comprising a multi-source data collection module, a data analysis module, an FET frequency domain equivalent current model construction module, an intelligent evaluation module and a calibration module;

[0044] The multi-source data collection module is used to collect FET related data in the working process of the field effect transistor device in real time, pre-process, and construct an FET data set S according to the pre-processed FET related data;

[0045] The data analysis module is used to analyze the potential shift and the capacitance series according to the FET data set S, so as to obtain the equivalent gate potential shift ΔPy and the equivalent total coupling capacitance Ccoup;

[0046] The FET frequency domain equivalent current model construction module is used to calculate and obtain the corrected threshold voltage Vthref, the effective input voltage Vin(ω) and the corrected transconductance gmeff(ω) based on the equivalent gate potential shift ΔPy and the equivalent total coupling capacitance Ccoup, so as to construct the FET frequency domain equivalent current model;

[0047] The intelligent evaluation module is used to obtain the amplitude shift rate ΔAMP(ω) according to the FET frequency domain equivalent current model, set the amplitude shift threshold ф, compare with the amplitude shift rate ΔAMP(ω), evaluate the working state of the field effect transistor device, and when the working state of the field effect transistor device is abnormal, mark the field effect transistor device as an abnormal device, and generate an equipment calibration instruction;

[0048] The calibration module is used for calibrating the field effect transistor device according to the device calibration instruction when the field effect transistor device is an abnormal device according to the field effect transistor device working state evaluation result.

[0049] The application provides an intelligent generation method and system for a field effect transistor device model.

[0050] (1) By introducing the equivalent gate potential offset and the equivalent total coupling capacitance parameters, a dynamic frequency domain modeling framework of the FET device in the physiological contact state is constructed, compared with the traditional modeling method based on the ideal static electrical parameters, the actual influence of physiological factors such as the chloride ion activity, the capacitance polarization and the temperature disturbance in the skin electrode contact interface on the device characteristics can be accurately reflected, the physical consistency and the electrical response prediction accuracy of the modeling result in the physiological contact state are significantly improved, the problem that the model in the prior art is difficult to adapt to the hot and humid environment, the motion state and the like is solved, and the real-time restoration ability of the amplification performance is ensured.

[0051] (2) By constructing the FET frequency domain equivalent current model based on the modified threshold voltage and the modified transconductance, and introducing the signal coupling coefficient and the amplitude offset rate and other response evaluation parameters, the amplification performance drift problem caused by the contact state or the environmental fluctuation can be identified in the actual operation process of the field effect transistor device, the amplitude offset rate ΔAMP(ω) and the threshold value ф obtained by calculation are compared and judged, and the device calibration instruction is automatically generated in the abnormal state, so that the adaptive correction of the bias state of the FET device is realized, the robustness of the system to the sudden disturbance is significantly improved, and the signal output consistency and the system stability of the field effect transistor device in the long-time operation are ensured.

[0052] (3) By automatically generating and executing the device calibration instruction after evaluating the abnormal field effect transistor device, the field effect transistor device is calibrated, the adaptability of the field effect transistor device to the complex physiological contact environment is improved, the non-stable factors such as temperature fluctuation, contact area change and electrochemical disturbance can be continuously coped with, the intelligent self-update of the FET modeling parameters and the amplification performance is realized, the signal distortion risk caused by the device drift is effectively reduced, and the neural signal acquisition quality and the user interaction experience are improved. BRIEF DESCRIPTION OF DRAWINGS

[0053] Figure 1 It is a step schematic diagram of the intelligent generation method of the field effect transistor device model of the application;

[0054] Figure 2 It is a block diagram of the intelligent generation system of the field effect transistor device model of the application;

[0055] Figure 3 It is a step schematic diagram of the working state evaluation process of the field effect transistor device of the application;

[0056] Figure 4 A broken line graph for evaluating the working state of the field effect transistor device of the application. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of the application will be apparently and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments. Based on the embodiments in the application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the application.

[0058] Embodiment 1

[0059] Please refer to Figure 1 The application provides an intelligent generation method of a field effect transistor device model. To achieve the above object, the application is implemented by the following technical scheme: comprising the following steps:

[0060] S1, collecting FET related data in a working process of a field effect transistor device in real time, pre-processing, and constructing an FET data set S according to the pre-processed FET related data;

[0061] S2, performing potential offset condition analysis and capacitance series analysis according to the FET data set S to obtain an equivalent gate potential offset amount ΔPy and an equivalent total coupling capacitance Ccoup;

[0062] S3, calculating and obtaining a corrected threshold voltage Vthref, an effective input voltage Vin(ω) and a corrected transconductance gmeff(ω) based on the equivalent gate potential offset amount ΔPy and the equivalent total coupling capacitance Ccoup to construct an FET frequency domain equivalent current model;

[0063] S4, obtaining an amplitude offset rate ΔAMP(ω) according to the FET frequency domain equivalent current model, comparing the amplitude offset rate ΔAMP(ω) with an amplitude offset threshold ф, evaluating the working state of the field effect transistor device, and marking the field effect transistor device as an abnormal device and generating a device calibration instruction when the working state of the field effect transistor device is abnormal;

[0064] S5, according to the evaluation result of the working state of the field effect transistor device, calibrating the field effect transistor device according to the device calibration instruction when the field effect transistor device is an abnormal device.

[0065] In this embodiment, through the whole-process closed-loop control method from data acquisition-electrical modeling-state evaluation-calibration feedback, compared with the limitations of relying on ideal static model in the prior art, which is difficult to adapt to the changes of physiological contact state, the scheme introduces equivalent potential and coupling capacitance parameters, realizes dynamic modeling of device performance changes under non-ideal contact interface, significantly improves the environmental adaptability and accuracy of modeling, at the same time, by setting the amplitude offset threshold, the abnormality of field effect transistor device is identified and processed in time, ensuring the stability and output consistency of field effect transistor device in signal amplification process, finally, through the automatic generation and execution of equipment calibration instruction, a feedback mechanism with intelligent evaluation and adaptive calibration capability is constructed, which effectively improves the long-term operation reliability and signal processing accuracy of field effect transistor device in complex application scenarios.

[0066] Embodiment 2

[0067] Please refer to Figure 1 , specifically: S1 specific steps include:

[0068] S11, embedding a wearable sensor array in the field effect transistor device to monitor the field effect transistor device in real time using the embedded wearable sensor array, to obtain FET related data at the gate-skin contact interface of the field effect transistor device during operation, and to preprocess the obtained FET related data, and to construct a FET data set S according to the preprocessed FET related data, wherein preprocessing refers to signal denoising, baseline calibration, time domain and frequency domain normalization, and outlier rejection;

[0069] Wherein, signal denoising refers to using wavelet denoising technology to denoise the signal, baseline calibration refers to using local extreme value smoothing method to calibrate the direct current drift in the signal, time domain and frequency domain normalization refers to using Z-score standardization processing parameter to eliminate parameter dimension influence, and outlier rejection refers to using quartile method for outlier identification and rejection;

[0070] The interface refers to the physical-chemical interaction boundary region formed by the contact of the gate of the field effect transistor device and its covered gate dielectric layer with the external medium, which simultaneously exists various physical processes such as electron transport, dielectric polarization, ion migration, heat conduction and fluid disturbance;

[0071] The FET related data includes voltage fluctuation signal V, electrochemical impedance Zelec, interface chloride ion activity aCI, double-layer capacitance density CdL, interface contact area Ac and interface contact temperature Tc;

[0072] The embedded wearable sensor array includes voltage sensor, micro impedance spectrum sensor, ion concentration sensor, capacitive contact sensor and contact type thermistor.

[0073] The voltage fluctuation signal V refers to the instantaneous fluctuation value of the gate input voltage of the field effect tube device under the action of the biological interface, which is collected and obtained by a voltage sensor;

[0074] The electrochemical impedance Zelec refers to the resistance of charge transfer at the electrode-electrolyte interface, which is collected and obtained by a micro-impedance spectrum sensor;

[0075] The interface chloride ion activity aCI refers to the effective activity of the chloride ion concentration at the interface between the gate and the skin, which is collected and obtained by an ion concentration sensor;

[0076] The double-layer capacitance density CdL refers to the charge storage capacity of the double-layer at the electrode surface, which is obtained by frequency domain capacitance measurement of a micro-impedance spectrum sensor;

[0077] The interface contact area Ac refers to the effective contact area of the interface between the gate of the field effect tube device and the skin, which is collected and obtained by a capacitive contact sensor;

[0078] The interface contact temperature Tc refers to the temperature of the interface between the gate of the field effect tube device and the skin, which is collected and obtained by a contact thermistor.

[0079] In this embodiment, by arranging an embedded wearable sensor array inside the field effect tube device, real-time monitoring of the gate-skin contact interface is realized, and signal preprocessing techniques such as wavelet denoising, extreme value smoothing, Z-score standardization, and quartile outlier removal are introduced, effectively improving the accuracy and robustness of FET-related data. At the same time, the collection of FET-related data can comprehensively characterize the interface state of the field effect tube device, enhance the physical authenticity and adaptability of the model construction, and the constructed FET data set S can provide high-quality basic data for subsequent offset calculation, frequency domain modeling, and amplifier performance evaluation. Compared with the existing single-dimensional signal collection method, the comprehensive and dynamic response capture ability of the interface modeling is significantly improved, which helps to improve the accuracy of amplifier performance evaluation and the timeliness of equipment calibration.

[0080] Embodiment 3

[0081] Please refer to Figure 1 , specifically: S2 specific steps include:

[0082] S21, according to the FET data set S, analyze the potential offset caused by the change of chloride ion activity under the condition that the field effect tube device is in contact with the skin at the gate, to obtain the equivalent gate level potential offset ΔPy, wherein the equivalent gate level potential offset ΔPy is obtained in the following way:

[0083]

[0084] In the formula, γ CI represents the chloride ion response sensitivity coefficient, k represents the Boltzmann constant, T c represents the interface contact temperature, q represents the elementary charge, a CI represents the interface chloride ion activity, a CIref represents the standard interface chloride ion activity, and ln represents the natural logarithm function.

[0085] Specific examples are as follows:

[0086] Assuming that the field effect transistor device collects FET-related data during operation, including a chloride ion response sensitivity coefficient γ CI of 0.92, an interface contact temperature T c of 310, an interface chloride ion activity a CI of 0.102, and a standard interface chloride ion activity a CIref of 0.050, the equivalent gate potential offset Δ Py is calculated as follows:

[0087]

[0088] It is shown that the equivalent gate potential offset calculated under the physiological change condition of the chloride ion activity rising from 0.050 to 0.102 at the current interface contact temperature of 310 is 0.0175.

[0089] Chloride ion response sensitivity coefficient γ CI represents the response degree of the FET device to the change in chloride ion activity, which is set by an expert according to the actual situation;

[0090] Formula derivation process and physical meaning: The basic theory of the equivalent gate potential offset Δ Py formula is derived from the Nernst equation, which describes the relationship between the electrode potential and the ion activity in the solution. In this scheme, the relative change amount of the gate potential is concerned. The formula uses the normalized FET-related data to eliminate the dimensionless FET-related data for coupling calculation, wherein, is the thermoelectric voltage coefficient, which represents the heat energy conversion coefficient of unit charge at the interface contact temperature T c , and the numerical value directly reflects the response strength of the interface charge to the ion activity change under the thermal activation condition, is the relative activity response term, which represents the change amplitude of the chloride ion activity relative to the reference state. In an electrochemical system, the chemical potential of an ion is logarithmically related to its activity. Therefore, the natural logarithm function ln is used to compress and map the activity ratio to ensure the stability and interpretability of the model under a large number of orders of magnitude of ion concentration changes. Since ideal Nernst behavior only holds under the premise of ideal electrodes, electrolytes, constant temperature, and ideal activity, there are non-ideal interfaces such as sweat and lipid layers in the contact between the FET device and the skin. A chloride ion response sensitivity coefficient γ CI is introduced to correct the actual response amplitude under the non-ideal electrical contact interface.

[0091] S22, according to the FET data set S, the double-layer capacitance density CdLand the interface contact area Ac in the physiological contact state are multiplied to obtain the interface polarization layer capacitance C1;

[0092] S23, through the field effect tube device process database, the gate oxide capacitance Cox of the field effect tube device is obtained, and according to the capacitance series principle, the interface polarization layer capacitance C1 and the gate oxide capacitance Cox are analyzed in series to obtain the equivalent total coupling capacitance Ccoup of the gate dielectric structure of the field effect tube device formed under the biological contact condition, wherein the equivalent total coupling capacitance Ccoup is obtained in the following specific way

[0093] The field effect tube device process database refers to a special database system for recording and managing the structure, electrical and response characteristic data of various types of field effect tube devices under different manufacturing process parameters.

[0094] In this embodiment, by introducing the Nernst equation theory and combining the chloride ion activity variation characteristics, the calculation model of the equivalent gate potential offset is constructed, which can truly reflect the potential response of the FET device caused by the interface ion disturbance under the physiological contact state, thereby improving the accuracy of the electrochemical dynamic process modeling. At the same time, the interface polarization layer capacitance C1 and the gate oxide capacitance Cox are integrated based on the capacitance series principle to construct the equivalent total coupling capacitance model under the biological interface, which effectively enhances the adaptability and parameter interpretability of the FET model in the physiological environment, improves the credibility of the frequency domain modeling basis, provides accurate support for the correction of transconductance, input impedance and other frequency domain indicators, and significantly optimizes the consistency between the model and the actual device behavior.

[0095] Embodiment 4

[0096] Please refer to Figure 1 and Figure 3 , specifically: S3 specific steps include:

[0097] S31, according to the field effect tube device process database, the calibration threshold voltage Vthref of the field effect tube device is obtained, and the calibration threshold voltage Vthref and the equivalent gate potential offset ΔPy are subtracted to correct the threshold voltage and obtain the corrected threshold voltage Vtheff under the physiological contact state;

[0098] S32, Fourier transform the voltage fluctuation signal V according to the FET data set S to identify the main frequency point f of the voltage fluctuation signal V, and convert the angular frequency ω of the main frequency point f, wherein ω = 2πf, and perform coupling analysis on the equivalent total coupling capacitance Ccoup and the angular frequency ω of the main frequency point f, perform complex domain division calculation using the capacitive impedance calculation formula, to obtain the gate input impedance value Zin(ω) of the field effect transistor device at the angular frequency ω, wherein the gate input impedance value Zin(ω) calculation formula is Where j represents the complex unit.

[0099] S33, based on the gate input impedance value Zin(ω) and the electrochemical impedance Zelec(ω), calculate the signal coupling coefficient η, wherein the signal coupling coefficient η calculation formula is

[0100] The signal coupling coefficient η represents the actual coupling degree of the input signal from the biological interface to the gate port of the FET device, and the formula as a whole is a typical anti-resistance voltage division coupling model, which reflects the influence of the electrochemical interface on the effective transmission ability of the signal. In a specific example, assuming that the voltage fluctuation signal V is Fourier transformed to identify the main frequency point as 50Hz, the angular frequency is 314.16, the gate input impedance value Zin(ω) is 12.0-j8.0Ω, and the electrochemical impedance Zelec(ω) is 18.0+j6.0Ω, the calculated signal coupling coefficient η is 0.48, indicating that the field effect transistor device at the input signal frequency of 50Hz, the current actual input signal has only about 48% successfully transmitted to the FET gate port, and the rest is lost at the interface electrode impedance.

[0101] S34, according to the field effect transistor device process database, and combined with the signal coupling coefficient η, calculate the modified transconductance gmeff(ω) of the field effect transistor device under physiological contact state, wherein the modified transconductance gmeff(ω) is specifically obtained as follows:

[0102]

[0103] Wherein, gmref represents the transconductance of the field effect transistor device under standard temperature state, η represents the signal coupling coefficient, Tref represents the reference temperature, Tc represents the interface contact temperature, and m represents the exponential factor.

[0104] Formula derivation process and physical meaning: the formula uses the normalized dimensionless data for coupling analysis, introduces a multi-factor coupling mechanism based on the traditional thermal degradation transconductance model, and modifies the transconductance combined with the signal coupling characteristics and contact temperature environment of the FET device under physiological contact state, wherein is a temperature normalization index correction term, which is constructed based on the exponential law of mobility decay with temperature, and can dynamically compensate the performance decay caused by the temperature rise of the physiological contact interface. The exponential factor m represents the temperature sensitivity response characteristics of the field effect transistor device material, and η represents the energy matching degree between the physiological signal at a specific frequency η and the FET structure. The formula uses a product structure to realize dynamic adjustment of transconductance, decouples and combines process basic parameters, signal sensing ability and environmental disturbance factors, improves the stability and adaptability of the device in non-ideal physiological environment, ensures the consistency and reliability of the transconductance modeling results under multiple state conditions, and has good engineering practicability and physical interpretation basis.

[0105] Specific examples are as follows:

[0106] Suppose that the transconductance gmaref of the field effect transistor device under standard temperature state is 5.0 according to the field effect transistor device process database, the reference temperature is 500, the interface contact temperature is 310, the signal coupling coefficient η is 0.48, and the exponential factor m is 1.8. At this time, the corrected transconductance gmeff(ω) of the field effect transistor device under physiological contact state is:

[0107] gmeff(ω) = 5.0 x 0.48 x 0.947 = 2.27, indicating that the corrected transconductance gmeff(ω) of the field effect transistor device under physiological contact state is 2.27, which is significantly lower than the ideal 5.0, indicating that the temperature rise and signal coupling degradation caused by physiological contact have a significant impact on the gain performance of the device.

[0108] S35, multiply the signal coupling coefficient η and the original input voltage Vinraw(ω) to obtain the effective input voltage Vin(ω), and combine the corrected transconductance gmeff(ω) and the corrected threshold voltage Vtheff to couple and construct the FET frequency domain equivalent current model, wherein the FET frequency domain equivalent current model has the following specific form:

[0109] Id(ω) = gmeff(ω) · [Vin(ω) - Vtheff];

[0110] In the formula, Id(ω) represents the drain output current of the field effect transistor device at the angular frequency ω.

[0111] Formula derivation process and physical meaning: The formula uses the normalized effective input voltage Vin(ω), the modified transconductance gmeff(ω) and the modified threshold voltage Vtheff, constructs an FET frequency domain equivalent current model, jointly models the gate input impedance, signal coupling ratio, threshold voltage temperature drift characteristics and transconductance variation, establishes the dynamic current response relationship of the field effect transistor device under the current physiological-electrochemical-thermal state, realizes the accurate expression of the current output characteristics of the FET device under the actual working state, and determines whether the current channel is opened and the channel strength in the saturation region working mode of the field effect transistor device. The linear small signal frequency domain model is adopted, which is suitable for the modeling of the sub-threshold or near-threshold working state of the frequency domain input signal, emphasizes the "amplification gain", the input term is [Vin(ω)-Vtheff], which represents the part of the input signal that exceeds the threshold voltage and can start the drain current, and gmeff(ω) gain factor is multiplied by [Vin(ω)-Vtheff] to obtain the drain output current Id(ω). The response strength of the amplification field effect transistor device under the physiological contact state is truly restored, and the interference of threshold shielding and gain drift on the output behavior is effectively avoided.

[0112] Specific examples are as follows:

[0113] Assuming that the obtained data includes: the modified transconductance gmeff(ω) is 2.27, the signal coupling coefficient η is 0.48, the original input voltage Vinraw(ω) is 0.75, and the modified threshold voltage Vtheff is 0.30;

[0114] The FET frequency domain equivalent current model is constructed to obtain the drain output current Id(ω) of the field effect transistor device at the angular frequency ω as 0.1362.

[0115] In this embodiment, the FET frequency domain equivalent current model based on the actual contact state is used to realize the accuracy improvement of the electrical performance modeling of the field effect transistor device under the non-ideal physiological environment. Compared with the traditional method of deriving the static transconductance or threshold voltage only according to the structural parameters, the present scheme introduces dynamic parameters such as signal coupling coefficient and interface temperature, and establishes a dynamic thermal-electric-signal linkage modeling mechanism. This method not only corrects the transmission distortion of the electrical signal in the complex physiological interface, but also effectively compensates for the mobility attenuation caused by the temperature rise, so that the model has stronger adaptability and scalability. The current model constructed has frequency domain analysis capability and is suitable for behavior prediction in the sub-threshold or low amplitude signal scenario, further improving the accuracy and stability of the amplifier device in the field of weak signal detection, biological interface monitoring and the like, and having good engineering applicability and modeling interpretability.

[0116] Embodiment 5

[0117] Please refer to Figure 1 , Figure 3 and Figure 4 , in particular: S4 specific steps include:

[0118] S41, according to the FET frequency domain equivalent current model, the drain output current Id(ω) of the field effect transistor device at the angular frequency ω is obtained, and the difference calculation is carried out with the theoretical drain output current Idref(ω) of the field effect transistor device under the standard reference state, the current offset of the field effect transistor device at the angular frequency ω is obtained, and the current offset is divided by the theoretical drain output current Idref(ω), and the normalization processing is carried out, so as to obtain the amplitude shift rate ΔAMP(ω);

[0119] Wherein, the theoretical drain output current Idref(ω) of the field effect transistor device under the standard reference state is obtained through the field effect transistor device process database;

[0120] S42, collect the historical FET related data of the field effect transistor device under the standard state as samples, carry out statistical analysis, and select the amplitude shift rate ΔAMP(ω) of the upper limit value of 95% confidence interval as the amplitude shift threshold ф, and compare the amplitude shift threshold ф with the amplitude shift rate ΔAMP(ω) to evaluate the working state of the field effect transistor device, the specific evaluation content is as follows:

[0121] If the amplitude shift rate ΔAMP(ω) is greater than or equal to the amplitude shift threshold ф, it is judged that the working state of the field effect transistor device is abnormal, at this time the field effect transistor device is marked as an abnormal device, and the equipment calibration instruction is generated;

[0122] If the amplitude shift rate ΔAMP(ω) is less than the amplitude shift threshold ф, it is judged that the working state of the field effect transistor device is normal, at this time no processing is needed.

[0123] In this embodiment, by constructing the working state evaluation mechanism based on the amplitude shift rate, the quantifiable judgment of the performance change of the field effect transistor device in the physiological contact environment is realized. Compared with the traditional way of taking time domain voltage drift or transconductance variation as the evaluation basis, this scheme introduces the amplitude shift rate ΔAMP(ω) as the dynamic response index, combines the data distribution under the historical standard state, sets the upper limit of 95% confidence interval as the shift threshold ф, ensures the scientificity and reliability of abnormal identification, effectively avoids the interference of multiple factors such as temperature and ion activity on single parameter evaluation, makes the abnormal judgment more adaptive and practical, at the same time, generates the equipment calibration instruction automatically, realizes the closed loop control from abnormal detection to intervention response, greatly improves the stability, accuracy and engineering controllability of the field effect transistor device model in the process of biological signal amplification and perception, meets the demand of performance self-adaptive adjustment in long-term and high-precision application scene.

[0124] Embodiment 6

[0125] Referring to Figure 1 , specifically: S5 specific steps include:

[0126] S51, according to the field effect tube device working state evaluation result, when the field effect tube device is an abnormal device, according to the equipment calibration instruction, the field effect tube device calibration is carried out, wherein the equipment calibration instruction includes electrical compensation adjustment instruction and mechanical contact compensation adjustment instruction;

[0127] The electrical compensation adjustment instruction refers to using a digital controller to control the gate input voltage in the interval {0.38V, 0.42V} and using a programmable band-pass filter to control the signal bandwidth of the field effect tube device input port in the interval {0.5Hz, 30Hz};

[0128] The mechanical contact compensation adjustment instruction refers to using a micro liquid pump to control the electrolyte liquid, and controlling the liquid film thickness in the interval {0.5Hz, 30Hz}.

[0129] In this embodiment, by introducing the equipment calibration instruction to modify the FET device, when the FET is detected to be in an abnormal state, the calibration operation is actively performed, and the consistency and stability of the device frequency domain response are improved, wherein the electrical compensation adjustment instruction adjusts the gate input voltage to the interval {0.38V, 0.42V} and limits the signal bandwidth to {0.5Hz, 30Hz}, effectively suppresses the electric field disturbance caused by environmental voltage fluctuation or high-frequency noise, and improves the excitation stability of the FET input end from the source. The mechanical contact compensation adjustment instruction optimizes the coupling state of the FET and the electrolyte interface by controlling the liquid film thickness, reduces the capacitance coupling drift caused by uneven contact pressure or interface interference, avoids the cost investment of replacing the device, improves the anti-interference ability and adaptability of the FET device in the non-ideal physiological contact environment, and overall, the calibration process of the embodiment realizes the closed-loop feedback control of the FET performance degradation state, solves the problem that the existing FET device lacks real-time adaptive adjustment capability in complex application scenarios, and ensures higher reliability and precision in the signal acquisition and amplification process.

[0130] Embodiment 7

[0131] Referring to Figure 2 , specifically: an intelligent generation system of a field effect tube device model, comprising a multi-source data collection module, a data analysis module, an FET frequency domain equivalent current model construction module, an intelligent evaluation module and a calibration module;

[0132] The multi-source data collection module is used for real-time acquisition of FET related data in the working process of the field effect tube device, preprocessing, and constructing an FET data set S according to the preprocessed FET related data;

[0133] The data analysis module is configured to analyze the potential offset and the series capacitance based on the FET data set S, to obtain an equivalent gate potential offset ΔPy and an equivalent total coupling capacitance Ccoup.

[0134] The FET frequency domain equivalent current model construction module is configured to calculate a corrected threshold voltage Vthref, an effective input voltage Vin(ω) and a corrected transconductance gmeff(ω) based on the equivalent gate potential offset ΔPy and the equivalent total coupling capacitance Ccoup, to construct the FET frequency domain equivalent current model.

[0135] The intelligent evaluation module is configured to obtain an amplitude offset rate ΔAMP(ω) based on the FET frequency domain equivalent current model, to compare the amplitude offset rate ΔAMP(ω) with an amplitude offset threshold ф, to evaluate the working state of the field effect transistor device, and to mark the field effect transistor device as an abnormal device and generate a device calibration instruction when the working state of the field effect transistor device is abnormal.

[0136] The calibration module is configured to calibrate the field effect transistor device based on the device calibration instruction when the field effect transistor device is an abnormal device according to the evaluation result of the working state of the field effect transistor device.

[0137] Although the embodiments of the present application have been shown and described, it is to be understood that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the present application.

Claims

1. A method for intelligently generating a field-effect transistor (FET) device model, characterized in that: Includes the following steps: S1. Real-time acquisition of FET-related data during the operation of the field-effect transistor device, preprocessing, and construction of FET data set S based on the preprocessed FET-related data; S2. Based on the FET data set S, perform potential offset analysis and capacitance series analysis to obtain the equivalent gate potential offset ΔPy and the equivalent total coupling capacitance Ccoup. S3. Based on the equivalent gate potential offset ΔPy and the equivalent total coupling capacitance Ccoup, calculate and obtain the corrected threshold voltage Vthref, the effective input voltage Vin(ω), and the corrected transconductance gmeff(ω) to construct the FET frequency domain equivalent current model. S4. Based on the FET frequency domain equivalent current model, obtain the amplitude offset rate ΔAMP(ω), set the amplitude offset threshold ф, compare it with the amplitude offset rate ΔAMP(ω), evaluate the working state of the field-effect transistor device, and when the working state of the field-effect transistor device is abnormal, mark the field-effect transistor device as an abnormal device and generate a device calibration command. S5. Based on the evaluation results of the operating status of the field-effect transistor (FET), if the FET is an abnormal device, calibrate the FET according to the equipment calibration instructions.

2. The intelligent generation method for a field-effect transistor device model according to claim 1, characterized in that: The specific steps in S1 include: S11. An embedded wearable sensor array is deployed inside the field-effect transistor device to monitor the field-effect transistor device in real time, acquire FET-related data at the gate-skin contact interface during the operation of the field-effect transistor device, preprocess the acquired FET-related data, and construct a FET data set S based on the preprocessed FET-related data. The preprocessing refers to signal denoising, baseline calibration, time-domain and frequency-domain normalization, and outlier removal. FET-related data include voltage fluctuation signal V, electrochemical impedance Zelec, interfacial chloride ion activity aCI, double layer capacitance density CdL, interfacial contact area Ac, and interfacial contact temperature Tc. Embedded wearable sensor arrays include voltage sensors, miniature impedance spectroscopy sensors, ion concentration sensors, capacitive contact sensors, and contact thermistors.

3. The intelligent generation method for a field-effect transistor device model according to claim 2, characterized in that: The specific steps in S2 include: S21. Based on the FET data set S, analyze the potential shift caused by the change in chloride ion activity when the gate of the field-effect transistor is in contact with the skin, in order to obtain the equivalent gate potential shift ΔPy. The specific method for obtaining the equivalent gate potential shift ΔPy is as follows: In the formula, γCI represents the chloride ion response sensitivity coefficient, k represents the Boltzmann constant, Tc represents the interface contact temperature, q represents the elementary charge, aCI represents the interface chloride ion activity, aCIref represents the standard interface chloride ion activity, and ln represents the natural logarithm function.

4. The intelligent generation method for a field-effect transistor device model according to claim 3, characterized in that: The specific steps in S2 also include: S22. Based on the FET data set S, the double-layer capacitance density CdL and the interface contact area Ac under physiological contact state are multiplied to calculate the interface polarization layer capacitance C1. S23. Obtain the gate oxide capacitance Cox of the field-effect transistor (FET) device from the FET device process database. Based on the capacitor series principle, perform series analysis on the interface polarization layer capacitance C1 and the gate oxide capacitance Cox to obtain the equivalent total coupling capacitance Ccoup formed by the gate dielectric structure of the FET device under biological contact conditions. The specific method for obtaining the equivalent total coupling capacitance Ccoup is as follows:

5. The intelligent generation method for a field-effect transistor device model according to claim 4, characterized in that: The specific steps of S3 include: S31. Based on the field-effect transistor device process database, obtain the calibration threshold voltage Vthref of the field-effect transistor device, and calculate the difference between the calibration threshold voltage Vthref and the equivalent gate potential offset ΔPy to correct the threshold voltage and obtain the corrected threshold voltage Vtheff under physiological contact conditions. S32. Based on the FET data set S, perform a Fourier transform on the voltage fluctuation signal V to identify the dominant frequency point f of the voltage fluctuation signal V, and calculate the angular frequency ω of the dominant frequency point f, where ω = 2πf. Then, perform a coupling analysis between the equivalent total coupling capacitance Ccoup and the angular frequency ω of the dominant frequency point f, and use the capacitive impedance calculation formula to perform complex domain division to obtain the gate input impedance value Zin(ω) of the field-effect transistor device at the angular frequency ω. The formula for calculating the gate input impedance value Zin(ω) is as follows: Where j represents the complex unit.

6. The intelligent generation method for a field-effect transistor device model according to claim 5, characterized in that: The specific steps in S3 also include: S33. Based on the gate input impedance Zin(ω) and the electrochemical impedance Zelec(ω), calculate the signal coupling coefficient η, where the formula for calculating the signal coupling coefficient η is: S34. Based on the field-effect transistor (FET) device technology database and combined with the signal coupling coefficient η, calculate and obtain the corrected transconductance gmeff(ω) of the FET device under physiological contact conditions. The specific method for obtaining the corrected transconductance gmeff(ω) is as follows: In the formula, gmref represents the transconductance of the field-effect transistor device under standard temperature conditions, η represents the signal coupling coefficient, Tref represents the reference temperature, Tc represents the interface contact temperature, and m represents the exponential factor.

7. The intelligent generation method for a field-effect transistor device model according to claim 6, characterized in that: The specific steps in S3 also include: S34. Multiply the signal coupling coefficient η with the original input voltage Vinraw(ω) to obtain the effective input voltage Vin(ω). Combine this with the corrected transconductance gmeff(ω) and the corrected threshold voltage Vtheff to construct the FET frequency domain equivalent current model. The specific form of the FET frequency domain equivalent current model is as follows: Id(ω)=gmeff(ω)·[Vin(ω)-Vtheff]; In the formula, Id(ω) represents the drain output current of the field-effect transistor device at angular frequency ω.

8. The intelligent generation method for a field-effect transistor device model according to claim 7, characterized in that: The specific steps of S4 include: S41. Based on the FET frequency domain equivalent current model, obtain the drain output current Id(ω) of the field-effect transistor device at angular frequency ω, and calculate the difference between it and the theoretical drain output current Idref(ω) of the field-effect transistor device under standard reference conditions to obtain the current offset of the field-effect transistor device at angular frequency ω. Divide the current offset by the theoretical drain output current Idref(ω) and normalize it to obtain the amplitude offset rate ΔAMP(ω). S42. Collect historical FET-related data under standard conditions as a sample, perform statistical analysis, and select the amplitude offset rate ΔAMP(ω) at the upper limit of the 95% confidence interval as the amplitude offset threshold ф. Compare and analyze the amplitude offset threshold ф with the amplitude offset rate ΔAMP(ω) to evaluate the operating status of the FET. The specific evaluation content is as follows: If the amplitude offset rate ΔAMP(ω) ≥ amplitude offset threshold ф, the field-effect transistor device is determined to be in an abnormal operating state. At this time, the field-effect transistor device is marked as an abnormal device, and a device calibration command is generated. If the amplitude offset rate ΔAMP(ω) < amplitude offset threshold ф, the field-effect transistor device is considered to be operating normally, and no processing is required.

9. The intelligent generation method for a field-effect transistor device model according to claim 8, characterized in that: The specific steps of S5 include: S51. Based on the evaluation results of the operating status of the field-effect transistor device, when the field-effect transistor device is an abnormal device, the field-effect transistor device shall be calibrated according to the equipment calibration instructions. The equipment calibration instructions include electrical compensation adjustment instructions and mechanical contact compensation adjustment instructions. The electrical compensation adjustment command refers to calling the digital controller to control the gate input voltage within the range of {0.38V, 0.42V} and using a programmable bandpass filter to control the signal bandwidth of the input port of the field-effect transistor device within the range of {0.5Hz, 30Hz}. The mechanical contact compensation adjustment command refers to using a micro-liquid pump to control the electrolyte liquid and keep the liquid film thickness within the range of {0.5Hz, 30Hz}.

10. An intelligent generation system for a field-effect transistor (FET) device model, applied to the intelligent generation method for a FET device model as described in any one of claims 1-9, characterized in that: It includes a multi-source data collection module, a data analysis module, a FET frequency domain equivalent current model construction module, an intelligent evaluation module, and a calibration module; The multi-source data acquisition module is used to collect FET-related data in real time during the operation of the field-effect transistor device, perform preprocessing, and construct the FET data set S based on the preprocessed FET-related data; The data analysis module is used to perform potential offset analysis and capacitance series analysis based on the FET data set S, in order to obtain the equivalent gate potential offset ΔPy and the equivalent total coupling capacitance Ccoup. The FET frequency domain equivalent current model construction module is used to calculate and obtain the corrected threshold voltage Vthref, effective input voltage Vin(ω), and corrected transconductance gmeff(ω) based on the equivalent gate potential offset ΔPy and the equivalent total coupling capacitance Ccoup, so as to construct the FET frequency domain equivalent current model. The intelligent evaluation module is used to obtain the amplitude offset rate ΔAMP(ω) based on the FET frequency domain equivalent current model, and set the amplitude offset threshold ф. It compares the amplitude offset rate ΔAMP(ω) with the threshold to evaluate the operating status of the field-effect transistor device. When the operating status of the field-effect transistor device is abnormal, it marks the field-effect transistor device as an abnormal device and generates a device calibration command. The calibration module is used to calibrate the field-effect transistor (FET) device according to the device calibration instructions when the FET device is an abnormal device, based on the evaluation results of the FET device's operating status.

Images