Nanofluid-based combined heat and power plant waste heat recovery system

Abstract

The application discloses a kind of power plant cogeneration waste heat recovery systems based on nanofluid.The system includes heat exchanger module, thermoelectric generator module, self-feedback excitation coil assembly, front alternating electromagnetic exciter and intelligent control module.Heat exchange flow passage is circulated with heat transfer medium containing magnetic nanoparticles.The thermoelectric generator module and coil assembly are connected in series to form a thermoelectric-excitation self-feedback loop, generating a non-uniform static magnetic field, which drives the particles to the pipe wall to reduce the local thermal resistance of the boundary layer through magnetic force enrichment.The front alternating electromagnetic exciter outputs high-frequency alternating magnetic field to induce particle spin and oscillation, break the agglomeration structure and flush the wall.The intelligent control module outputs flow control instructions to match the heat flow-flow rate.The application reduces the local thermal resistance without significantly increasing the overall fluid apparent viscosity, improving the thermal energy conversion efficiency of the system.

Description

Technical Field

[0001] This invention relates to the field of power plant waste heat recovery technology, specifically to a waste heat recovery system for power plant cogeneration based on nanofluids. Background Technology

[0002] Cogeneration systems in power plants generate a significant amount of low- to medium-grade waste heat during operation. To improve overall energy efficiency, industrial applications typically employ waste heat recovery systems to recover and utilize this heat energy. With the development of thermal management technology, nanofluids, as a novel heat transfer medium, are increasingly being introduced into waste heat recovery heat exchange networks due to their higher thermal conductivity than conventional base fluids. Furthermore, combined with thermoelectric conversion technology, the system can directly convert the recovered waste heat into electrical energy, providing an effective way for power plants to save energy and reduce consumption.

[0003] In existing industrial practices of waste heat recovery in power plants, to enhance the heat transfer effect of the pipeline network, the common practice is to uniformly disperse a certain concentration of metal or carbon-based high thermal conductivity nanoparticles in a basic heat exchange medium such as water or thermal oil, and then use a circulating water pump to drive this uniformly mixed nanofluid through the heat exchanger pipes. A thermoelectric generator module is directly attached to the outer wall of the heat exchanger. The waste heat carried by the nanofluid is conducted through the heat exchanger pipe wall to the hot end of the thermoelectric generator. The thermoelectric generator uses the temperature difference between the hot and cold ends to generate a Seebeck electromotive force to output electrical energy. In actual operation, the system mainly relies on adjusting the speed of the circulating water pump to change the macroscopic transport velocity of the fluid, thereby coping with the heat load fluctuations under different operating conditions of the power plant.

[0004] However, existing technologies for waste heat recovery using nanofluids combined with thermoelectric generators present a trade-off between heat transfer efficiency and pumping power consumption. In conventional approaches, achieving significant thermal conductivity enhancement often requires increasing the volume fraction of nanoparticles. However, with increasing particle concentration, the apparent viscosity of the fluid increases non-linearly, leading to a significant increase in system pumping power consumption. This presents a trade-off between heat transfer benefits and pumping costs. More critically, in conventional forced convection heat transfer, the fluid velocity near the heat exchanger tube wall approaches zero due to viscous forces, forming a laminar thermal boundary layer with high thermal resistance. In existing technologies, the distribution of nanoparticles within the thermal boundary layer is primarily dominated by passive mechanisms such as thermophoretic force, Brownian diffusion, and gravitational settling. There is currently no systematic solution coupling thermoelectric conversion with magnetophoretic directional enrichment, making it difficult to achieve controllable high-concentration enrichment within the thermal boundary layer. This makes it difficult to effectively overcome the thermal resistance barrier at the laminar sublayer, hindering the transfer of high-temperature heat from the center of the heat exchange channel to the outside of the tube wall. This static thermal boundary layer constraint prevents the thermoelectric generator's hot end from acquiring sufficient local heat flux. Local thermal resistance on the wall leads to insufficient hot-end temperature of the thermoelectric module and a reduced effective temperature difference, ultimately limiting the overall thermal energy conversion efficiency of the system. Furthermore, under long-term high-temperature operation and fluid shearing, existing nanofluids are prone to agglomeration, sedimentation, and fouling on the heat exchanger walls. This not only causes a continuous decline in the fluid's thermal conductivity but also increases local flow resistance, severely impacting the system's long-term stability and maintenance costs. Summary of the Invention

[0005] The technical problem to be solved by this invention is the contradiction between heat transfer efficiency and pumping power consumption in existing power plant waste heat recovery systems, as well as the defects of performance degradation caused by the easy agglomeration and sedimentation of nanoparticles and scaling on the wall surface.

[0006] This invention provides a waste heat recovery system for combined heat and power (CHP) plants based on nanofluids, comprising a heat exchanger module, a thermoelectric generator module arranged outside the heat exchanger module, a self-feedback excitation coil assembly wound around the outer wall of the heat exchanger module, a pre-amplified alternating electromagnetic exciter located upstream of the heat exchanger module inlet, and an intelligent control module connected to the heat exchanger module and the thermoelectric generator module. The heat exchanger module has internal heat exchange channels for the flow of nanofluid heat transfer medium. The heat exchange channels have an elliptical cross-sectional shape and standardized mechanical interfaces for pipeline connection at both ends, with quick-release mechanical connectors externally. The thermoelectric material inside the thermoelectric generator module is a bismuth telluride-based alloy or a cobaltite-based alloy. Its hot end makes direct heat conduction contact with the outer wall of the heat exchanger module, and its cold end is connected to an external cooling medium pipeline. The electrical output terminal of the thermoelectric generator module is physically connected in series with the self-feedback excitation coil assembly, forming a thermoelectric-excitation self-feedback loop. The self-feedback excitation coil assembly is wound with non-equidistant spacing along the axial direction of the heat exchanger module, forming a radially decreasing magnetic field distribution structure within the heat exchange channel.

[0007] During system startup, the power plant's auxiliary power supply provides initial excitation. Once the thermoelectric generator module's output stabilizes, it switches to self-feedback mode. In self-feedback mode, the thermoelectric generator module's output voltage, after being processed by a DC boost circuit, drives the coil. This overcomes the equivalent DC internal resistance of the thermoelectric generator module, the equivalent coil resistance of the self-feedback excitation coil assembly, and the equivalent load resistance of the external downstream electrical equipment, generating an excitation current in the resulting low-power active excitation circuit. As this excitation current flows through the self-feedback excitation coil assembly, it generates a non-uniform static magnetic field with a spatial gradient within the heat exchange channel.

[0008] The nanofluid heat transfer medium contains core-shell magnetic nanoparticles. A non-uniform static magnetic field applies a Kelvin magnetophore force to the core-shell magnetic nanoparticles, causing them to tend towards the heat exchange tube wall. Under the influence of the Kelvin magnetophore force, the core-shell magnetic nanoparticles accumulate towards the inner wall region, establishing a high-volume-fraction particle layer in the wall-attached region. This particle-enriched layer reconstructs the microscopic material distribution of the laminar thermal boundary layer inside the heat exchange channel. Through a nonlinear increase in the local equivalent thermal conductivity, it reduces the local thermal resistance parameter of the wall-attached thermal boundary layer. High-temperature heat at the center of the flow field is conducted to the hot end of the thermoelectric generator module through this thermally reduced thermal boundary layer, maintaining a local heat transfer equilibrium. Since the thickness of the particle-enriched layer is much smaller than the thickness of the laminar sublayer, and the particle shell is made of a low-adhesion metallic thermally conductive material, the local enrichment of particles does not significantly affect the macroscopic flow resistance of the fluid.

[0009] A pre-amplified alternating electromagnetic exciter is fixedly installed on the periphery of the buffer tube section in front of the inlet end. It is connected to an independent AC power supply and generates a high-frequency alternating magnetic field inside the buffer tube section. The direction of the high-frequency alternating magnetic field is perpendicular to the fluid flow direction. When the nanofluid heat transfer medium flows through the buffer tube section, the alternating magnetic field applies an alternating magnetic torque to the magnetic nanoparticles. Under the influence of this torque, the magnetic nanoparticles generate in-situ high-frequency spin and translational oscillations, inducing microscopic turbulence within the fluid and generating localized fluid shear forces (estimated to be on the order of 10^-1 to 10^1 Pa). This localized fluid shear force is greater than the van der Waals forces between particles (on the order of approximately 10^-3 Pa), physically disrupting the particle aggregation structure. The inner wall of the heat exchanger module is physically coated with a fluorosilane-modified composite titanium dioxide-based self-cleaning coating. This coating possesses superhydrophobic physical properties and anti-fouling boundary conditions. The localized fluid shear force induced by the high-frequency spin is transmitted to the inner wall surface with the fluid, creating a high-frequency scouring effect on the self-cleaning coating surface, which, in conjunction with the superhydrophobic properties, inhibits fouling adhesion to the wall surface.

[0010] The nanofluidic heat transfer medium comprises various core-shell magnetic nanoparticle clusters, with the core being a manganese-zinc ferrite material. By adjusting the doping ratio of zinc in the manganese-zinc ferrite material, magnetic particle clusters with different Curie temperatures are configured, with the doping ratio inversely proportional to the Curie temperature parameter. Particle clusters with different Curie temperatures are mixed and dispersed in a base liquid according to their mass ratio. When the local flow field temperature is lower than the Curie temperature of a specific particle cluster, the particles are in a ferromagnetic state and respond to Kelvin magnetophore force; when the flow field temperature is higher than the Curie temperature, the volumetric magnetic susceptibility of the particles decreases and they transition to a paramagnetic state. Different particle clusters achieve selective magnetic response in the spatial temperature field through multi-stage phase transition characteristics.

[0011] The intelligent control module collects temperature, pressure, and flow rate parameters of the nanofluid heat transfer medium through physical sensors, and simultaneously acquires the output voltage and excitation current parameters in the thermoelectric-magnetic self-feedback loop. After normalizing these parameters, the intelligent control module inputs them into a built-in multivariate machine learning model, which outputs a predicted value of the heat conversion efficiency via a neural network mapping structure. The intelligent control module iteratively optimizes the heat conversion efficiency, obtaining the target flow rate parameter and outputting flow control commands to the circulating water pump module. The circulating water pump module adjusts the system flow rate parameters accordingly, achieving dynamic global heat flux-flow rate matching between the system flow rate parameters and the local spatial gradient distribution of the non-uniform static magnetic field.

[0012] At the steady-state convergence node, the intelligent control module extracts the flow margin data of the circulating water pump module. By calculating the ratio of the flow margin to the rated flow requirement parameter of a single heat exchanger module and rounding down, it obtains the maximum number of heat exchanger modules that can be connected in parallel, generating a horizontal expansion plan for the system. Simultaneously, the intelligent control module extracts various operational monitoring data and aligns them with the initial settings to construct a global operational state matrix. The intelligent control module uses the Euclidean distance weighted equation to calculate the system's global performance degradation index. When the global performance degradation index exceeds the safety redundancy threshold, the intelligent control module sorts the internal deviation values ​​to extract error sources. These error sources are input into a pre-defined fault mapping lookup table containing key-value pairs of associated specific performance parameters and corresponding hardware occurrence nodes. This table matches and outputs the physical hardware-level device performance degradation nodes, thereby generating corresponding operational optimization suggestions and operation guidance documents for single-point module plug-in / plug-out replacement.

[0013] The present invention, by adopting the above technical solution, can bring the following beneficial effects:

[0014] 1. This invention constructs a thermoelectric-excitation self-feedback loop by connecting a thermoelectric generator module and a self-feedback excitation coil assembly in series, and generates a non-uniform static magnetic field with a spatial gradient inside the flow channel to apply a magnetophore force to the magnetic particles towards the heat exchange tube wall. This causes the particles to accumulate in the wall region and reconstruct the micro-material distribution law of the laminar thermal boundary layer, thereby reducing the local thermal resistance parameter. This achieves the effect of breaking through the heat transfer limitation without significantly increasing the overall fluid apparent viscosity and pumping power consumption, so as to maintain the local heat transfer balance in the thermoelectric conversion process and improve the system's thermal energy conversion efficiency.

[0015] 2. This invention uses a pre-positioned alternating electromagnetic exciter in the buffer tube section at the inlet of the heat exchanger module to output a high-frequency alternating magnetic field, which forces magnetic nanoparticles to generate spin and oscillation and generate local fluid shear force inside the fluid. This, combined with a titanium dioxide-based self-cleaning coating with superhydrophobic physical properties applied to the inner wall of the heat exchange channel, constructs an anti-fouling boundary. This achieves the effect of using purely physical micro-convective shear force to destroy the particle agglomeration structure and inhibiting the adhesion of wall fouling in conjunction with the coating interface properties, thereby maintaining the dispersion stability of the heat transfer medium during long-term operation and reducing the frequency of equipment cleaning and maintenance.

[0016] 3. This invention prepares a core-shell magnetic nanoparticle cluster with multi-stage phase transition characteristics so that the particles exhibit different magnetization states according to the local flow field temperature response. At the same time, it uses a multivariable machine learning model embedded in the intelligent control module to collect multi-dimensional physical state parameters of the system and execute iterative calculations to output flow control commands. This achieves selective magnetic response of particles with different Curie temperatures under the space temperature field. Combined with real-time optimization of the algorithm, it performs global heat flow-flow velocity dynamic matching to ensure that the system can maintain steady-state convergence operation under drastic fluctuations in external operating conditions. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the overall system architecture of the present invention;

[0018] Figure 2 This is a schematic diagram of the heat exchanger-thermoelectric generator-excitation coil of the present invention;

[0019] Figure 3 This is a diagram illustrating the system control logic and workflow of the present invention.

[0020] Figure 4 This is a diagram showing the working principle and electrical connection of the thermoelectric-excitation self-feedback loop of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Example 1

[0023] See attached document Figure 1-2 The present invention provides a waste heat recovery system for cogeneration in power plants based on nanofluids, comprising: a heat exchanger module 10, a thermoelectric generator module 20 arranged outside the heat exchanger module 10, a self-feedback excitation coil assembly 30 wound around the outer wall of the heat exchanger module 10, a front-mounted alternating electromagnetic exciter 40 disposed upstream of the inlet of the heat exchanger module 10, and an intelligent control module 50 connected to the heat exchanger module 10 and the thermoelectric generator module 20.

[0024] The heat exchanger module 10 has a heat exchange channel inside for the flow of nanofluid heat transfer medium 11. Both ends of the heat exchanger module 10 are equipped with standardized mechanical interfaces. The mechanical interfaces adopt a pipe joint structure, and the size tolerance of the mechanical interfaces is controlled within a preset threshold range. This is used to realize the docking and connection of various fluid pipelines in the system. The heat exchanger module 10 is equipped with quick-release mechanical connectors on the outside, so that the individual heat exchanger module 10 has the characteristics of independent disassembly and replacement, and can realize the instant interchange of heat exchanger modules 10 without changing the physical configuration of the overall system.

[0025] The thermoelectric generator module 20 is tightly attached to the outer wall of the heat exchange tube section of the heat exchanger module 10. The thermoelectric generator module 20 has a hot end and a cold end. The hot end of the thermoelectric generator module 20 is in direct heat conduction contact with the outer wall of the heat exchanger module 10. The cold end of the thermoelectric generator module 20 is connected to the external cooling medium pipeline. The end of the thermoelectric generator module 20 is equipped with a standardized electrical interface. The electrical interface has a fixed contact resistance threshold and current carrying range standard.

[0026] The self-feedback excitation coil assembly 30 is wrapped and arranged on the outer wall area of ​​the heat exchanger module 10 corresponding to the thermoelectric generator module 20 with a preset number of winding turns and axial spacing. The system constructs a thermoelectric-excitation self-feedback loop. The positive and negative terminals of the electrical output of the thermoelectric generator module 20 are physically connected in series with the two terminals of the self-feedback excitation coil assembly 30 through electrical interfaces. The self-feedback excitation coil assembly 30 is wired with an axial non-equal spacing winding structure along the outside of the heat exchanger module 10 so that a magnetic field distribution structure with decreasing magnetic induction intensity is formed in the radial direction from the pipe wall to the center of the pipe of the heat exchanger module 10.

[0027] Furthermore, for the above system, the following process is provided for waste heat recovery:

[0028] Step S1. Nanofluid property calculation: The thermophysical properties of the nanofluid are calculated in detail using a computational model, including but not limited to thermal conductivity, viscosity, and heat capacity. This step involves selecting metal oxides or carbon nanotubes as nanoparticles and water or oil as the base liquid. Combined with the effective medium theory model, it is estimated that under the configuration parameters of 1-5% by mass, the thermal conductivity of the target nanofluid can be increased to 1.2 times that of the traditional base liquid, the viscosity is controlled within the Newtonian fluid range, and the heat capacity is calculated according to the specific heat capacity formula Cp=dQ / dT, where dQ is the heat change and dT is the temperature change.

[0029] Step S2. Heat Exchanger Design Optimization: Computational Fluid Dynamics (CFD) simulation technology is applied to optimize the flow channel design of the heat exchanger. This step includes simulating fluid flow in the Reynolds number (Re) range of 100 to 2000 and heat transfer process in the Nusselt number (Nu) range of 5 to 50 to determine the optimal flow channel geometry, size and layout, thereby improving heat exchange efficiency.

[0030] Step S3. Thermoelectric material performance evaluation: Evaluate the performance of various thermoelectric materials, including their thermoelectric conversion efficiency, stability and durability. This step involves experimental testing and theoretical analysis, selecting materials with a maximum thermoelectric figure of merit (ZT value) greater than 1 at a temperature difference of 50-100°C.

[0031] Step S4. System thermal energy conversion efficiency simulation: Simulate the thermal energy conversion efficiency of the entire waste heat recovery system under different operating conditions. This step includes simulating the flow and heat transfer process of nanofluids in the heat exchanger, as well as the electrical energy output of thermoelectric materials, and predicting that the theoretical thermoelectric conversion efficiency of the system under specific optimized operating conditions with a temperature difference of 50-100°C will reach 5-10%.

[0032] Step S5. Intelligent Algorithm Development: Develop intelligent algorithms for real-time monitoring and adjustment of key parameters of the waste heat recovery system. This step includes designing sensor layout, controller logic, and actuator response strategies to achieve optimal system operation and ensure that the system's control command output response time does not exceed 5 seconds under different operating conditions.

[0033] Step S6. Self-cleaning coating effect prediction: Predict the effect of the self-cleaning coating on the heat exchanger surface, including its effect on reducing fouling and biofilm adhesion. This step involves the selection of coating materials and simulation of surface properties, and evaluates the coating's effect on reducing fouling adhesion to less than 10%.

[0034] Step S7. Modular Design Evaluation: Evaluate the impact of modular design on system flexibility and scalability. This step includes designing standardized interfaces and interchangeable modules, as well as evaluating the impact of these designs on system upgrades and maintenance. Modular design supports rapid replacement and can significantly reduce maintenance time under typical operating conditions.

[0035] Step S8. Economic and Environmental Impact Assessment: Conduct a system economic analysis and environmental impact assessment. This step includes calculating the system's initial investment, operating costs, and expected energy-saving effects, as well as assessing the system's contribution to reducing greenhouse gas emissions and improving energy sustainability. Based on typical operating conditions, the investment payback period is estimated to be competitive.

[0036] Step S9. Optimization suggestion generation: Based on the above calculation and simulation results, optimization suggestions for system design and operation are generated. This step includes specific suggestions for improving heat exchanger design, thermoelectric material selection, intelligent control strategy and modular design to achieve efficient operation and long-term stability of the system.

[0037] Based on the above system structure, the workflow of the waste heat recovery system for combined heat and power plants based on nanofluids includes the following steps:

[0038] Q1: When the nanofluid heat transfer medium 11 inside the heat exchanger module 10 flows through the heat exchange channel, a temperature gradient exists between the hot end and the cold end of the thermoelectric generator module 20, thereby generating a Seebeck thermoelectric electromotive force (the thermoelectric generator module 20 uses a series-parallel array design of multiple thermoelectric units to improve the basic open-circuit voltage). During the system startup phase, the power plant auxiliary power supply is used to provide initial excitation for the self-feedback excitation coil assembly 30. After the temperature difference between the hot and cold ends of the thermoelectric generator module 20 is established and the open-circuit voltage reaches the startup threshold of the DC boost circuit, it switches to self-feedback mode; in the self-feedback... In the feed-through mode, the output voltage of the thermoelectric generator module 20 is branched by the power distribution module: the main circuit, after being processed by a DC boost circuit with a high boost ratio (design parameters of 1:10 to 1:50), drives a coil with a high turns density (N>1000 turns) to generate a central magnetic field strength greater than 10mT. To reduce average power consumption and match the limited output power of the thermoelectric generator module 20, this main circuit is configured in pulse excitation mode or resonant operation mode to ensure the system's energy self-sustaining balance; the auxiliary circuit is connected to a supercapacitor energy storage buffer to provide regulated power to the intelligent control module and external loads. In the low-power active excitation circuit constructed above, the generated excitation current... Satisfying the formula:

[0039]

[0040] in, The open-circuit thermoelectric electromotive force generated by thermoelectric generator module 20; This is the equivalent DC internal resistance of the thermoelectric generator module 20; The equivalent coil resistance of the self-feedback excitation coil assembly 30; The load equivalent resistance of the external subsequent actual electrical equipment connected in series in the thermoelectric-excitation self-feedback loop. In order to ensure the excitation current threshold, the Rload needs to meet the short-circuit excitation constraint that approaches zero, or the load equipment can be connected in parallel bypass form.

[0041] Q2: Excitation current When the flow passes through the self-feedback excitation coil assembly 30, a non-uniform static magnetic field with a spatial gradient is generated inside the flow channel of the heat exchanger module 10. The non-uniform static magnetic field forms a direct electrical feedback coupling with the real-time temperature difference output of the thermoelectric generator module 20, which is used to apply a Kelvin magnetophore force towards the heat exchange tube wall to the magnetic particles inside the nanofluid heat transfer medium 11.

[0042] Q3: The pre-amplified alternating electromagnetic exciter 40 is fixedly installed on the periphery of the buffer tube section in front of the inlet end of the heat exchanger module 10. The pre-amplified alternating electromagnetic exciter 40 is connected to an independent AC power supply outside the system. It generates a high-frequency alternating magnetic field in the internal space of the buffer tube section. The main magnetic flux direction of the high-frequency alternating magnetic field is not parallel to the fluid flow direction of the nanofluid heat transfer medium in the buffer tube section. When the nanofluid heat transfer medium 11 flows through the buffer tube section, the high-frequency alternating magnetic field applies an alternating magnetic torque to the magnetic nanoparticles in the nanofluid heat transfer medium 11, inducing relative slip between the particles and the fluid, thereby generating microscopic shear force. Combined with the high-frequency reversal of the particle magnetic moment, it directly destroys the magnetic dipole interaction and van der Waals force connection between the magnetic nanoparticles.

[0043] Q4: The intelligent control module 50 is connected to the sensor probes fixed on the heat exchanger module 10 and the thermoelectric generator module 20 via a standardized data communication bus. The intelligent control module 50 has an independent data communication protocol for collecting and recording fluid temperature, pipe pressure and thermoelectric electromotive force status data. The intelligent control module 50 is independent of the thermoelectric-excitation self-feedback loop formed by the thermoelectric generator module 20 and the self-feedback excitation coil assembly 30, and monitors and diagnoses the module compatibility and operating status of the system.

[0044] Example 2

[0045] See attached document Figure 3-4 Based on Example 1, the calculation of nanofluid properties specifically includes the following steps:

[0046] Step S101: Prepare and screen core-shell magnetic nanoparticles with multi-stage phase transition characteristics. The core of the core-shell magnetic nanoparticles is made of ferromagnetic material, and the outer shell is made of metallic thermally conductive material. The ferromagnetic material is manganese-zinc ferrite material. By preparing manganese-zinc ferrite particles with different zinc doping ratios, multiple magnetic particle clusters with first, second, and third Curie temperatures are obtained. These clusters are then mixed and dispersed in a base liquid according to a preset mass ratio. The highest-order third Curie temperature parameter is set to be higher than the maximum rated operating temperature of the power plant waste heat recovery system to ensure that even in the highest temperature core heat exchange region, the temperature remains stable. At least one particle cluster maintains ferromagnetic response and magnetophoretic enrichment capability, ensuring thermal resistance reconstruction effect across the entire temperature range and avoiding local thermal resistance rebound in high-temperature areas due to all particles converting to paramagnetic state. The doping ratio of zinc element is set to have a negative correlation with the Curie temperature parameter, and the doping ratio of zinc element is set to have a monotonically decreasing relationship with the Curie temperature parameter within a set mole fraction range. The first particle cluster with the first Curie temperature, the second particle cluster with the second Curie temperature, and the third particle cluster with the third Curie temperature are extracted and mixed and dispersed in the base liquid according to the preset mass ratio parameters to generate nanofluid heat transfer medium 11.

[0047] Step S102: Construct a molecular dynamics simulation model of the nanofluid heat transfer medium 11. Input the density parameters and surface tension parameters of the base liquid into the molecular dynamics simulation model, and set the particle size parameters and initial mass fraction of the core-shell magnetic nanoparticles. Calculate the microscopic interaction potential energy between the core-shell magnetic nanoparticles. The microscopic interaction potential energy includes the van der Waals force attraction potential energy and electrostatic repulsion potential energy between the core-shell magnetic nanoparticles.

[0048] Step S103: Introduce the Lennard-Jones potential function into the molecular dynamics simulation model to numerically solve the van der Waals gravitational potential energy between core-shell magnetic nanoparticles. Based on DLVO theory, the system linearly superimposes the van der Waals gravitational potential energy with the double-layer electrostatic repulsion potential energy to construct the total interaction potential energy equation. The discrete time step iterative solution is performed on the particle aggregation tendency and sedimentation rate of the nanofluid heat transfer medium 11 at multiple temperature nodes to obtain the critical aggregation concentration threshold of the nanofluid heat transfer medium 11.

[0049] Step S104: Establish a magnetic phase transition response determination model for core-shell magnetic nanoparticles. Based on the Curie-Weiss law, define the phase transition response law of the volume magnetic susceptibility of core-shell magnetic nanoparticles as a function of local flow field temperature. When the flow field ambient temperature... Below a specific Curie temperature parameter At this time, core-shell magnetic nanoparticles exhibit a ferromagnetic structure; when the ambient temperature of the flow field... Exceeding a specific Curie temperature parameter At this time, the volume magnetic susceptibility decreases during the phase transition, and the core-shell magnetic nanoparticles are converted into a paramagnetic structure;

[0050] Step S105: Feed back the critical agglomeration concentration threshold of the nanofluid heat transfer medium 11 to the preparation equipment to limit the maximum mass percentage of particles in the preparation process of the nanofluid heat transfer medium 11, obtain the design operating temperature node data of each level of heat exchanger module 10 along the power plant cogeneration waste heat recovery system, align and map the parameter values ​​of the first Curie temperature, the second Curie temperature and the third Curie temperature to the design operating temperature nodes of the heat exchanger module 10 at different levels, and complete the multi-stage phase change preparation of the nanofluid heat transfer medium 11.

[0051] Furthermore, the specific procedures for heat exchanger design optimization and system thermal energy conversion efficiency simulation include:

[0052] Step S201: Establish a three-dimensional geometric model of the heat exchanger module 10, set the cross-sectional shape of the heat exchange channel inside the heat exchanger module 10 to be elliptical, set the width of the heat exchange channel to be 1.5 to 3.5 mm, the depth of the channel to be 0.8 to 1.5 mm, the length of the channel to be 200 to 400 mm, set the torsion angle of the heat exchange channel to be 2 to 5 degrees, and arrange the spacing between adjacent heat exchange channels to be 5 to 10 mm;

[0053] Step S202: Configure the physical property parameters of the nanofluid heat transfer medium 11, including its thermal conductivity, viscosity, and density. Set the fluid boundary conditions for the computational fluid dynamics simulation, including the velocity and temperature input values ​​at the inlet boundary, the pressure output value at the outlet boundary, and the fixed heat flux on the outer wall of the heat exchanger module 10. Introduce the Kelvin magnetophoretic force generated by the spatially non-uniform static magnetic field into the volume force term of the momentum conservation equation. This Kelvin magnetophoretic force is calculated using the following formula:

[0054]

[0055] in, Kelvin magnetophoresis force; The volume of a single magnetic nanoparticle inside the nanofluid heat transfer medium 11; This is the difference in volumetric magnetic susceptibility between the magnetic nanoparticles and the underlying liquid (dimensionless). This formula is applicable when the particle size is much smaller than the scale of magnetic field change and the particle magnetization is in the unsaturated linear response region. When the local magnetic field is extremely high and causes magnetic saturation, the system algorithm will automatically introduce the Langevin function for nonlinear saturation magnetization correction. Permeability of free space; The magnetic flux density vector generated by the self-feedback excitation coil assembly 30; As a spatial gradient operator, the volume force term is numerically correlated with the spatial distribution data of magnetic induction intensity generated by the self-feedback excitation coil component 30 to establish the coupled calculation input conditions of force displacement of magnetic nanoparticles and local velocity distribution change.

[0056] Step S203: Solve the flow and heat transfer process in the heat exchange channel using the continuity equation, momentum conservation equation, and energy conservation equation;

[0057] Step S204: Perform numerical iterative calculations, extract temperature field distribution data, velocity field distribution data and pressure distribution data inside the heat exchanger module 10, obtain fluid flow behavior characteristic data of the heat exchange channel in the Reynolds number range of 200 to 2000, and extract local heat flux boundary data at the outer wall of the heat exchanger module 10 with the thermoelectric generator module 20 attached.

[0058] Step S205: Combining the extracted temperature field distribution data and velocity field distribution data, calculate the total heat transfer coefficient of the heat exchanger module 10 and the pressure drop values ​​at both ends of the heat exchange channel. Use the total heat transfer coefficient and pressure drop values ​​as feedback basis for adjusting the geometric parameters of the heat exchanger module 10. Iteratively adjust the geometric parameters and spatial layout of the heat exchange channel according to the feedback basis to complete the channel design and system heat conversion simulation.

[0059] Furthermore, the procedure for evaluating the performance of thermoelectric materials includes the following steps:

[0060] Step S301: Extract the thermophysical and electrical parameters of the thermoelectric material within the thermoelectric generator module 20. The thermoelectric material uses bismuth telluride-based alloy or cobaltite-based alloy. Calculate the thermoelectric figure of merit parameters of the thermoelectric material, set the threshold value of the figure of merit standard to 1.0, and screen... Thermoelectric materials with values ​​greater than the merit threshold standard are used to construct thermoelectric generator modules 20. Mechanical vibration fatigue tests and temperature cycle load tests are performed on the selected thermoelectric materials to obtain the contact resistance drift rate and structural integrity parameters of the thermoelectric materials.

[0061] Step S302: Within the heat exchange pipes of the heat exchanger module 10, the nanofluid heat transfer medium 11 generates radial particle migration distribution in the non-uniform static magnetic field generated by the self-feedback excitation coil assembly 30. Based on the convection-diffusion equation, the particle mass flux equation of the core-shell magnetic nanoparticles on the radial cross-section of the heat exchange channel is established:

[0062]

[0063] in, is the total mass flux vector of the core-shell magnetic nanoparticles; is the Brownian diffusion coefficient of the core-shell magnetic nanoparticles in the base liquid; This represents the local particle volume fraction. is the magnetostrictive drift velocity vector of the core-shell magnetic nanoparticles under the action of Kelvin magnetophoresis. To simplify the analysis, this equation ignores the thermophoretic force and gravitational sedimentation terms and only considers the Brownian diffusion and magnetophoresis-dominated working conditions.

[0064] Step S303: Combining the Stokes drag balance relationship, the magnetic field disturbance force is substituted into the solution process of the magnetostrictive drift velocity vector. The formula for solving the magnetostrictive drift velocity vector is:

[0065]

[0066] in, The Kelvin magnetophore force vector experienced by the core-shell magnetic nanoparticles; The dynamic viscosity of the base liquid; The kinetic diameter of the core-shell magnetic nanoparticles is given. These nanoparticles accumulate on the inner wall of the heat exchanger module 10, establishing a high-volume-fraction particle layer in the wall-attached region. This reconstructs the microscopic material distribution of the laminar thermal boundary layer inside the heat exchanger module 10. Since the thickness of the particle-rich layer is much smaller than the thickness of the laminar sublayer, and the outer shell uses a low-adhesion metallic thermally conductive material, simulation data shows that under conditions where the particle mass fraction is 1% to 5%, the increase in local viscosity leads to an increase of less than 3% in the overall pipe network flow resistance, and its impact on macroscopic flow resistance is negligible. Based on the interfacial thermal resistance theory of heat transfer, the local thermal resistance parameters of the reconstructed thermal boundary layer are calculated:

[0067]

[0068] in, The local thermal resistance parameter of the wall-attached thermal boundary layer; The effective thickness of the thermal boundary layer; This represents the local equivalent thermal conductivity corresponding to the high volume fraction particle layer.

[0069] Step S304: Local particle volume fraction The increase leads to an increase in the local equivalent thermal conductivity. The thermal resistance increases nonlinearly, thereby reducing the local thermal resistance parameter of the attached thermal boundary layer. The high-temperature heat at the center of the internal flow field of heat exchanger module 10 is conducted through the local thermal resistance parameter. The reduced thermal boundary layer is transferred to the hot end of the thermoelectric generator module 20, forming a steady-state heat flux boundary at the interface between the heat exchanger module 10 and the thermoelectric generator module 20.

[0070] Step S305: Monitor the change in thermoelectric electromotive force output of thermoelectric generator module 20 under steady-state heat flux boundary conditions. An increase in thermoelectric electromotive force synchronously increases the loop current of the series-connected self-feedback excitation coil assembly 30, increasing the spatial gradient value of the non-uniform static magnetic field. When the total mass flux vector... When the calculated result is a zero vector state under the coupling effect of Brownian diffusion and Kelvin magnetophoresis, the system reaches the adaptive equilibrium node of boundary layer dynamic reconstruction. At this time, impedance matching is formed between thermoelectric generator module 20 and heat exchanger module 10 to maintain the local heat transfer equilibrium state of the thermoelectric conversion process.

[0071] Furthermore, the development of intelligent algorithms specifically includes the following steps:

[0072] Step S401: The intelligent control module 50 collects the inlet temperature, outlet temperature, flow channel pressure, and flow velocity parameters of the nanofluid heat transfer medium 11 flowing inside the heat exchanger module 10 through physical sensors. The intelligent control module 50 also collects the output voltage parameters of the thermoelectric generator module 20 and the excitation current parameters of the self-feedback excitation coil assembly 30 in the thermoelectric-excitation self-feedback loop. The intelligent control module 50 performs normalization processing on the above physical state parameters and concatenates the normalized physical state parameters to generate a system state feature vector.

[0073] Step S402: The intelligent control module 50 is embedded with a multivariate machine learning model that has been pre-trained offline based on a large-scale multiphysics dataset generated by computational fluid dynamics (CFD) simulation. The multivariate machine learning model receives the system state feature vector and outputs the predicted value of thermal energy conversion efficiency through a multi-layer neural network mapping structure (including an input feature normalization layer, a multi-layer perceptron hidden layer, and an output layer with a ReLU activation function).

[0074] Step S403: The intelligent control module 50 calculates the actual measured value of the thermal energy conversion efficiency based on the ratio of the actual output power of the system to the input waste heat power. The intelligent control module 50 constructs a specific loss function that includes the mean square error (MSE) and L2 regularization constraint terms.

[0075] (Specifically in the form: Loss = MSE + λ||W||², where λ is the regularization coefficient and W is the weight matrix), and the momentum gradient descent algorithm is used to perform backpropagation operation during the idle period of the non-control cycle to fine-tune the connection weight variables of the model online, constraining the numerical convergence of the loss function;

[0076] Step S404: Within the preset flow rate constraint range, the intelligent control module 50 iteratively optimizes the flow rate parameters in the system state feature vector to obtain the target flow rate parameters that maximize the predicted value of the thermal energy conversion efficiency. The intelligent control module 50 outputs a flow rate control command containing the target flow rate parameters. The flow rate control command is transmitted to the circulating water pump module 60 connected to the heat exchanger module 10. The circulating water pump module 60 adjusts the system flow rate parameters of the nanofluid heat transfer medium 11 injected into the heat exchanger module 10 according to the target flow rate parameters. The thermoelectric-excitation self-feedback loop responds to the temperature field fluctuations in the local space of the heat exchanger module 10 and uses the excitation current of the self-feedback excitation coil assembly 30 to adjust the local magnetic field gradient, thereby performing the spatial redistribution of particle concentration of the nanofluid heat transfer medium 11.

[0077] Step S405: The intelligent control module 50 continuously monitors the loss function value and the excitation current value. When the loss function value is lower than the set loss judgment threshold, and the variance of the excitation current fluctuation of the self-feedback excitation coil assembly 30 is lower than the set variance judgment threshold within a continuously set time period, the intelligent control module 50 determines that the cogeneration waste heat recovery system of the power plant has reached the self-feedback steady-state convergence node. Under the self-feedback steady-state convergence node, the system flow velocity parameter and the local spatial gradient distribution of the non-uniform static magnetic field form a global heat flow-flow velocity dynamic match, and the intelligent control module 50 locks the current flow control command output state. Through the above closed-loop iterative optimization mechanism and momentum gradient optimization constraint, it is mathematically guaranteed that the model parameters can achieve rapid convergence within a 5-second control response period.

[0078] Furthermore, the specific process for predicting the effectiveness of self-cleaning coatings includes the following steps:

[0079] Step S501: Using the sol-gel method or spraying process, a fluorosilane-modified composite titanium dioxide-based self-cleaning coating is physically coated on the inner wall of the heat exchange channel of the heat exchanger module 10. The modified composite titanium dioxide-based self-cleaning coating has superhydrophobic surface physical properties. The surface free energy parameter of the titanium dioxide-based self-cleaning coating is set to be lower than the set surface energy threshold. The contact angle of the droplet on the inner wall of the heat exchange channel is determined according to Young's equation. The apparent contact angle of the titanium dioxide-based self-cleaning coating is configured to be greater than 150 degrees, and the physical boundary conditions for anti-fouling adhesion of the inner wall of the heat exchange channel are established.

[0080] Step S502: The pre-amplifier alternating electromagnetic exciter 40 outputs a high-frequency alternating magnetic field inside the inlet front buffer tube section of the heat exchanger module 10. When the nanofluid heat transfer medium 11 flows through the inlet front buffer tube section, the high-frequency alternating magnetic field applies an alternating electromagnetic torque to the core-shell magnetic nanoparticles with inherent magnetic moments inside the nanofluid heat transfer medium 11. This alternating electromagnetic torque forces the core-shell magnetic nanoparticles to generate in-situ high-frequency spin and translation oscillations in the base liquid.

[0081] Step S503: The high-frequency spin and translational oscillations of the core-shell magnetic nanoparticles induce micro-turbulence inside the nanofluid heat transfer medium 11. The micro-turbulence generates local fluid shear stress inside the nanofluid heat transfer medium 11. Combined with the high-frequency reversal of the particle magnetic moment caused by the alternating magnetic field, it directly breaks the magnetic dipole interaction between particles. Moreover, the fluid shear force (on the order of 10^-1 to 10^1 Pa) generated by the local fluid shear stress on the surface of the core-shell magnetic nanoparticles is greater than the van der Waals attraction between the core-shell magnetic nanoparticles (on the order of approximately 10^-3 Pa). Through the above-mentioned coupled physical mechanism, the agglomeration structure between the core-shell magnetic nanoparticles is destroyed. The local fluid shear stress is transferred to the inner wall of the heat exchange channel of the heat exchanger module 10 with the fluid flow, forming a high-frequency fluid scouring effect on the surface of the titanium dioxide-based self-cleaning coating.

[0082] Step S504: Obtain the temperature cycle parameters, fluid pressure parameters, and operating time parameters of the heat exchanger module 10 under operating conditions. Input the temperature cycle parameters and fluid pressure parameters into the coating aging prediction model to calculate the contact angle decay rate of the titanium dioxide-based self-cleaning coating after a specific operating cycle. The coating aging adopts the empirical prediction model exponential decay equation:

[0083]

[0084] in, For runtime Real-time contact angle of the titanium dioxide-based self-cleaning coating surface; The initial intrinsic contact angle of the titanium dioxide-based self-cleaning coating is given. Let be the contact angle aging rate constant. The temperature cycling parameters and fluid pressure parameters are positively correlated. The local fluid shear stress value generated by micro-turbulence is extracted. The particle adhesion and retention on the surface of the titanium dioxide-based self-cleaning coating under high-frequency fluid scouring is calculated. The contact angle attenuation rate and particle adhesion and retention are comprehensively evaluated. The effective self-cleaning life cycle parameters of the titanium dioxide-based self-cleaning coating are calculated and output.

[0085] Furthermore, the specific process for economic and environmental impact assessment includes the following steps:

[0086] Step S601: The intelligent control module 50 acquires the equipment purchase cost data and installation cost data of the heat exchanger module 10, thermoelectric generator module 20, self-feedback excitation coil assembly 30, front-end alternating electromagnetic exciter 40 and circulating water pump module 60, and sums the equipment purchase cost data and installation cost data to calculate the initial total investment cost parameter of the system. At the same time, the intelligent control module 50 extracts the maintenance depreciation cost data and operating energy consumption cost data within the preset operating cycle, and accumulates them to generate the system operating cost parameter.

[0087] Step S602: The intelligent control module 50 extracts the total power generation data of the thermoelectric generator module 20 within a preset operating cycle, and calculates the economic benefit value of energy recovery based on the set local grid access price parameters. When calculating the system operating cost parameters, the system includes the real-time operating power consumption of the front-end alternating electromagnetic exciter and circulating water pump module in the deduction items to ensure that the total power generation benefit is much greater than the internal energy consumption for maintaining the dynamic magnetic field and fluid circulation. Then, based on the initial total investment cost parameters, system operating cost parameters, and net energy recovery economic benefit value, the intelligent control module 50 establishes a dynamic investment payback period solution equation, and solves the dynamic investment payback period solution equation through numerical iteration, outputting the dynamic investment payback period parameters.

[0088] Step S603: The intelligent control module 50 has a built-in life cycle assessment model. The material list details of each hardware module in the cogeneration waste heat recovery system of the power plant are input into the life cycle assessment model. The life cycle assessment model calls the environmental impact database to match the carbon footprint emission factors of the manufacturing stage corresponding to the material list details. The intelligent control module 50 calculates the total carbon emissions of the power plant cogeneration waste heat recovery system throughout its entire life cycle. The calculation formula is:

[0089]

[0090] in, Total carbon emissions over the entire life cycle; The cumulative carbon emission figures for each hardware module's manufacturing stage; The cumulative carbon emissions generated during system operation due to the power consumption of the pre-amplifier alternating electromagnetic exciter 40 and the circulating water pump module 60 are calculated and listed separately in the text assessment report. This serves as the carbon compensation accounting value for the recycling benefits after the system is dismantled and recycled once it reaches its set service life limit.

[0091] Step S604: The intelligent control module 50 obtains the standard coal conversion factor and the standard coal carbon emission factor constant, multiplies the total power generation data of the cogeneration generator module 20 by the standard coal conversion factor, calculates the equivalent standard coal consumption value of the cogeneration waste heat recovery system of the power plant, and then multiplies the standard coal consumption value by the standard coal carbon emission factor constant to calculate the carbon dioxide emission reduction equivalent value within the preset operating cycle.

[0092] Step S605: The intelligent control module 50 calculates the difference between the carbon dioxide emission reduction equivalent value and the total carbon emissions over the entire life cycle to obtain the net carbon emission reduction environmental gain data of the system. The intelligent control module 50 constructs a two-dimensional evaluation matrix with the dynamic investment payback period parameters and the net carbon emission reduction environmental gain data to generate an economic and environmental comprehensive evaluation index for the cogeneration waste heat recovery system of the power plant. The intelligent control module 50 stores the economic and environmental comprehensive evaluation index in a local database to provide a quantitative basis for the subsequent deployment of the heat exchanger module 10 and the scale expansion of the cogeneration generator module 20.

[0093] Furthermore, the specific steps for generating optimization suggestions include:

[0094] Step S701: The intelligent control module 50 extracts the total heat transfer coefficient of the heat exchanger module 10, the critical agglomeration concentration threshold of the nanofluid heat transfer medium 11, the contact angle attenuation rate of the titanium dioxide-based self-cleaning coating, the contact resistance drift rate of the thermoelectric generator module 20, and the comprehensive economic and environmental evaluation indicators of the power plant cogeneration waste heat recovery system stored in the local database. The extracted operation monitoring data are aligned with the initial setting parameters according to a unified timestamp sequence to construct a global operation status matrix containing multiple performance characteristics.

[0095] Step S702: Based on the global operating state matrix, the intelligent control module 50 uses the Euclidean distance weighted equation to calculate the global performance degradation index of the power plant's cogeneration waste heat recovery system. The specific calculation formula is as follows:

[0096]

[0097] in, The global performance degradation index of the combined heat and power waste heat recovery system in a power plant. The total number of key performance parameters involved in the evaluation; For the first The constant weight coefficients corresponding to each key performance parameter are preset according to the analytic hierarchy process. For the first The actual measured values ​​of each key performance parameter during the current operating cycle; For the first The initial reference setting values ​​of each key performance parameter at the factory state. Before being substituted into the Euclidean distance weighted equation, each key performance parameter has been uniformly processed by the intelligent control module to be dimensionless and normalized.

[0098] Step S703: The intelligent control module 50 compares the global performance degradation index with the set safety redundancy judgment threshold. When the value of the global performance degradation index exceeds the safety redundancy judgment threshold, the intelligent control module 50 sorts the deviation values ​​of each item in the global performance degradation index equation in descending order, extracts the error source item that produces the largest deviation value, and inputs the error source item into the built-in fault mapping lookup table. The fault mapping lookup table has key-value pairs that are associated with specific performance parameters and corresponding hardware occurrence nodes. The device performance degradation node at the physical hardware level is output through the key-value pair matching mechanism.

[0099] Step S704: The intelligent control module 50 generates corresponding operation optimization suggestion text based on the equipment performance degradation nodes. Specifically, when the error source item points to the contact angle attenuation rate of the titanium dioxide-based self-cleaning coating exceeding the set limit, the intelligent control module 50 generates control suggestions to improve the alternating magnetic field frequency parameters of the pre-amplifier alternating electromagnetic exciter 40 and outputs a cleaning and maintenance schedule for the inner wall of the heat exchanger module 10; when the error source item points to the contact resistance drift rate of the thermoelectric generator module 20 exceeding the limit, the intelligent control module 50 extracts the mechanical interface coordinate parameters of the thermoelectric generator module 20 and generates an operation guidance document for single-point module plug-in and replacement.

[0100] Step S705: When the global performance degradation index is lower than the safety redundancy threshold and the economic and environmental comprehensive assessment indicators meet the expected investment return setting, the intelligent control module 50 extracts the flow margin data of the circulating water pump module 60 and the current load rate of the self-feedback excitation coil assembly 30. The intelligent control module 50 divides the flow margin data by the rated flow requirement parameter of a single heat exchanger module 10 and rounds down to calculate the maximum number of heat exchanger modules 10 that can be added in parallel under the current pipeline network conditions. The specific formula is as follows:

[0101]

[0102] in, This is the parameter for the maximum number of parallel connections; This is the flow margin data for the circulating water pump module 60; The rated flow rate requirement parameters for a single heat exchanger module 10; As a function for rounding down, the intelligent control module 50 finally generates a horizontal expansion plan for the waste heat recovery system of the power plant cogeneration based on the maximum number of parallel connections parameter, and converts the operation optimization suggestion text, operation guidance document and horizontal expansion plan into a standard industrial text format and transmits them to the human-machine interaction terminal equipment in the power plant's central control room.

Claims

1. A waste heat recovery system for combined heat and power (CHP) plants based on nanofluids, characterized in that, include: The heat exchanger module (10), the thermoelectric generator module (20) arranged outside the heat exchanger module (10), the self-feedback excitation coil assembly (30) wound around the outer wall of the heat exchanger module (10), the front alternating electromagnetic exciter (40) arranged upstream of the inlet of the heat exchanger module (10), and the intelligent control module (50) connected to the heat exchanger module (10) and the thermoelectric generator module (20); The heat exchanger module (10) has a heat exchange channel inside for the flow of nanofluid heat transfer medium (11); The thermoelectric generator module (20) has a hot end and a cold end. The hot end of the thermoelectric generator module (20) is in direct thermal conduction contact with the outer wall of the heat exchanger module (10), and the cold end of the thermoelectric generator module (20) is connected to an external cooling medium pipeline. The system is constructed with a thermoelectric-excitation self-feedback loop. The positive and negative terminals of the electrical output of the thermoelectric generator module (20) are physically connected in series with the two terminals of the self-feedback excitation coil assembly (30) through an electrical interface. The self-feedback excitation coil assembly (30) is wired along the outside of the heat exchanger module (10) using an axial non-equal spacing winding structure, so that a magnetic field distribution structure with decreasing magnetic induction intensity is formed in the radial direction from the pipe wall to the center of the pipe of the heat exchanger module (10).

2. The waste heat recovery system for combined heat and power plants based on nanofluids according to claim 1, characterized in that, During the system startup phase, the auxiliary power supply of the power plant is used to provide initial excitation for the self-feedback excitation coil assembly (30). After the output of the thermoelectric generator module (20) stabilizes, it switches to the self-feedback mode. In the self-feedback mode, the output voltage of the thermoelectric generator module (20) is processed by a DC boost circuit to drive the coil, generating an excitation current in the low-power active excitation circuit. Satisfying the formula: in, The open-circuit thermoelectric electromotive force generated by the thermoelectric generator module (20); The equivalent DC internal resistance of the thermoelectric generator module (20); The equivalent coil resistance of the self-feedback excitation coil assembly (30); The equivalent load resistance of the external downstream electrical equipment connected in series in the thermoelectric-excitation self-feedback loop is used to ensure the excitation current threshold. The short-circuit excitation constraint must be close to zero, or the load equipment must be connected in parallel bypass mode; The excitation current When flowing through the self-feedback excitation coil assembly (30), a non-uniform static magnetic field with a spatial gradient is generated inside the flow channel of the heat exchanger module (10) to apply a Kelvin magnetophore force towards the heat exchange tube wall to the magnetic particles inside the nanofluid heat transfer medium (11).

3. The waste heat recovery system for combined heat and power plants based on nanofluids according to claim 1, characterized in that, The pre-mounted alternating electromagnetic exciter (40) is fixedly installed on the periphery of the buffer tube section in front of the inlet end of the heat exchanger module (10), and connected to an independent AC power supply outside the system. It generates a high-frequency alternating magnetic field in the space inside the buffer tube section. The main magnetic flux direction of the high-frequency alternating magnetic field is not parallel to the fluid flow direction of the nanofluid heat transfer medium (11) inside the buffer tube section. When the nanofluid heat transfer medium (11) flows through the buffer pipe section, the high-frequency alternating magnetic field, with a frequency range of 1kHz to 50kHz to avoid the skin effect, applies an alternating magnetic torque to the magnetic nanoparticles in the nanofluid heat transfer medium (11), inducing a local fluid shear force on the order of 10^-1 to 10^1 Pa, which is greater than the van der Waals attraction between particles on the order of approximately 10^-3 Pa, thereby physically breaking the van der Waals force connection between the magnetic nanoparticles.

4. The waste heat recovery system for combined heat and power plants based on nanofluids according to claim 1, characterized in that, The nanofluid heat transfer medium (11) comprises core-shell magnetic nanoparticles, wherein the core of the core-shell magnetic nanoparticles is made of manganese-zinc ferrite material and the outer shell is made of metal thermally conductive material. By preparing manganese-zinc ferrite particles with different zinc doping ratios, multiple magnetic particle clusters with first, second, and third Curie temperatures were obtained. These clusters were then mixed and dispersed in a base liquid according to a preset mass ratio. The zinc doping ratio was set to monotonically decrease with respect to the Curie temperature parameter within a set mole fraction range. Multiple particle clusters with different Curie temperatures were extracted and mixed and dispersed in a base liquid according to a preset mass ratio. The highest-order third Curie temperature was set to be higher than the system's maximum rated operating temperature to ensure that the magnetism is maintained and an effective magnetophore enrichment effect is generated in the high-temperature heat exchange core region, so that the particles exhibit different magnetization states according to the local flow field temperature response.

5. The waste heat recovery system for combined heat and power plants based on nanofluids according to claim 1, characterized in that, The inner wall of the heat exchange channel of the heat exchanger module (10) is physically coated with a fluorosilane-modified composite titanium dioxide-based self-cleaning coating. The fluorosilane-modified composite titanium dioxide-based self-cleaning coating has superhydrophobic surface physical properties. The apparent contact angle of the titanium dioxide-based self-cleaning coating is greater than 150 degrees, thus establishing the physical boundary conditions for anti-fouling adhesion of the inner wall of the heat exchange channel. The high-frequency spin and translational oscillation inside the nanofluid heat transfer medium (11) generates local fluid shear stress. The local fluid shear stress is transmitted to the inner wall of the heat exchange channel of the heat exchanger module (10) with the fluid flow, forming a high-frequency fluid scouring effect on the surface of the composite titanium dioxide-based self-cleaning coating.

6. The waste heat recovery system for combined heat and power plants based on nanofluids according to claim 1, characterized in that, The heat exchanger module (10) is equipped with standardized mechanical interfaces at both ends. The mechanical interfaces adopt a pipe joint structure to realize the docking and connection of each fluid pipeline in the system. The heat exchanger module (10) is equipped with quick-release mechanical connectors on the outside, so that a single heat exchanger module (10) has the characteristics of independent disassembly and replacement. The heat exchange channel has an elliptical cross-sectional shape, and the channel width is set to be 1.5 to 3.5 mm, the channel depth to be 0.8 to 1.5 mm, the channel length to be 200 to 400 mm, the channel twist angle to be 2 to 5 degrees, and the spacing between adjacent heat exchange channels is 5 to 10 mm.

7. The waste heat recovery system for combined heat and power plants based on nanofluids according to claim 1, characterized in that, The thermoelectric material in the thermoelectric generator module (20) is bismuth telluride-based alloy material or cobaltite-based alloy material; Under the action of a non-uniform static magnetic field, the core-shell magnetic nanoparticles in the nanofluid heat transfer medium (11) are enriched in the inner wall region of the heat exchanger module (10), and a high volume fraction particle layer is established in the wall region, reconstructing the microscopic material distribution law of the laminar thermal boundary layer inside the heat exchanger module (10); since the thickness of the particle enrichment layer is much smaller than the thickness of the laminar bottom layer, and the outer shell is made of low adhesion metal thermal conductive material, its influence on macroscopic flow resistance can be ignored. By reducing the local thermal resistance parameter of the attached thermal boundary layer through the nonlinear increase of the local equivalent thermal conductivity, the high-temperature heat at the center of the flow field is transferred through the thermal boundary layer to the hot end of the thermoelectric generator module (20).

8. The waste heat recovery system for combined heat and power plants based on nanofluids according to claim 1, characterized in that, The system also includes a circulating water pump module (60); the intelligent control module (50) is connected to sensor probes fixed on the heat exchanger module (10) and the thermoelectric generator module (20) via a standardized data communication bus; The intelligent control module (50) is embedded with a multivariate machine learning model. The intelligent control module (50) is configured to: collect the temperature, pressure and flow rate parameters of the nanofluid heat transfer medium (11), combine the output voltage and excitation current parameters in the thermoelectric-excitation self-feedback loop, perform normalization processing and splice to generate a system state feature vector, output the predicted value of heat energy conversion efficiency through a multi-layer neural network mapping structure; and iteratively optimize the target flow rate parameter with the goal of maximizing heat energy conversion efficiency, output the flow rate control command to the circulating water pump module (60) to adjust the system flow rate parameter, and realize the global heat flow-flow rate dynamic matching between the system flow rate parameter and the local spatial gradient distribution of the non-uniform static magnetic field.

9. The waste heat recovery system for combined heat and power plants based on nanofluids according to claim 8, characterized in that, Under the condition of satisfying the steady-state convergence node, the intelligent control module (50) is further configured as follows: Extract the flow margin data of the circulating water pump module (60) and calculate the maximum number of heat exchanger modules (10) that can be added in parallel under the current pipeline network conditions. The specific formula is as follows: in, The flow margin data of the circulating water pump module (60); The rated flow rate requirement parameters for a single heat exchanger module (10); This is the floor function; Furthermore, the intelligent control module (50) is configured to generate a horizontal scaling scheme for the system based on the maximum number of parallel connections parameter.

10. The waste heat recovery system for combined heat and power plants based on nanofluids according to claim 1, characterized in that, The intelligent control module (50) is configured as follows: Various operational monitoring data were extracted and aligned with the initial set parameters to construct a global operational state matrix containing multiple performance characteristics. The global performance degradation index of the cogeneration waste heat recovery system of the power plant was calculated using the Euclidean distance weighted equation. When the value of the global performance degradation index exceeds the set safety redundancy judgment threshold, the internal deviation values ​​are sorted to extract the error source item that produces the largest deviation value. The error source item is input into the internally preset fault mapping lookup table with key-value pairs of related specific performance parameters and corresponding hardware occurrence nodes. The device performance degradation node at the physical hardware level is output, and the corresponding operation optimization suggestion text and single-point module plug-in and replacement operation guidance document are generated.

Images