Enhanced artificial ground freezing working medium and optimized operation method and system thereof

Abstract

The application provides an enhanced artificial ground freezing working medium and an optimized operation method and system thereof, constructs a ground and working medium thermophysical parameter database, prepares a freezing working medium on site, inverses the thermophysical parameters of the freezing working medium and the ground in the site in real time through online monitoring of on-site data, establishes a multi-parameter and multi-physical field coupling simulation model based on the monitoring data and the inversely obtained thermophysical parameters, predicts freezing development, obtains a prediction result, optimizes the prediction result through an artificial intelligence algorithm based on the monitoring data, the inversely obtained thermophysical parameters and the prediction result, and intelligently controls the freezing process through an artificial intelligence decision algorithm based on the optimized prediction result. Through the construction of an integrated technical chain of "data collection-working medium preparation-online monitoring-thermophysical inversion-numerical prediction-artificial intelligence optimization control", the dynamic regulation of the thermal conductivity and viscosity of the CaCl2-based freezing working medium under low-temperature conditions is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of freezing medium design, thermophysical property regulation and intelligent operation on site in artificial ground freezing method, and in particular to an enhanced artificial ground freezing medium and an optimization operation method and system thereof. BACKGROUND

[0002] The artificial ground freezing method is widely used in tunnel, shaft, subway station, hydraulic and mine construction due to its strong applicability, small disturbance to ground and other characteristics. However, the actual ground often has high complexity, which is reflected in soil properties, seepage and even local cavities, and therefore there are high requirements for the formation speed, thickness uniformity and steady strength of the freezing wall.

[0003] Currently, CaCl2 or NaCl solution is generally used as the freezing medium in engineering. However, it has the following problems: 1. The thermal conductivity of the freezing medium changes significantly with the decrease of temperature, and the thermal conductivity decreases significantly at low temperature, which reduces the heat transfer efficiency of the freezing wall. 2. The viscosity of the freezing medium increases sharply at low temperature, and the circulating resistance increases accordingly, which leads to high energy consumption and low heat transfer efficiency. 3. The formula of the traditional freezing medium is static, and it is difficult to dynamically adjust the performance of the medium according to the changes of the ground on site. 4. Numerical simulation prediction lags behind, and cannot be used for real-time regulation and control. SUMMARY

[0004] The present application is carried out to solve the above problems, and aims to provide an enhanced artificial ground freezing medium and an optimization operation method and system thereof.

[0005] The application provides an enhanced artificial stratum freezing working medium optimized operation method, which has the following characteristics and comprises the following steps: S1, constructing a stratum and working medium thermophysical property parameter database; S2, preparing a freezing working medium on site; S3, through online monitoring of on-site data, inversing the thermophysical property parameters of the freezing working medium and the stratum in real time; S4, based on the monitoring data and the inversing thermophysical property parameters, establishing a multi-parameter multi-physical field coupling simulation model, predicting freezing development, and obtaining a prediction result; S5, based on the monitoring data, the inversing thermophysical property parameters and the prediction result, optimizing the prediction result through an artificial intelligence algorithm; and S6, based on the optimized prediction result, intelligently regulating and controlling the freezing process through an artificial intelligence decision algorithm, wherein in S1, the method for constructing the stratum and working medium thermophysical property parameter database is as follows: based on the survey data of an engineering site, a three-dimensional stratum model is established, and the stratum thermophysical property parameters are obtained through on-site soil sample experiments to construct a stratum thermophysical property parameter database; the working medium thermophysical property parameter database is constructed through experiments on the freezing working medium; and the parameters in the stratum thermophysical property parameter database and the working medium thermophysical property parameter database are dynamically calibrated and corrected by using on-site monitoring data, in S2, the preparation method of the freezing working medium is as follows: taking a CaCl2 solution as a basic solution, and adding a heat conduction enhancer, a dispersing agent, a low-temperature viscosity reducer and an anti-crystallization stabilizer, and in S3, the method for online monitoring of on-site data is as follows: multiple types of sensors are arranged in the freezing circuit, the freezing pipe and the stratum to monitor the real-time temperature, flow state and concentration information of the freezing working medium.

[0006] In the enhanced artificial stratum freezing working medium optimized operation method provided by the application, the three-dimensional stratum model can also have the following characteristics: in S1, the method for establishing the three-dimensional stratum model is as follows: based on the drilling results, the rock and soil test data and the geological survey report of an engineering site, a three-dimensional stratum model is established, which comprises the distribution of soil layers, water content, an initial temperature field, a permeability coefficient, pore structure and mechanical parameters.

[0007] In the enhanced artificial stratum freezing working medium optimized operation method provided by the application, the method for constructing the stratum thermophysical property parameter database can also have the following characteristics: in S1, the method for constructing the stratum thermophysical property parameter database is as follows: the multi-temperature zone thermal conductivity coefficient, specific heat, freezing-melting curve and phase change latent heat of the on-site soil sample are tested to obtain the stratum thermophysical property parameters under different depths and different ice contents, and the stratum thermophysical property parameter database is constructed; the stratum thermophysical property parameters comprise a thermal conductivity coefficient, a specific heat capacity and a thermal diffusivity coefficient; and the method for constructing the working medium thermophysical property parameter database is as follows: the low-temperature viscosity of the freezing working medium is tested, the thermal conductivity coefficient is tested, and the phase equilibrium characteristics of the freezing working medium solution under different CaCl2 concentrations are tested to construct the working medium thermophysical property parameter database.

[0008] In the enhanced artificial ground freezing working medium optimized operation method provided by the application, the multiple types of sensors can be arranged at the measuring points of the freezing pipe inlet, the freezing pipe outlet and the heat exchange key nodes, and the multiple types of sensors include low-temperature temperature probes, flow meters, pressure transmitters and density and refractive concentration meters.

[0009] In the enhanced artificial ground freezing working medium optimized operation method provided by the application, the method for inversely calculating the thermal physical parameters of the freezing working medium and the ground in the field can be that the ground thermal conductivity and the specific heat parameter are inversely calculated by using a machine learning algorithm based on the field ground temperature data, and the freezing working medium thermal conductivity and the dynamic viscosity are inversely calculated in real time based on the freezing pipe inlet temperature, the freezing pipe outlet temperature and the freezing working medium circulation speed.

[0010] In the enhanced artificial ground freezing working medium optimized operation method provided by the application, the method for establishing the multiple-parameter and multiple-physical-field coupling simulation model in S4 can be that based on the monitoring data and the inversely calculated thermal physical parameters, a numerical simulation software is used to establish a multiple-parameter and multiple-physical-field coupling simulation model of the soil-pipeline-working medium, simulate the heat exchange process, set the boundary conditions, and perform grid division and parameter adjustment to optimize the multiple-parameter and multiple-physical-field coupling simulation model.

[0011] In the enhanced artificial ground freezing working medium optimized operation method provided by the application, the method for optimizing and predicting the results in S5 can be that based on the monitoring data, the inversely calculated thermal physical parameters and the prediction results, an artificial intelligence prediction model is constructed, the dynamic evolution law of the freezing process is learned through a deep time sequence network, the freezing process is predicted, and the prediction results are mutually calibrated and fused with the prediction results in S4 to optimize the prediction results.

[0012] In the enhanced artificial ground freezing working medium optimized operation method provided by the application, the intelligent control method for the freezing process in S6 can be that a reinforcement learning algorithm is introduced, the inversely calculated thermal physical parameters, the monitoring data and the optimized prediction results are collectively taken as environmental inputs, a reward function is set with the freezing rate, the energy consumption level and the freezing wall uniformity as the targets, the algorithm gradually learns the optimal operation strategy, and the related parameters in the freezing process are intelligently controlled through the optimal operation strategy.

[0013] The application also provides an enhanced artificial ground freezing working medium, based on the enhanced artificial ground freezing working medium optimization operation method, having the characteristics that the mass fraction of CaCl2 solution of the freezing working medium is 25%-33%, the thermal conductivity enhancer uses CuO nanoparticles with a particle size of 45-55 nm, the addition ratio is 0.4%-0.6%, and the dispersing agent uses gum arabic, the addition ratio is 0.4%-0.6%.

[0014] The application also provides an enhanced artificial ground freezing working medium optimization operation system, having the characteristics that the system comprises: a database construction module for constructing a ground and working medium thermophysical property database; a freezing working medium configuration module for preparing the freezing working medium on site; a thermophysical property inversion module for inversing the thermophysical properties of the freezing working medium and the ground on site in real time through online monitoring of the on-site data; a prediction result obtaining module for establishing a multi-parameter and multi-physical field coupling simulation model based on the monitoring data and the inversely obtained thermophysical properties, predicting the freezing development, and obtaining the prediction result; a prediction result optimization module for optimizing the prediction result through an artificial intelligence algorithm based on the monitoring data, the inversely obtained thermophysical properties, and the prediction result; and a regulation and control module for intelligently regulating and controlling the freezing process through an artificial intelligence decision algorithm based on the optimized prediction result.

[0015] Compared with the prior art, the application has the following advantages:

[0016] The application adjusts the concentration of CaCl2 solution and the proportion of additives in real time, so that the freezing working medium still has a good thermal conductivity and controllable viscosity in the deep low temperature zone, and the freezing efficiency is significantly improved.

[0017] The application continuously collects multi-source data such as temperature, pressure difference, flow rate, and energy consumption, establishes a dynamic perception system of the freezing wall temperature field and the state of the freezing working medium, reduces blind freezing and missed freezing, and improves the controllability and safety of on-site operation.

[0018] Based on the monitoring data, the thermophysical inversion result, and the simulation prediction, the application automatically adjusts the circulating pump flow rate, the cold brine loop temperature difference, and the working medium formula, so that the entire freezing process always maintains an efficient operation state, significantly reduces energy consumption, and shortens the freezing period.

[0019] Through the closed-loop architecture of "monitoring-inversion-prediction-optimization", the optimization operation method of the application has self-perception, self-learning, and self-optimization capabilities, significantly improves the quality of the freezing wall formation, the operation reliability, and the engineering safety, and has outstanding engineering application and popularization value. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 is a flowchart of the enhanced artificial ground freezing working medium optimization operation method in the embodiment of the application. DETAILED DESCRIPTION

[0021] To make the technical means, creative features, objectives and effects of the present invention easy to understand, the following embodiments, in conjunction with the accompanying drawings, specifically illustrate the enhanced artificial formation freezing medium and its optimized operation method and system of the present invention.

[0022] This embodiment provides an enhanced artificial formation freezing medium and its optimized operation method, including the following steps:

[0023] Figure 1 This is a schematic flowchart of the optimized operation method for enhanced artificial formation freezing working medium in an embodiment of the present invention.

[0024] like Figure 1 As shown, S1: Constructing a database of stratigraphic and working fluid thermal property parameters. Based on the site survey data, a three-dimensional stratigraphic model is established, and stratigraphic thermal property parameters are obtained through field soil sample experiments to construct a database of stratigraphic thermal property parameters. A database of working fluid thermal property parameters is constructed through experiments on the frozen working fluid. Field monitoring data is used to dynamically calibrate and correct the parameters in the stratigraphic and working fluid thermal property parameter databases, specifically as follows:

[0025] Based on the drilling results, geotechnical test data and geological survey report of the engineering site, a three-dimensional stratigraphic model was established, including the distribution of soil layers, water content, initial temperature field, permeability coefficient, pore structure and mechanical parameters.

[0026] Multi-temperature zone tests were conducted on soil samples to measure thermal conductivity, specific heat, freeze-thaw curves, and latent heat of phase change, obtaining formation thermal property parameters at different depths and with varying ice content. A database of formation thermal property parameters was constructed, including thermal conductivity, specific heat capacity, and thermal diffusivity. These parameters were then compared with the temperature gradient distribution monitored in the field to achieve spatiotemporal dynamic correction of the formation thermal property parameters.

[0027] Low-temperature viscosity and thermal conductivity tests were conducted on the freezing working fluid, and the phase equilibrium characteristics of the freezing working fluid solutions at different CaCl2 concentrations were investigated to construct a database of the working fluid's thermophysical parameters. Furthermore, online calibration of the freezing working fluid concentration, crystallization trend, and low-temperature viscosity changes was achieved by using real-time monitoring data of the return water temperature, flow rate, conductivity, and solution density from the on-site pipeline.

[0028] S2: Prepare the freezing working fluid on-site, using CaCl2 solution as the base solution, and add thermal conductivity enhancer, dispersant, low-temperature viscosity reducer, and anti-crystallization stabilizer via an automatic dosing device. Specifically:

[0029] According to the indoor experiment, the CaCl2 solution mass fraction of the freezing working medium is 25%-33%, which is suitable for-25--40℃ freezing environment. The thermal conductivity enhancer uses CuO nanoparticles with a particle size of 50 nm, and the addition ratio is 0.5%. The dispersant uses gum arabic, and the addition ratio is 0.5%. And by using circulating stirring, ultrasonic dispersion and online mixing structure, various additives can be fully dispersed and kept stable for a long time in the high-salt low-temperature environment.

[0030] Specifically, CaCl2 or dilute water can be automatically added according to the real-time feedback of the refractive concentration and the AI control instruction, so as to ensure that the concentration of the freezing working medium remains in the optimal range in dynamic operation. Therefore, the thermal conductivity and viscosity of the freezing working medium can be adjusted at any time according to the formation condition of the freezing wall, so that the freezing working medium has self-adaptive thermal performance and avoids the problems of crystallization and sudden increase of viscosity of the conventional CaCl2 working medium at low temperature.

[0031] S3: A plurality of types of sensors are arranged in the freezing circuit, the freezing pipe and the stratum to monitor the real-time temperature, flow state and concentration information of the freezing working medium on site, and the thermal physical parameters of the freezing working medium and the stratum on site are inversely calculated in real time, specifically:

[0032] The plurality of types of sensors are arranged at the measuring points of the inlet and outlet of the freezing pipe and the key nodes of heat exchange to monitor the freezing hole temperature, the monitoring hole temperature, the inlet and outlet temperatures of the freezing pipe (i.e. the inlet and outlet temperatures of the freezing working medium), the freezing working medium flow and the equipment operation power and the like.

[0033] The plurality of types of sensors include low-temperature temperature probes, flow meters, pressure transmitters, density and refractive concentration meters and the like. The sensors are all designed with low-temperature protection, and can still be stably operated for a long time below-35℃. Through the multi-point measurement of the freezing wall temperature field, the evolution trend of the freezing ring thickness, whether there is local melting hidden danger in the freezing wall and the dynamic response of the stratum to the cold and heat changes can be reflected in real time, so as to provide continuous and reliable on-site data support for the subsequent thermal physical property inversion and intelligent control.

[0034] Through the stratum temperature data on site, the thermal conductivity and specific heat parameters of the stratum are inversely calculated by using machine learning algorithms (such as nonlinear least squares LM offline and ensemble Kalman filter EnKF online). And the thermal conductivity and dynamic viscosity of the freezing working medium are inversely calculated in real time by using the inlet temperature of the freezing pipe, the outlet temperature of the freezing pipe and the circulating speed of the freezing working medium.

[0035] Specifically, by time-synchronizing and noise-filtering the field-monitored temperature, flow rate, pressure and concentration data, transient interference caused by factors such as pump start-stop and flow disturbance is eliminated, and a stable operation data stream is constructed. Subsequently, based on the temperature difference between the inside and outside of the freeze pipe, the circulating flow rate and the energy conservation relationship, the instantaneous thermal conductivity λ of the frozen working medium is solved by a steady-state or quasi-steady-state heat transfer model. At the same time, the dynamic viscosity μ of the working medium is inversely calculated by using the Darcy-Weisbach formula combined with the measured pressure difference. In order to improve the inversion accuracy, the λ and μ are further dynamically corrected by using indoor test determination, historical data regression and correction model, so that they can accurately reflect the real thermophysical property changes of the frozen working medium under the conditions of complex low temperature, high salt and multiple additives.

[0036] S4: Based on the monitoring data and the inverted thermophysical property parameters, a multi-parameter multi-physical field coupled simulation model is established to predict the freeze development and obtain the prediction results, specifically:

[0037] Based on the monitoring data and the inverted thermophysical property parameters, a numerical simulation software is used to construct a real-time updated multi-parameter multi-physical field coupled simulation model (freeze field finite element model) of soil-pipeline-working medium, combined with the freeze pipe arrangement, formation properties and construction conditions, to simulate the heat transfer process and dynamically predict the temperature field, phase change interface, growth thickness and stability of the frozen wall.

[0038] Specifically, the multi-parameter multi-physical field coupled simulation model adopts three-dimensional thermal-hydraulic three-field coupling, and if necessary, factors such as formation seepage, groundwater recharge or local melting can be superimposed, which can simulate the development trend of the frozen wall in the future several hours to several tens of hours.

[0039] The prediction results include the spatial distribution of the frozen circle thickness, the possible weak areas, the change of the freezing speed, the different proportions of the frozen working medium components and the thermal response differences under different circulating conditions of the frozen working medium.

[0040] By feeding the prediction results back to the subsequent artificial intelligence prediction model, early warning and working condition pre-adjustment can be realized, so that the freezing process has foresight and active control ability.

[0041] By setting boundary conditions such as adiabatic boundary of the freeze pipe inlet temperature, and through grid division and parameter adjustment, the multi-parameter multi-physical field coupled simulation model is optimized to improve its accuracy and practicality. The prediction results are compared with the field measured results and the prediction results of the subsequent artificial intelligence prediction model to ensure the accuracy of the three.

[0042] S5: Based on the monitoring data, the inverted thermal physical parameters, and the prediction results, the prediction results are optimized by an artificial intelligence algorithm. Based on the monitoring data, the inverted thermal physical parameters, and the prediction results, an artificial intelligence prediction model is constructed. The dynamic evolution law of the freezing process is learned through a deep time sequence network, the freezing process is predicted, and it is calibrated and fused with the prediction results in S4, the prediction results are optimized, and the specific steps are as follows:

[0043] The coupling relationship between the working medium thermal physical property, the freezing efficiency, and the energy consumption is established by using the artificial intelligence prediction model. The dynamic evolution law of the freezing process is learned through a deep time sequence network (for example, a long short-term memory network LSTM or a gated recurrent network GRU). The future state of the freezing wall growth is evaluated, that is, the freezing process is predicted, and the prediction results of the freezing wall temperature field, the freezing front advancing speed, and the local temperature anomaly area in the future hours to days are generated. The prediction results are calibrated and fused with the prediction results in S4, and the prediction results are optimized.

[0044] The artificial intelligence prediction model also has self-learning ability. It can accumulate running data under different strata, different formulations, and different environmental temperatures in the long-term running process, continuously correct the prediction accuracy, make the whole freezing process have self-adaptive and intelligent control ability, and provide safe and stable protection for freezing construction in complex strata and extremely low temperature environment.

[0045] S6: Based on the optimized prediction results, the freezing process is intelligently controlled by an artificial intelligence decision algorithm, and the specific steps are as follows:

[0046] The reinforcement learning algorithm is introduced, and the optimized prediction results, the inverted thermal physical parameters, and the monitoring data are collectively used as environmental inputs. The reward function is set by taking the freezing rate, the energy consumption level, and the freezing wall uniformity as the target, so that the algorithm gradually learns the optimal running strategy. According to the real-time running state of the equipment, the related parameters in the freezing process are controlled through the optimal running strategy, and the intelligent control of the freezing process is realized. The related parameters include: (1) the frequency and flow of the cold salt water circulating pump; (2) the working medium temperature difference (ΔT) control; (3) the CaCl2 solution concentration and the additive proportion; (4) the refrigeration host start-stop and partial load operation mode.

[0047] Specifically, when freezing unevenness, thermal instability, abnormal increase of energy consumption, and the like are detected, the working medium concentration, the enhancer and viscosity reducer dosage, the circulating pump speed, and the heat exchanger power can be automatically adjusted, so that the whole freezing process always runs in the best thermal efficiency interval.

[0048] The embodiment also provides an enhanced artificial stratum freezing working medium optimized running system, which comprises:

[0049] A database construction module is configured to implement step S1, that is, to construct a stratum and working medium thermal physical parameter database.

[0050] A freezing working medium configuration module is configured to implement step S2, i.e., field preparation of the freezing working medium.

[0051] A thermophysical parameter inversion module is configured to implement step S3, i.e., real-time inversion of the thermophysical parameters of the freezing working medium and the stratum at the site through online monitoring of the site data.

[0052] A prediction result obtaining module is configured to implement step S4, i.e., establishment of a multi-parameter multi-physical field coupling simulation model based on the monitoring data and the inverted thermophysical parameters to predict the freezing development and obtain the prediction result.

[0053] A prediction result optimization module is configured to implement step S5, i.e., optimization of the prediction result through an artificial intelligence algorithm based on the monitoring data, the inverted thermophysical parameters and the prediction result.

[0054] A regulation module is configured to implement step S6, i.e., intelligent regulation of the freezing process through an artificial intelligence decision algorithm based on the optimized prediction result.

[0055] Effects of the embodiment

[0056] The enhanced artificial stratum freezing working medium and the optimized operation method and system thereof have the following beneficial effects:

[0057] 1. The freezing efficiency is significantly improved by real-time adjustment of the concentration of the CaCl2 solution and the proportion of the additive so that the freezing working medium still has a good thermal conductivity and controllable viscosity in the deep low temperature zone.

[0058] 2. The dynamic perception system of the freezing wall temperature field and the freezing working medium state is established by continuous collection of multi-source data such as temperature, pressure difference, flow and energy consumption, thereby reducing blind freezing and missed freezing and improving the controllability and safety of the field operation.

[0059] 3. The entire freezing process is kept in a high-efficiency operation state by automatic adjustment of the circulating pump flow, the cold brine loop temperature difference and the working medium formula based on the monitoring data, the thermophysical inversion result and the simulation prediction, thereby significantly reducing the energy consumption and shortening the freezing period.

[0060] 4. The optimized operation method has self-perception, self-learning and self-optimization capabilities through the closed-loop architecture of "monitoring-inversion-prediction-optimization", thereby significantly improving the freezing wall formation quality, the operation reliability and the engineering safety, and having outstanding engineering application and promotion value.

[0061] Those skilled in the art should understand that this invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to this invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. An optimized operation method for enhanced artificial formation freezing working fluid, characterized in that, Includes the following steps: S1: Construct a database of formation and working fluid thermophysical parameters; S2: On-site preparation of freezing medium; S3: By monitoring the field data online, the thermal properties of the freezing medium and the formation at the field can be retrieved in real time; S4: Based on the monitoring data and the thermophysical parameters obtained by inversion, a multi-parameter multi-physics field coupled simulation model is established to predict the freezing development and obtain the prediction results; S5: Based on the monitoring data, the thermophysical parameters obtained by inversion, and the prediction results, the prediction results are optimized using artificial intelligence algorithms; S6: Based on the optimized prediction results, the freezing process is intelligently controlled through an artificial intelligence decision-making algorithm. Specifically, in S1, the method for constructing the database of stratigraphic and working fluid thermal property parameters is as follows: based on the survey data of the engineering site, a three-dimensional stratigraphic model is established, and stratigraphic thermal property parameters are obtained through field soil sample experiments to construct a stratigraphic thermal property parameter database. A working fluid thermal property parameter database is constructed through experiments on the frozen working fluid. The parameters in the stratigraphic and working fluid thermal property parameter databases are dynamically calibrated and corrected using field monitoring data. In step S2, the freezing medium is prepared by using CaCl2 solution as the base solution and adding a thermal conductivity enhancer, a dispersant, a low-temperature viscosity reducer, and an anti-crystallization stabilizer. In step S3, the method for online monitoring of field data is as follows: multiple types of sensors are deployed in the freezing loop, freezing pipe, and stratum to monitor the real-time temperature, flow state, and concentration information of the freezing working fluid. In S6, the intelligent control method for the freezing process is as follows: a reinforcement learning algorithm is introduced, and the optimized prediction results, the inverted thermophysical parameters and monitoring data are used as environmental inputs. The reward function is set with freezing rate, energy consumption level and freezing wall uniformity as objectives, so that the algorithm can learn the optimal operating strategy step by step. The relevant parameters in the freezing process are intelligently controlled through the optimal operating strategy. The relevant parameters include: CaCl2 solution concentration and additive ratio.

2. The method for optimizing the operation of the enhanced artificial formation freezing working medium according to claim 1, characterized in that: in, In S1, the method for establishing the three-dimensional stratigraphic model is as follows: based on the drilling results, geotechnical test data and geological survey report of the engineering site, a three-dimensional stratigraphic model is established, including the distribution of soil layers, water content, initial temperature field, permeability coefficient, pore structure and mechanical parameters.

3. The optimized operation method for enhanced artificial formation freezing working fluid according to claim 1, characterized in that: in, In S1, the method for constructing the formation thermal property parameter database is as follows: Multi-temperature zone tests are conducted on field soil samples to measure thermal conductivity, specific heat, freeze-thaw curves, and latent heat of phase change, obtaining formation thermal property parameters at different depths and under different ice content conditions. The formation thermal property parameters include: thermal conductivity, specific heat capacity, and thermal diffusivity. The method for constructing the working fluid thermophysical parameter database is as follows: low-temperature viscosity test, thermal conductivity test, and phase equilibrium characteristics of the frozen working fluid solution at different CaCl2 concentrations are performed to construct the working fluid thermophysical parameter database.

4. The optimized operation method for enhanced artificial formation freezing working fluid according to claim 1, characterized in that: in, In S3, multiple types of sensors are installed at the measuring points of the freezing tube inlet, freezing tube outlet, and key heat exchange nodes. These multiple types of sensors include: low temperature probe, flow meter, pressure transmitter, and density and refractometer.

5. The optimized operation method for enhanced artificial formation freezing working fluid according to claim 4, characterized in that: in, In S3, the method for inverting the thermal properties of the freezing medium and the formation at the site is as follows: using the formation temperature data at the site, the formation thermal conductivity and specific heat parameters are inverted using machine learning algorithms, and the thermal conductivity and dynamic viscosity of the freezing medium are inverted in real time using the freezing pipe inlet temperature, freezing pipe outlet temperature and freezing medium circulation speed.

6. The optimized operation method for enhanced artificial formation freezing working fluid according to claim 5, characterized in that: in, In S4, the method for establishing the multi-parameter multi-physics coupling simulation model is as follows: based on the monitoring data and the inverted thermophysical parameters, a multi-parameter multi-physics coupling simulation model between soil, pipe and working fluid is established using numerical simulation software to simulate the heat transfer process. Boundary conditions are set and meshing and parameter adjustment are performed to optimize the multi-parameter multi-physics coupling simulation model.

7. The optimized operation method for enhanced artificial formation freezing working fluid according to claim 1, characterized in that: in, In step S5, the method for optimizing the prediction results is as follows: based on monitoring data, inverted thermophysical parameters and prediction results, an artificial intelligence prediction model is constructed, the dynamic evolution law of the freezing process is learned through a deep time series network, the freezing process is predicted, and the prediction results are mutually calibrated and fused with the prediction results in step S4 to optimize the prediction results.

8. An enhanced artificial formation freezing medium, obtained based on the optimized operation method of the enhanced artificial formation freezing medium according to any one of claims 1-7, characterized in that, The CaCl2 solution of the freezing medium has a mass fraction of 25%~33%, the thermal conductivity enhancer is CuO nanoparticles with a particle size of 45-55 nm, the addition ratio is 0.4%-0.6%, and the dispersant is gum arabic, the addition ratio is 0.4%-0.6%.

9. An enhanced artificial formation freezing working medium optimization operation system, implemented based on any one of the artificial formation freezing working medium optimization operation methods of claims 1-7, characterized in that, include: The database construction module builds a database of formation and working fluid thermophysical parameters. Freezing medium preparation module for on-site preparation of freezing medium; The thermal property parameter inversion module can invert the thermal property parameters of the freezing medium and the formation in real time by monitoring the field data online. The prediction result acquisition module establishes a multi-parameter multi-physics field coupled simulation model based on monitoring data and the inverted thermophysical parameters to predict freezing development and obtain prediction results. The prediction result optimization module optimizes the prediction result using an artificial intelligence algorithm based on the monitoring data, the inverted thermophysical parameters, and the prediction result. The control module, based on the optimized prediction results, intelligently controls the freezing process through an artificial intelligence decision-making algorithm.

Images