A temperature control system and method for a microfluidic chip

Abstract

This invention discloses a temperature control system and method for a microfluidic chip, relating to the field of microfluidic temperature control technology. The method includes: during the microfluidic chip manufacturing process, embedding topological insulator nanowires into a substrate in a directional arrangement to form a pre-locked array; simultaneously generating an anisotropic thermally conductive network under continuous acoustic-magnetic field action; calculating topological vortex invariants to generate thermal conductivity weighting coefficients; setting a fluorescent layer on the surface of the micro-reaction cavity of the anisotropic thermally conductive network and acquiring fluorescence lifetime decay curves in real time; calculating transient temperature values ​​and temperature change rates based on a thermal conductivity-constrained variational model and the thermal conductivity weighting coefficients; calculating nonlinear feedforward compensation power based on the transient temperature values ​​and temperature change rates; calculating feedback compensation power based on the deviation field between the target temperature and the real-time temperature; and generating compensation power commands. This invention constructs a thermal conductivity frequency domain filtering function through compensation power commands, improving the uniformity and response speed of the thermal response.

Description

Technical Field

[0001] This invention relates to the field of microfluidic temperature control technology, and in particular to a temperature control system and method for a microfluidic chip. Background Technology

[0002] With the rapid development of microfluidics technology, it has shown broad application prospects in various fields such as biomedicine, chemical analysis, and environmental monitoring. Microfluidic chips, as the core carrier of this technology, enable the precise control and processing of small amounts of liquid at a microscale, realizing the transformation from laboratory research to practical applications. Temperature control is a crucial aspect of microfluidic systems because many biochemical reactions are extremely sensitive to temperature, requiring precise temperature regulation to ensure the accuracy and repeatability of experimental results. In recent years, researchers have attempted to introduce nanomaterials into microfluidic devices to improve their temperature control performance.

[0003] Although existing technologies have improved the temperature control problem of microfluidic chips to some extent, there are still two major shortcomings: First, existing temperature control devices often cannot achieve nanometer-level spatial resolution, resulting in an uneven temperature field distribution, which is a major drawback for experiments that require high-precision temperature control; Second, most existing methods rely on empirical formulas or simple mathematical models to predict and adjust the temperature, lacking a real-time feedback mechanism, making it difficult to dynamically adjust to cope with complex temperature changes. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a temperature control method for microfluidic chips to solve the problem of accurate spatial resolution of temperature field in the prior art.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a temperature control method for a microfluidic chip, comprising: embedding topological insulator nanowires into a substrate in a directional arrangement during the microfluidic chip manufacturing process to form a pre-locked array; simultaneously generating an anisotropic thermally conductive network under continuous acoustic-magnetic field action, and calculating topological vortex invariants to generate thermal conductivity weighting coefficients; setting a fluorescent layer on the surface of the micro-reaction cavity of the anisotropic thermally conductive network and acquiring fluorescence lifetime decay curves in real time; calculating transient temperature values ​​and temperature change rates based on a thermal conductivity-constrained variational model and in conjunction with the thermal conductivity weighting coefficients; calculating nonlinear feedforward compensation power based on the transient temperature values ​​and temperature change rates, calculating feedback compensation power in conjunction with the deviation field between the target temperature and the real-time temperature, and generating compensation power commands; constructing a thermal conductivity frequency domain filter function based on the compensation power commands, and performing inverse Fourier transform to generate array driving commands; transferring heat through the thermally conductive network to execute the array driving commands, while monitoring the substrate strain distribution, verifying temperature uniformity based on the strain-temperature mapping relationship, and establishing a steady-state temperature field.

[0008] In a preferred embodiment of the temperature control method for the microfluidic chip described in this invention, the following steps are taken during the microfluidic chip manufacturing process: topological insulator nanowires are embedded into the substrate in a directional arrangement to form a pre-locked array.

[0009] In the process of microfluidic chip manufacturing, topological insulator nanowires and magnetic modifiers are incorporated to generate magnetically responsive composite fluids.

[0010] A magnetically responsive composite fluid is injected into a microfluidic chip molding die, and a three-dimensional phase-modulated acoustic field and a pulsed gradient magnetic field are applied simultaneously to drive the nanowires to align and form a pre-locked array.

[0011] As a preferred embodiment of the temperature control method for the microfluidic chip described in this invention, the steps of generating an anisotropic thermal conductive network and calculating topological vortex invariants to generate thermal conductivity weight coefficients under the continuous action of acoustic-magnetic field are as follows: Based on a pre-locked array, ultraviolet gradient exposure is performed under the continuous action of acoustic-magnetic field to generate an anisotropic thermal conductive network, while monitoring the changes in magnetic field strength and fluid viscosity to generate a rheological-magnetic coupling parameter set.

[0012] A three-dimensional microstructure scan of anisotropic heat-conducting networks was performed, and topological vortex invariants were calculated using the spatial configuration of nanowires.

[0013] Based on the topological vortex invariant and the rheological-magnetic coupling parameter set, the thermal conductivity weighting coefficient is generated.

[0014] As a preferred embodiment of the temperature control method for the microfluidic chip described in this invention, the steps are as follows: a fluorescent layer is set on the surface of the micro-reaction cavity of the anisotropic thermally conductive network and the fluorescence lifetime decay curve is collected in real time. The transient temperature value and temperature change rate are calculated according to the thermal conductivity constrained variational model and combined with the thermal conductivity weight system. The specific steps are as follows: a quantum dot fluorescent layer is set on the surface of the micro-reaction cavity of the anisotropic thermally conductive network, and the fluorescence lifetime decay curve is collected by a single photon counter.

[0015] The fluorescence lifetime decay curve is input into the thermal conductivity-constrained variational model, and the transient temperature value is calculated by combining the thermal conductivity weighting coefficient and the spatiotemporal gradient of the temperature field is generated.

[0016] The rate of change of the temperature field is calculated by solving the coupled partial differential equation of heat conduction and convection based on the spatiotemporal gradient of the temperature field.

[0017] As a preferred embodiment of the temperature control method for the microfluidic chip described in this invention, the specific steps for calculating the nonlinear feedforward compensation power based on the transient temperature value and the rate of temperature change are as follows: based on the transient temperature value, the rate of temperature change, and the thermal conductivity weight coefficient, a latent variable feature vector is generated through a physically constrained variational autoencoder.

[0018] The nonlinear feedforward compensation power is calculated based on the latent variable eigenvectors and topological vortex invariants.

[0019] As a preferred embodiment of the temperature control method for the microfluidic chip described in this invention, the specific steps for calculating the feedback compensation power by combining the deviation field between the target temperature and the real-time temperature and generating the compensation power command are as follows: combining the deviation field between the target temperature and the real-time temperature and the spatial gradient of the thermal conductivity weight coefficient, the feedback compensation power is generated through a graph convolutional network.

[0020] Variational collaborative optimization is performed on the nonlinear feedforward compensation power and feedback compensation power to generate compensation power commands.

[0021] As a preferred embodiment of the temperature control method for the microfluidic chip of the present invention, the specific steps of constructing a thermal conductivity frequency domain filter function based on the thermal conductivity weight coefficient according to the compensation power command are as follows: based on the compensation power command, the power spectrum is decomposed into a frequency domain power spectrum through a three-dimensional fast Fourier transform.

[0022] A thermal conductivity frequency domain filter function is constructed based on the spatial gradient of the thermal conductivity weighting coefficient and the topological vortex invariant.

[0023] As a preferred embodiment of the temperature control method for the microfluidic chip of the present invention, the specific steps for performing inverse Fourier transform to generate array driving instructions are as follows: performing complex domain convolution operation between the frequency domain power spectrum and the thermal conductivity filter function to generate the filtered power spectrum.

[0024] The filtered power spectrum is subjected to inverse Fourier transform to construct a spatial domain driving power field, and the spatial domain driving power field is integrated and averaged according to the edge array partition to generate array driving commands.

[0025] As a preferred embodiment of the temperature control method for the microfluidic chip of the present invention, the steps of transferring heat by executing array driving commands through a heat-conducting network, simultaneously monitoring substrate strain distribution, verifying temperature uniformity based on strain-temperature mapping relationship, and establishing a steady-state temperature field are as follows: transferring heat by executing array driving commands through a heat-conducting network, simultaneously monitoring substrate strain distribution using dual-mode response characteristics, and obtaining dual-mode fluorescence response data.

[0026] A strain-temperature correlation reconstruction function is constructed based on dual-mode fluorescence response data to generate a nonlocal temperature field.

[0027] Calculate the topology alignment index based on the spatial gradient of the nonlocal temperature field and the thermal conductivity weighting coefficient, and output the verification results.

[0028] Based on the verification results, a steady-state temperature field is established when the rate of temperature change is below the quantum thermal steady-state threshold and the topological alignment index meets the standard.

[0029] Secondly, this invention provides a temperature control system for a microfluidic chip, comprising a nano-arrangement module, a fluorescence temperature measurement module, a power compensation module, a frequency domain filtering module, and a strain verification module. The nano-arrangement module is used to embed topological insulator nanowires into a substrate in a directional arrangement during the microfluidic chip manufacturing process, forming a pre-locked array. Simultaneously, under the continuous action of an acoustic-magnetic field, an anisotropic thermally conductive network is generated, and topological vortex invariants are calculated to generate thermal conductivity weighting coefficients. The fluorescence temperature measurement module is used to set a fluorescence layer on the surface of the micro-reaction cavity of the anisotropic thermally conductive network and acquire fluorescence lifetime decay curves in real time, based on thermal conductivity constraints. The system employs a variational model combined with a thermal conductivity weighting system to calculate transient temperature values ​​and the rate of temperature change. A power compensation module calculates nonlinear feedforward compensation power based on the transient temperature values ​​and the rate of temperature change, and calculates feedback compensation power by combining the deviation field between the target temperature and the real-time temperature, generating compensation power commands. A frequency domain filtering module constructs a thermal conductivity frequency domain filtering function based on the compensation power commands, performs an inverse Fourier transform, and generates array drive commands. A strain verification module executes array drive commands through the thermally conductive network to transfer heat, while simultaneously monitoring the substrate strain distribution, verifying temperature uniformity based on the strain-temperature mapping relationship, and establishing a steady-state temperature field.

[0030] The beneficial effects of this invention are as follows: by constructing a thermal conductivity frequency domain filtering function through a compensation power command, the uniformity and response speed of the thermal response are improved, and precise spatial frequency domain control of heat transfer is achieved; by executing array drive commands and synchronously monitoring substrate strain distribution, the long-term operational reliability and structural safety are improved while ensuring temperature control accuracy, and physical consistency verification and thermal steady-state confirmation in closed-loop control are achieved. Attached Figure Description

[0031] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a flowchart of a temperature control method for microfluidic chips.

[0033] Figure 2 A flowchart for generating a pre-locked array and anisotropic thermally conductive network.

[0034] Figure 3 This is a flowchart for transient temperature field calculation and power compensation.

[0035] Figure 4 A flowchart for generating compensation power commands. Detailed Implementation

[0036] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0037] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0038] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0039] Reference Figures 1-4 As one embodiment of the present invention, this embodiment provides a temperature control method for a microfluidic chip, comprising the following steps:

[0040] S1: In the microfluidic chip manufacturing process, topological insulator nanowires are embedded into the substrate in an oriented manner to form a pre-locked array. At the same time, under the continuous action of acoustic-magnetic field, an anisotropic thermal conductive network is generated, and the topological vortex invariant is calculated to generate thermal conductivity weight coefficients.

[0041] In the process of microfluidic chip manufacturing, topological insulator nanowires and magnetic modifiers are incorporated to generate magnetically responsive composite fluids.

[0042] The specific process involves uniformly dispersing topological insulator nanowires and magnetic modifiers in a specific ratio within a liquid polymer substrate during microfluidic chip manufacturing to form a magnetically responsive composite fluid. When an external magnetic field is applied, the magnetically responsive composite fluid exhibits directional alignment behavior. The topological insulator nanowires, due to their unique surface electronic state structure, can effectively regulate the heat conduction path, while the magnetic modifier changes the magnetic susceptibility of the fluid medium, enabling the nanowires to achieve controllable orientation under the influence of the magnetic field. The resulting magnetically responsive composite fluid retains the high thermal conductivity of the topological insulator nanowires and also possesses the ability to regulate the spatial distribution of the nanowires through an external magnetic field.

[0043] A magnetically responsive composite fluid is injected into a microfluidic chip molding die, and a three-dimensional phase-modulated acoustic field and a pulsed gradient magnetic field are applied simultaneously to drive the nanowires to align and form a pre-locked array.

[0044] The specific process involves injecting a magnetically responsive composite fluid into a microfluidic chip molding die while simultaneously applying a three-dimensional phase-modulated acoustic field and a pulsed gradient magnetic field. The three-dimensional phase-modulated acoustic field forms a complex standing wave field structure in the fluid by precisely controlling the phase difference of the acoustic waves, while the pulsed gradient magnetic field generates a time-varying magnetic field gradient distribution. The synergistic effect of these two fields causes the topological insulator nanowires to be arranged in an orderly manner in a preset direction in three-dimensional space. The mechanical vibration generated by the three-dimensional phase-modulated acoustic field prompts the nanowires to overcome fluid resistance and move into place, while the gradient magnetic field provides a continuous orientation guiding force. Ultimately, the nanowires form a pre-locked array with a specific spatial configuration before solidification. The pre-locked array retains the directional arrangement characteristics of the nanowires, laying the foundation for the subsequent construction of an anisotropic thermally conductive network.

[0045] The preset direction is determined by analyzing the optimal heat flow distribution based on the heat conduction path requirements of the micro-reaction cavity in the microfluidic chip.

[0046] Based on a pre-locked array, ultraviolet gradient exposure is performed under the continuous action of acoustic-magnetic field to generate an anisotropic thermal conductive network. At the same time, the magnetic field strength and fluid viscosity changes are monitored to generate a rheological-magnetic coupling parameter set.

[0047] The specific process includes, after the pre-locked array is formed, a three-dimensional phase-modulated acoustic field and a pulsed gradient magnetic field are continuously applied while ultraviolet light gradient exposure is performed. The ultraviolet light irradiates the magnetically responsive composite fluid with a spatial gradient intensity distribution, causing selective cross-linking and curing of the polymer substrate in the magnetically responsive composite fluid, thereby permanently fixing the directional alignment configuration of the topological insulator nanowires and forming an anisotropic thermally conductive network with directional thermal conductivity. During this process, the magnetic field strength change recorded by the magnetic field strength sensor and the fluid viscosity change detected by the rheometer are measured in real time. After correlation analysis of these two types of dynamic parameters, a rheological-magnetic coupling parameter set reflecting the magnetic field-rheological properties is established. The rheological-magnetic coupling parameter set fully describes the orientation dynamics of the topological insulator nanowires during the curing process.

[0048] A three-dimensional microstructure scan of the anisotropic heat-conducting network was performed, and the topological vortex invariants were calculated using the nanowire spatial configuration. The expression is as follows:

[0049] ;

[0050] in, Represents topological vortex invariants. Indicates the volume of the microfluidic chip. Represents the three-dimensional integration region. Indicates the spatial coordinates inside the microfluidic chip. Indicates the internal spatial location of the microfluidic chip The average orientation density vector of nanowires at that location.

[0051] The specific process includes using high-resolution X-ray tomography to perform three-dimensional microstructure scanning on the anisotropic thermal conductive network, obtaining spatial distribution images of the topological insulator nanowires in the solidified matrix, extracting the position coordinates and orientation angle data of the nanowires through image processing algorithms, calculating the curl distribution of the nanowire orientation field based on these spatial configuration data, integrating and normalizing the curl distribution over the entire microfluidic chip volume to obtain the topological vortex invariant characterizing the orderliness of the nanowire arrangement, which reflects the topological characteristics of the nanowire arrangement in the anisotropic thermal conductive network.

[0052] Based on the topological vortex invariant and the rheological-magnetic coupling parameter set, the thermal conductivity weighting coefficient is generated.

[0053] The specific process includes inputting topological vortex invariants and rheological-magnetic coupling parameter sets into a thermal conductivity weighting algorithm. By analyzing the orderliness of the nanowire spatial configuration reflected by the topological vortex invariants and the dynamic characteristics of the curing process recorded by the rheological-magnetic coupling parameter set, a mapping relationship between nanowire arrangement density and local thermal conductivity is established. Tensor operations are used to couple the orientation distribution characteristics represented by the topological vortex invariants with the magnetic field-viscosity correlation information contained in the rheological-magnetic coupling parameter set. Finally, a thermal conductivity weighting coefficient matrix describing the thermal conductivity of different regions of the microfluidic chip is output. The thermal conductivity weighting coefficient matrix quantifies the thermal conductivity directionality and intensity distribution characteristics of the anisotropic thermal conductive network.

[0054] S2: A fluorescent layer is set on the surface of the micro-reaction cavity of the anisotropic thermally conductive network and the fluorescence lifetime decay curve is collected in real time. The transient temperature value and temperature change rate are calculated based on the thermal conductivity-constrained variational model and the thermal conductivity weight system.

[0055] A quantum dot fluorescent layer was deposited on the surface of a microreactor with an anisotropic thermally conductive network, and the fluorescence lifetime decay curve was collected using a single-photon counter.

[0056] The specific process includes uniformly coating the surface of the microreactor with a quantum dot fluorescent layer using physical vapor deposition (PVD) technology. Quantum dot materials have temperature-sensitive fluorescence properties. When the quantum dot fluorescent layer is excited by a pulsed laser, the fluorescence intensity decays exponentially over time. The fluorescence decay process is recorded using time-correlated single-photon counting technology. The arrival time of each fluorescent photon is measured using a high-precision time-to-digital converter. After statistically analyzing a large number of photon events, a fluorescence lifetime decay curve reflecting the lifetime distribution of the quantum dot excited state is constructed. The decay time constant contained in the fluorescence lifetime decay curve has a definite correspondence with the surface temperature of the microreactor.

[0057] The fluorescence lifetime decay curve is input into the thermal conductivity-constrained variational model. The transient temperature value is calculated using the thermal conductivity weighting coefficient, and the spatiotemporal gradient of the temperature field is generated. The expression is:

[0058] ;

[0059] in, Indicates the spatial location inside the microfluidic chip and time The transient temperature value, Represents a time variable. Indicates the fluorescence temperature coefficient of the material. Indicates the intrinsic fluorescence lifetime at the reference temperature. Indicates the measured fluorescence lifetime. Indicates the thermal conductivity constraint strength coefficient. This represents the characteristic volume of a microfluidic chip. This represents the spatial gradient of the thermal conductivity weighting coefficient. Represents the temperature field gradient. This represents the topology-temperature conversion coefficient.

[0060] The specific process includes: the fluorescence lifetime decay curve is processed by the thermal conductivity-constrained variational model. First, the curve is analyzed to obtain the quantum dot fluorescence lifetime parameters. The quantum dot fluorescence lifetime parameters are matched with the temperature-lifetime correspondence. At the same time, combined with the anisotropic thermal conductivity characteristic data provided by the thermal conductivity weighting coefficient matrix, the transient temperature values ​​at each position on the surface of the microreactor are calculated by solving the variational problem that couples the thermal conductivity equation and the fluorescence thermal response equation. Then, based on the distribution of transient temperature values, spatiotemporal differential operations are performed to obtain the spatiotemporal gradient distribution of the temperature field that reflects the rate and direction of temperature change. The whole process ensures that the temperature field calculation includes both the transient temperature information detected by the quantum dot fluorescence layer and the influence of the thermal conductivity characteristics of the anisotropic thermal conductivity network is considered through the thermal conductivity weighting coefficient matrix.

[0061] Furthermore, the pre-training process of the thermal conductivity-constrained variational model first requires constructing a calibration dataset containing quantum dot fluorescence lifetime decay curves and corresponding temperature values. This is achieved by precisely controlling an isothermal platform to generate different temperature environments and recording the quantum dot fluorescence response, collecting a sufficient number of temperature-fluorescence lifetime sample pairs. Subsequently, a neural network architecture containing a thermal conductivity weight coefficient matrix as prior knowledge is established, and a variational autoencoder framework is used to encode the fluorescence lifetime features into a latent variable space representation. During the training phase, KL divergence is used to constrain the latent variable distribution and minimize the temperature prediction error, and the network parameters are optimized through a backpropagation algorithm. The final thermal conductivity-constrained variational model can accurately capture the nonlinear mapping relationship between fluorescence lifetime and temperature, while maintaining physical consistency with the thermal conductivity characteristics represented by the thermal conductivity weight coefficient matrix.

[0062] Based on the spatiotemporal gradient of the temperature field, the rate of change of the temperature field is calculated by solving the coupled partial differential equation of heat conduction and convection, and the expression is:

[0063] ;

[0064] in, This represents the rate of change of the temperature field over time. Indicates the thermal diffusivity. The Laplace operator represents the temperature field gradient. This represents the fluid velocity field vector.

[0065] The specific process includes the following steps: the spatiotemporal gradient of the temperature field is input into the solution process of the heat conduction-convection coupled partial differential equation. First, the thermal conductivity weighting coefficient matrix is ​​substituted into the heat conduction-convection coupled partial differential equation as anisotropic thermal conductivity tensor. At the same time, combined with the fluid velocity field distribution data in the microfluidic chip, a differential operator containing heat diffusion and convection terms is established on the spatiotemporal discrete grid. The partial differential equation is transformed into a system of algebraic equations through the finite volume method discretization. The partial derivative of the temperature field with respect to time is calculated using an iterative solution method to obtain the temperature field change rate, which reflects the instantaneous rate and direction characteristics of temperature change at each location in the microreactor. This temperature field change rate includes both the temperature diffusion information caused by heat conduction and the temperature field migration characteristics caused by convection effect.

[0066] S3: Calculate the nonlinear feedforward compensation power based on the transient temperature value and the rate of temperature change, and calculate the feedback compensation power by combining the deviation field between the target temperature and the real-time temperature, and generate the compensation power command.

[0067] Based on transient temperature values, temperature change rate, and thermal conductivity weighting coefficients, a latent variable feature vector is generated through a physically constrained variational autoencoder.

[0068] The specific process involves inputting transient temperature values, temperature change rate, and thermal conductivity weighting coefficients into the encoder network of a physically constrained variational autoencoder. The encoder network consists of multiple fully connected layers, each of which is subject to physical constraints based on thermodynamic laws. The transient temperature values, temperature change rate, and thermal conductivity weighting coefficients are normalized and then gradually reduced in dimensionality through nonlinear transformation of the encoder network to generate latent variable feature vectors that conform to a Gaussian distribution in the latent space. At the same time, the decoder network ensures that the latent variable feature vectors retain the thermodynamic characteristics of the original input through reconstruction loss. The final output latent variable feature vectors contain both the dynamic change information of the temperature field and the thermal conductivity characteristics of the material through the thermal conductivity weighting coefficients, forming a compact physical feature representation.

[0069] The nonlinear feedforward compensation power is calculated based on the latent variable eigenvectors and topological vortex invariants, expressed as follows:

[0070] ;

[0071] in, This represents the nonlinear feedforward compensation power. Indicates matrix transpose. Represents the weight vector The transpose of the matrix, Represents the latent variable eigenvector. Represents the gradient gain coefficient. Indicates the heat capacity conversion coefficient. Indicates the current time point, Indicates the start time point. This represents the time variable for integration.

[0072] The specific process includes inputting the latent variable eigenvector and the topological vortex invariant into the nonlinear feedforward compensation power calculation unit. First, the latent variable eigenvector is nonlinearly transformed using the hyperbolic tangent activation function. Simultaneously, the topological vortex invariant is used as a scaling factor to modulate the transformation result. Then, the spatial gradient of the thermal conductivity weight coefficient is logarithmically calculated and multiplied by the modulated eigenvector. Finally, the product result is multiplied by the integral of the temperature change rate to generate a nonlinear feedforward compensation power that reflects the dynamic characteristics of heat conduction. This power value considers both the temperature field dynamic information encoded by the latent variable eigenvector and the topological characteristics of the nanowire array through the topological vortex invariant.

[0073] By combining the deviation field between the target temperature and the real-time temperature and the spatial gradient of the thermal conductivity weighting coefficient, a feedback compensation power is generated through a graph convolutional network.

[0074] The specific process includes: the deviation field between the target temperature and the real-time temperature, and the spatial gradient of the thermal conductivity weight coefficient are input into a graph convolutional network for processing. The graph convolutional network abstracts the topology of the microfluidic chip into graph data, where nodes represent temperature monitoring points, edges represent heat conduction paths, and node features include temperature deviation values ​​and local thermal conductivity weight gradients. The temperature deviation information of neighboring nodes is aggregated through multi-layer graph convolution operations, and each convolution layer adjusts the information transmission intensity by combining the spatial gradient of the thermal conductivity weight coefficient. After transformation by a nonlinear activation function, the final output layer maps the graph node features into feedback compensation power that matches the thermal compensation requirements of each region. The distribution of feedback compensation power reflects the spatial correlation of the global temperature deviation field and maintains physical consistency with the thermal conductivity characteristics of the material through the thermal conductivity weight gradient.

[0075] Variational collaborative optimization is performed on the nonlinear feedforward compensation power and feedback compensation power to generate compensation power commands.

[0076] The specific process includes: nonlinear feedforward compensation power and feedback compensation power are input into a variational co-optimization processor. First, a joint optimization objective function containing both power contributions is established. The objective function sets the weights of the feedforward term reflecting dynamic response characteristics and the weights of the feedback term reflecting steady-state accuracy. The optimal power combination that minimizes the objective function is solved using a variational method. During the optimization process, thermodynamic constraints are applied using the Lagrange multiplier method to ensure that the power allocation conforms to the principle of energy conservation. The mixing ratio of nonlinear feedforward compensation power and feedback compensation power is iteratively adjusted until the convergence condition is met. The final output compensation power command retains the ability of nonlinear feedforward compensation power to suppress rapid disturbances and integrates the correction effect of feedback compensation power on steady-state deviation, forming a comprehensive control signal that takes into account both dynamic response speed and steady-state accuracy.

[0077] The convergence condition is a stopping criterion that is automatically triggered when the variational derivative of the compensated power distribution satisfies the energy conservation boundary condition, and is pre-calibrated through the Lagrange function extremum problem.

[0078] S4: Based on the compensation power command, construct the thermal conductivity frequency domain filter function, perform inverse Fourier transform, and generate array drive command.

[0079] Based on the compensation power command, it is decomposed into a frequency domain power spectrum through three-dimensional fast Fourier transform.

[0080] The specific process includes: the compensation power command is input into a three-dimensional fast Fourier transform processor, which performs grid-based discrete sampling of the spatially distributed compensation power values ​​according to the three-dimensional structure of the microfluidic chip to form a three-dimensional power matrix; then, fast Fourier transform calculations are performed along the three spatial dimensions XYZ to convert the spatial domain power distribution into a frequency domain representation; complex number operations are used in the fast Fourier transform to process the power amplitude and phase information, generating a three-dimensional frequency domain power spectrum containing real and imaginary parts; the three-dimensional frequency domain power spectrum accurately characterizes the distribution characteristics of the compensation power in the spatial frequency domain, where the low-frequency components correspond to large-scale thermal compensation requirements, and the high-frequency components reflect local fine temperature adjustment requirements, providing an input basis for subsequent thermal conduction frequency domain filtering.

[0081] A thermal conductivity frequency domain filter function is constructed based on the spatial gradient of the thermal conductivity weighting coefficient and the topological vortex invariant.

[0082] The specific process includes: the spatial gradient of the thermal conductivity weighting coefficients and the topological vortex invariant are input into the frequency domain filter function generator. First, the spatial gradient of the thermal conductivity weighting coefficients is normalized to obtain the distribution of the thermal conductivity weighting gradient magnitude, while the topological vortex invariant is converted into a frequency domain scaling factor. Then, a basis filter based on a Gaussian kernel function is constructed in the three-dimensional frequency domain space. The normalized thermal conductivity weighting gradient magnitude is multiplied by the scaling factor of the topological vortex invariant to generate a modulation coefficient matrix with spatial frequency selectivity. Finally, the modulation coefficient matrix and the basis filter are subjected to a Hadamard product operation to form a thermal conductivity frequency domain filter function. The thermal conductivity frequency domain filter function has differentiated filtering intensity at different positions in the frequency domain space, and can automatically adjust the transmittance of each frequency band component of the power spectrum according to the change of the thermal conductivity weighting gradient, while maintaining consistency with the topology of the nanowire array through the topological vortex invariant.

[0083] The power spectrum in the frequency domain is convolved with the thermal conductivity filter function in the complex domain to generate the filtered power spectrum.

[0084] The specific process includes feeding the frequency domain power spectrum and the thermal conductivity frequency domain filter function into a complex domain convolution processor. First, the real and imaginary parts of the frequency domain power spectrum are separated, and the thermal conductivity frequency domain filter function is similarly decomposed into real and imaginary parts. In the three-dimensional frequency domain space, the real part of the frequency domain power spectrum and the real part of the thermal conductivity frequency domain filter function are multiplied point by point. At the same time, the same operation is performed on the imaginary part of the frequency domain power spectrum and the imaginary part of the thermal conductivity frequency domain filter function, and the real and imaginary products are cross-combined according to the rules of complex multiplication. The convolution results of each frequency component are integrated by summation operation to generate a filtered power spectrum containing both real and imaginary parts. The filtered power spectrum retains the energy distribution characteristics of the original power spectrum and, based on the selective passability of the thermal conductivity frequency domain filter function, enhances the effective frequency band that matches the spatial gradient of the thermal conductivity weighting coefficient, while suppressing interference frequency components that may cause uneven heat conduction.

[0085] The filtered power spectrum is subjected to inverse Fourier transform to construct a spatial domain driving power field, and the spatial domain driving power field is integrated and averaged according to the edge array partition to generate array driving commands.

[0086] The specific process includes: the filtered power spectrum enters a three-dimensional inverse fast Fourier transform process, and the inverse transformation is carried out synchronously along the three frequency domain dimensions of XYZ. The complex form of the frequency domain power spectrum is gradually reconstructed into a spatial domain signal; the real part of the transformation result is retained to form a spatial domain driving power field, which presents a three-dimensional continuous spatial distribution characteristic; according to the actual physical layout of the edge heating array of the microfluidic chip, the power field is divided into corresponding regions, and the power value in each region is volume averaged to obtain the standard driving power of each array unit; the final output array driving command contains the power parameters required by the edge heating unit, and the array driving command fully retains the thermal conduction optimization characteristics of the power spectrum, while being completely matched with the physical structure of the heating array.

[0087] S5: The array drive command is executed through the heat conduction network to transfer heat, while the substrate strain distribution is monitored. The temperature uniformity is verified according to the strain-temperature mapping relationship, and a steady-state temperature field is established.

[0088] Heat is transferred by executing array drive commands through a thermally conductive network, and the strain distribution of the substrate is monitored simultaneously using dual-mode response characteristics to obtain dual-mode fluorescence response data.

[0089] The specific process includes: array driving commands are executed through heating units in the heat-conducting network; the heat-conducting network converts electrical energy into heat energy according to the commands and transmits it directionally along the topological insulator nanowire array; at the same time, the dual-mode fluorescent material embedded in the substrate generates a spectral response under thermo-mechanical coupling; a high-sensitivity photodetector is used to capture fluorescence wavelength shift and intensity change signals, which contain information on fluorescence peak position shift caused by strain and fluorescence lifetime change caused by temperature, forming dual-mode fluorescence response data reflecting the strain distribution of the substrate; the data acquisition process is strictly synchronized with heat transfer to ensure the time consistency of strain monitoring and thermal field control.

[0090] A strain-temperature correlation reconstruction function is constructed based on dual-mode fluorescence response data to generate a nonlocal temperature field.

[0091] The specific process includes the construction of a strain-temperature correlation reconstruction function based on dual-mode fluorescence response data input. First, two modes of data, fluorescence wavelength shift and lifetime decay, are separated. Wavelength shift reflects the strain distribution characteristics of the substrate, while lifetime decay corresponds to local temperature changes. Through the dual-mode conversion relationship established by calibration experiments, the wavelength shift is mapped to the strain tensor distribution, and the lifetime decay is converted into a temperature change. A coupled field reconstruction algorithm is used to jointly optimize the strain tensor and temperature change, solving for the optimal solution that satisfies the thermoelastic constitutive equation, generating a non-local temperature field that simultaneously contains information from both the strain field and the temperature field. The non-local temperature field not only reflects the temperature value at the directly measured temperature point, but also infers the temperature distribution in the area where no sensors are deployed through the strain-temperature coupling relationship, achieving a complete reconstruction of the temperature field of the entire microfluidic chip.

[0092] Based on the spatial gradient of the nonlocal temperature field and the thermal conductivity weighting coefficient, the topology alignment index is calculated and the verification results are output. The expression is as follows:

[0093] ;

[0094] in, This represents the topological alignment index. This represents the minimum value of regularization. This represents the regularization term in the denominator.

[0095] The specific process includes the following steps: the spatial gradient of the non-local temperature field and the thermal conductivity weight coefficient is incorporated into the topological alignment index calculation process. First, the gradient vector field of the non-local temperature field is extracted, and the spatial gradient vector of the thermal conductivity weight coefficient is normalized. The cosine value of the direction angle between the temperature gradient and the thermal conductivity weight gradient is calculated at the three-dimensional spatial grid points, and the overall alignment degree is quantified by volume integration. The integration result is compared with a preset threshold to generate a verification result. When the integration value exceeds the quantum thermal steady-state judgment threshold, the topological alignment is judged to be up to standard; otherwise, a failure signal is output. This verification result reflects the degree of matching between the heat conduction direction of the nanowire array and the actual temperature gradient distribution, ensuring that the heat flow transfer path and the anisotropic properties of the material maintain the optimal correspondence.

[0096] The quantum thermal steady-state determination threshold is a benchmark parameter predetermined by solving the eigenvalue problem of the time-dependent heat conduction equation based on the quantized thermal conductivity characteristics of topological insulator nanowires and the thermal relaxation time constant of microfluidic chip materials.

[0097] Based on the verification results, a steady-state temperature field is established when the rate of temperature change is below the quantum thermal steady-state threshold and the topological alignment index meets the standard.

[0098] The specific process includes verifying that the temperature change rate is lower than the quantum thermal steady-state threshold and the topological alignment index reaches the standard. At this time, the heat conduction process inside the microfluidic chip enters a state of equilibrium. At this time, the spatiotemporal gradient change of the nonlocal temperature field tends to stabilize, the thermal conductivity weight coefficient and the temperature gradient distribution form an optimal match, and the heat conduction path of the nanowire array is completely coordinated with the heat flow direction. Finally, a steady-state temperature field that does not change with time is formed in the entire micro-reaction cavity. This steady-state temperature field satisfies the thermodynamic equilibrium condition and maintains spatial uniformity, providing a precise and stable temperature environment for subsequent biochemical reactions.

[0099] This embodiment also provides a temperature control system for a microfluidic chip, including: a nano-arrangement module, a fluorescence temperature measurement module, a power compensation module, a frequency domain filtering module, and a strain verification module. The nano-arrangement module is used to embed topological insulator nanowires into a substrate in a directional arrangement during the microfluidic chip manufacturing process, forming a pre-locked array. Simultaneously, under the continuous action of an acoustic-magnetic field, an anisotropic thermally conductive network is generated, and the topological vortex invariant is calculated to generate a thermal conductivity weighting coefficient. The fluorescence temperature measurement module is used to set a fluorescence layer on the surface of the micro-reaction cavity of the anisotropic thermally conductive network and acquire the fluorescence lifetime decay curve in real time, based on the thermal conductivity constraint variational... The model calculates transient temperature values ​​and temperature change rates using a thermal conductivity weighting system. A power compensation module calculates nonlinear feedforward compensation power based on the transient temperature values ​​and temperature change rates, calculates feedback compensation power using the deviation field between the target temperature and the real-time temperature, and generates compensation power commands. A frequency domain filtering module constructs a thermal conductivity frequency domain filtering function based on the compensation power commands, performs inverse Fourier transform, and generates array drive commands. A strain verification module executes array drive commands through the thermal conductivity network to transfer heat, while simultaneously monitoring the substrate strain distribution, verifying temperature uniformity based on the strain-temperature mapping relationship, and establishing a steady-state temperature field.

[0100] This embodiment also provides a computer device applicable to the temperature control method for microfluidic chips, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the temperature control method for microfluidic chips as proposed in the above embodiment.

[0101] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0102] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the temperature control method for a microfluidic chip as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0103] In summary, this invention improves the uniformity and speed of thermal response by constructing a thermal conductivity frequency domain filter function through compensation power commands, and also achieves precise spatial frequency domain control of heat transfer; by executing array drive commands and simultaneously monitoring substrate strain distribution, it improves long-term operational reliability and structural safety while ensuring temperature control accuracy, and realizes physical consistency verification and thermal steady-state confirmation in closed-loop control.

[0104] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A temperature control method of a microfluidic chip, characterized by: The application relates to a microfluidic chip manufacturing method based on topological vortex invariants and a heat conduction weight coefficient. In the microfluidic chip manufacturing process, topological insulator nanowires are embedded into a base body in a directional arrangement mode to form a pre-locked array, meanwhile, an anisotropic heat conduction network is generated under the continuous action of an acoustic-magnetic field, and a topological vortex invariant is calculated to generate a heat conduction weight coefficient. A fluorescent layer is arranged on the surface of a micro-reaction cavity of the anisotropic heat conduction network, and a fluorescent lifetime decay curve is collected in real time, and transient temperature values and temperature change rates are calculated according to a heat conduction constraint variation model and in combination with the heat conduction weight. Nonlinear feedforward compensation power is calculated according to the transient temperature values and the temperature change rates, feedback compensation power is calculated in combination with a deviation field of a target temperature and a real-time temperature, and compensation power instructions are generated. Based on the compensation power instructions, a heat conduction frequency domain filtering function is constructed, and array driving instructions are generated through inverse Fourier transform. The array driving instructions are executed through the heat conduction network to transfer heat, meanwhile, a substrate strain distribution is monitored, temperature uniformity is verified according to a strain-temperature mapping relationship, and a steady-state temperature field is established.

2. The temperature control method of a microfluidic chip according to claim 1, wherein: In the microfluidic chip manufacturing process, topological insulator nanowires are embedded into a base body in a directional arrangement mode to form a pre-locked array, specifically as follows, In the microfluidic chip manufacturing process, topological insulator nanowires and magnetic modifiers are doped to generate a magnetic response composite fluid; The magnetic response composite fluid is injected into a microfluidic chip forming mold, a three-dimensional phase modulation acoustic field and a pulsed gradient magnetic field are synchronously applied, the nanowires are driven to arrange directionally, and the pre-locked array is formed.

3. The temperature control method of a microfluidic chip according to claim 2, wherein: The application relates to a microfluidic chip manufacturing method based on topological vortex invariants and a heat conduction weight coefficient. Based on the pre-locked array, ultraviolet gradient exposure is performed under the continuous action of the acoustic-magnetic field to generate the anisotropic heat conduction network, meanwhile, a rheological-magnetic coupling parameter set is generated by monitoring the magnetic field intensity and the fluid viscosity change; A three-dimensional microstructure of the anisotropic heat conduction network is scanned, and a topological vortex invariant is calculated through a nanowire space configuration; The heat conduction weight coefficient is generated according to the topological vortex invariant and the rheological-magnetic coupling parameter set.

4. The temperature control method of a microfluidic chip according to claim 3, wherein: A fluorescent layer is arranged on the surface of a micro-reaction cavity of the anisotropic heat conduction network, and a fluorescent lifetime decay curve is collected in real time, and transient temperature values and temperature change rates are calculated according to a heat conduction constraint variation model and in combination with the heat conduction weight. A quantum dot fluorescent layer is arranged on the surface of the micro-reaction cavity of the anisotropic heat conduction network, and a fluorescent lifetime decay curve is collected through a single-photon counter; The fluorescent lifetime decay curve is input into a heat conduction constraint variation model, and a temperature field space-time gradient is generated by combining the heat conduction weight coefficient to calculate the transient temperature values and the temperature field change rate. The application relates to a microfluidic chip manufacturing method based on topological vortex invariants and a heat conduction weight coefficient.

5. The temperature control method of a microfluidic chip according to claim 4, wherein: Based on the transient temperature values, the temperature change rates and the heat conduction weight coefficient, a latent variable feature vector is generated through a physical constraint variation autoencoder; The nonlinear feedforward compensation power is calculated according to the latent variable feature vector and the topological vortex invariant. The application relates to a microfluidic chip manufacturing method based on topological vortex invariants and a heat conduction weight coefficient.

6. The temperature control method of a microfluidic chip according to claim 5, wherein: ​ The deviation field of the target temperature and the real-time temperature and the spatial gradient of the heat conduction weight coefficient are combined, and feedback compensation power is generated through a graph convolution network; The nonlinear feedforward compensation power and the feedback compensation power are optimized in a variational collaborative manner to generate compensation power instructions.

7. The temperature control method of a microfluidic chip according to claim 6, wherein: Based on the compensation power instructions, a heat conduction frequency domain filtering function is constructed according to the heat conduction weight coefficient, and the specific steps are as follows, Based on the compensation power instructions, a three-dimensional fast Fourier transform is performed to decompose into a frequency domain power spectrum; The heat conduction frequency domain filtering function is constructed according to the spatial gradient of the heat conduction weight coefficient and the topological vortex invariant.

8. The temperature control method of a microfluidic chip according to claim 7, wherein: The inverse Fourier transform is performed to generate array driving instructions, and the specific steps are as follows, The frequency domain power spectrum is subjected to complex domain convolution operation with the heat conduction filtering function to generate a filtered power spectrum; The filtered power spectrum is subjected to inverse Fourier transform to construct a spatial domain driving power field, and the spatial domain driving power field is integrated and averaged according to the edge array partition to generate array driving instructions.

9. The temperature control method of a microfluidic chip according to claim 8, wherein: The array driving instructions are executed through the heat conduction network to transfer heat, and the substrate strain distribution is monitored, the temperature uniformity is verified according to the strain-temperature mapping relationship, and a steady-state temperature field is established, and the specific steps are as follows, The array driving instructions are executed through the heat conduction network to transfer heat, and the substrate strain distribution is monitored, the temperature uniformity is verified according to the strain-temperature mapping relationship, and a steady-state temperature field is established, and the specific steps are as follows, Based on the double-mode fluorescence response data, a strain-temperature correlation reconstruction function is constructed to generate a non-local temperature field; According to the non-local temperature field and the spatial gradient of the heat conduction weight coefficient, the topological alignment index is calculated and the verification result is outputted; Based on the verification result, when the temperature change rate is lower than the quantum heat steady-state threshold and the topological alignment index meets the standard, a steady-state temperature field is established.

10. A temperature control system of a microfluidic chip, based on the temperature control method of the microfluidic chip according to any one of claims 1 to 9, characterized in that: It comprises a nano arrangement module, a fluorescence temperature measurement module, a power compensation module, a frequency domain filtering module and a strain verification module, The nano arrangement module is used for embedding topological insulator nanowires in a base body in a directional arrangement manner to form a pre-locked array during the microfluidic chip manufacturing process, generating an anisotropic heat conduction network under the continuous action of an acoustic-magnetic field, and calculating a topological vortex invariant to generate a heat conduction weight coefficient; The fluorescence temperature measurement module is used for setting a fluorescence layer on the surface of the micro-reaction cavity of the anisotropic heat conduction network and collecting a fluorescence lifetime decay curve in real time, calculating a transient temperature value and a temperature change rate according to a heat conduction constraint variational model and in combination with the heat conduction weight coefficient; The power compensation module is used for calculating a nonlinear feedforward compensation power according to the transient temperature value and the temperature change rate, calculating a feedback compensation power in combination with the deviation field of the target temperature and the real-time temperature, and generating compensation power instructions; The frequency domain filtering module is used for constructing a heat conduction frequency domain filtering function based on the compensation power instructions, and performing inverse Fourier transform to generate array driving instructions; The strain verification module is used for transferring heat through the array driving instructions executed by the heat conduction network, monitoring the substrate strain distribution, verifying the temperature uniformity according to the strain-temperature mapping relationship, and establishing a steady-state temperature field.

Images