Desert Photovoltaic Bracket Pile Foundation Uplift Bearing Capacity Detection and Construction Parameter Optimization Method

Abstract

The present invention provides a method for detecting the uplift bearing capacity of desert photovoltaic support pile foundations and optimizing construction parameters, which relates to the technical field of construction optimization. The method includes dividing the construction sub-areas according to the soil parameter information of the desert target area, conducting test pile detection in each construction sub-area, and establishing the distribution law of the uplift bearing capacity at the pile-soil interface. Taking the pile foundation diameter, pile foundation burial depth, and pile foundation surface treatment method as variables, a multi-objective optimization function including the uplift bearing capacity stability target, the uplift bearing capacity redundancy target, and the construction efficiency target is established, and the weight coefficients of the multi-objective optimization function are dynamically adjusted with the ambient temperature and sand layer migration. The particle swarm algorithm is used to solve the multi-objective optimization function to obtain the optimal construction parameter combination for each construction sub-area. The optimal construction parameter combination is used for pile foundation construction, and piles are selected for checking the uplift bearing capacity, and the check result is fed back to the multi-objective optimization function to dynamically fine-tune the construction parameters.

Description

Technical Field

[0001] The present invention relates to construction optimization technology, and particularly to a method for detecting the uplift bearing capacity of desert photovoltaic support pile foundations and optimizing construction parameters. Background Art

[0002] The construction design of desert photovoltaic support pile foundations mainly selects parameters based on empirical values, without fully considering the differences in different construction areas, resulting in unstable bearing performance of the pile foundations. Existing detection methods use single-point strain measurement or top load detection, which cannot obtain the stress distribution law of the pile foundation along the depth direction and do not consider the influence of soil moisture content changes on the uplift bearing capacity.

[0003] Currently, the determination of pile foundation construction parameters lacks a systematic optimization method and fails to simultaneously consider multiple objectives such as the stability, redundancy, and construction efficiency of the uplift bearing capacity. Especially in the desert environment, the influence of environmental temperature changes and sand layer migration on the pile foundation performance is relatively large, but the existing technology lacks a dynamic adjustment mechanism for these environmental factors.

[0004] During the construction process, fixed construction parameters are generally used, and an effective parameter feedback and adjustment mechanism is not established. Even if it is found that the uplift bearing capacity does not meet the requirements, it is difficult to optimize and adjust the construction parameters specifically, which easily causes a large amount of rework or over-design, affecting the project quality and construction efficiency. Summary of the Invention

[0005] The embodiments of the present invention provide a method for detecting the uplift bearing capacity of desert photovoltaic support pile foundations and optimizing construction parameters, which can solve the problems in the existing technology.

[0006] In the first aspect of the embodiments of the present invention,

[0007] A method for detecting the uplift bearing capacity of desert photovoltaic support pile foundations and optimizing construction parameters is provided, including:

[0008] Dividing construction sub-areas according to the soil parameter information of the desert target area, and conducting test pile detection in each construction sub-area, including: arranging multiple groups of strain sensors along the depth direction of the test pile, performing layered loading detection on each test pile, and obtaining strain data at each depth; combining the soil moisture content data collected in real time during the detection process, calculating the contribution values of soil layers to the uplift bearing capacity of the pile foundation under different depths and different moisture content conditions, and establishing the distribution law of the uplift bearing capacity at the pile-soil interface;

[0009] Based on the above distribution law, taking the pile foundation diameter, pile foundation burial depth, and pile foundation surface treatment method as variables, a multi-objective optimization function including the anti-pulling bearing capacity stability target, anti-pulling bearing capacity redundancy target, and construction efficiency target is established. The weight coefficients of the multi-objective optimization function are dynamically adjusted according to the environmental temperature and sand layer migration; the particle swarm algorithm is used to solve the multi-objective optimization function to obtain the optimal construction parameter combination for each construction sub-region;

[0010] The pile foundation construction is carried out using the optimal construction parameter combination; during the construction process, some pile foundations are selected for the anti-pulling bearing capacity review, and the review results are fed back to the multi-objective optimization function to dynamically fine-tune the construction parameters.

[0011] In an alternative embodiment,

[0012] Establishing the distribution law of the anti-pulling bearing capacity of the pile-soil interface based on sensor data includes:

[0013] The strain data is obtained by hierarchical loading detection, including gradually increasing the load in percentage of the designed ultimate load, and continuously loading each level of load until the strain change rate measured by the strain sensor is less than the first preset threshold; based on the strain data, the axial force of each depth section of the pile body is calculated, and the shear stress of the pile-soil interface is calculated according to the axial force difference between adjacent depth sections;

[0014] The functional relationship between the shear stress of the pile-soil interface, depth, and water content is established. The functional relationship includes a depth influence function, a water content influence function, and a coupling correction coefficient. The depth influence function uses a power function form to characterize the influence law of depth on the interface shear stress. The water content influence function uses a piecewise linear function, and the coupling correction coefficient characterizes the interaction between depth and water content;

[0015] The contribution rate of the soil layer per unit depth to the total anti-pulling bearing capacity of the pile foundation is calculated using the functional relationship. The contribution rate is expressed as the percentage of the product of the shear stress of the pile-soil interface at this depth and the pile circumference area to the total anti-pulling force at the pile top; the pile foundation burial depth range is divided into a shallow layer section, a middle layer section, and a deep layer section, and the cumulative contribution values under different water content conditions in each depth section are calculated to determine the distribution law of the anti-pulling bearing capacity of the pile-soil interface.

[0016] In an alternative embodiment,

[0017] Establishing the functional relationship between the shear stress of the pile-soil interface, depth, and water content includes:

[0018] The depth influence function is constructed in the form of a power function. Taking the shear stress of the pile-soil interface at the ground surface as the reference shear stress, the ratio of the interface shear stress at different depths to the reference shear stress is fitted by the least squares method to determine the initial exponent of the power function; the final exponent of the power function is determined by the linear combination of the initial exponent of the power function, the coefficient of earth pressure at rest, and the relative density;

[0019] The moisture content influence function is constructed as a piecewise linear function, with the in-situ moisture content of the target area before construction as the demarcation point. When the moisture content is less than the in-situ moisture content, the first slope is adopted, and when the moisture content is greater than or equal to the in-situ moisture content, the second slope is adopted, respectively representing the influence of different moisture content intervals on the interface shear stress;

[0020] The product of the depth exponent and the moisture content deviation is constructed as the basic coupling term, the product of the depth power term and the moisture content is constructed as the dynamic coupling term, and the product of the depth demarcation function and the moisture content is constructed as the interlayer coupling term. The depth demarcation function takes the value of 1 at the mutation of soil layer properties and 0 at other positions. The basic coupling term, dynamic coupling term and interlayer coupling term are superimposed to form the coupling correction coefficient.

[0021] In an alternative embodiment,

[0022] Establishing a multi-objective optimization function and solving to obtain the optimal construction parameter combination includes:

[0023] The stability objective of the uplift bearing capacity is determined by the dispersion degree of the uplift bearing capacity under different depths and different moisture contents. The surplus degree objective of the uplift bearing capacity is determined by the deviation between the actual uplift bearing capacity and the required uplift bearing capacity. The construction efficiency objective is determined by the ratio of the pile foundation construction time to the reference construction time;

[0024] Based on the environmental temperature and sand layer migration, the weight coefficients of the multi-objective optimization function are dynamically adjusted. The environmental temperature is described by an annual cycle sine function. The sand layer migration determines the adjustment value of the weight coefficient through the coupling effect of the wind-induced migration rate and the migration duration, combined with the non-uniformity of the influence of the wind direction angle on the pile-soil interface;

[0025] The particle swarm algorithm is used to solve the multi-objective optimization function. The particle swarm algorithm is improved by dynamically adjusting the inertia weight and learning factor through the environmental temperature and sand layer migration. The particle positions including the pile foundation diameter, pile foundation burial depth and surface treatment method are iteratively updated to obtain the optimal construction parameter combination that meets the uplift bearing capacity constraint and geometric constraint conditions.

[0026] In an alternative embodiment,

[0027] Dynamically adjusting the weight coefficients of the multi-objective optimization function based on the environmental temperature and sand layer migration includes:

[0028] Establish an environmental temperature fluctuation equation, which describes the desert environmental temperature change using an annual cycle sine function. Calculate the pile-soil interface contact stress based on the environmental temperature fluctuation equation, and use the ratio of the pile-soil interface contact stress to the reference contact stress as the temperature influence coefficient. Establish a temperature weight correction coefficient using a sine function containing a first harmonic term and a second harmonic term;

[0029] Establish a wind-induced migration rate calculation equation, and determine the instantaneous migration rate using the wind force magnitude and wind direction angle; take the integral value of the instantaneous migration rate along the migration duration as the soil stress change amount; based on the soil stress change amount, establish a migration influence coefficient by combining the non-uniform influence of the wind direction angle on the soil around the pile;

[0030] Multiply the temperature weight correction coefficient by the migration influence coefficient and introduce a coupling enhancement term based on the soil stress change amount to establish an environmental coupling coefficient. Dynamically adjust the weight coefficients of the multi-objective optimization function based on the environmental coupling coefficient, where the weight coefficient of the anti-pulling bearing capacity stability target is positively correlated with the ratio of the pile-soil interface contact stress, the weight coefficient of the anti-pulling bearing capacity redundancy target is positively correlated with the soil stress change amount, and the weight coefficient of the construction efficiency target is negatively correlated with the environmental coupling coefficient; perform normalization processing on the weight coefficients.

[0031] In an alternative embodiment,

[0032] Use the particle swarm algorithm to solve the multi-objective optimization function, and obtain the optimal construction parameter combination including:

[0033] Establish a dynamic inertia weight calculation formula based on the environmental coupling coefficient. The dynamic inertia weight calculation formula adopts the exponential difference form of the maximum inertia weight and the minimum inertia weight, and adjusts the change rate of the inertia weight through the environmental coupling coefficient;

[0034] Establish calculation formulas for the individual optimal solution learning factor and the global optimal solution learning factor in the form of exponential functions, and perform adaptive adjustment through the ratio of the difference between the current particle fitness and the maximum fitness and the minimum fitness of the population;

[0035] Construct a default penalty function, which includes an anti-pulling bearing capacity constraint term and a geometric constraint term. The anti-pulling bearing capacity constraint term adopts the square form of the difference between the actual anti-pulling bearing capacity and the required anti-pulling bearing capacity, and the geometric constraint term adopts the square form of the difference between the pile foundation diameter and the pile foundation burial depth and their constraint boundaries;

[0036] Encode the particles including the pile foundation diameter, pile foundation burial depth, and surface treatment method, iteratively update the particle swarm using a dynamic inertia weight and an adaptive learning factor, combine the default penalty function and the multi-objective optimization function to calculate the particle fitness, and iteratively optimize to obtain the optimal construction parameter combination that meets the constraint conditions.

[0037] In an alternative embodiment,

[0038] Performing pile foundation construction using the optimal construction parameter combination and performing dynamic fine-tuning includes:

[0039] Perform pile foundation construction in each construction sub-area according to the optimal construction parameter combination, and perform uplift bearing capacity detection after every preset number of pile foundation constructions to obtain the actual uplift bearing capacity data;

[0040] Use time series analysis method to process the actual uplift bearing capacity data at multiple detection time points in the same construction sub-area to obtain the time-varying characteristics of the uplift bearing capacity;

[0041] Compare the actual uplift bearing capacity data with the theoretical calculated value. When the deviation rate of the uplift bearing capacity exceeds the second preset threshold, correct the weight coefficient in the multi-objective optimization function based on the time-varying characteristics, and re-optimize using the corrected multi-objective optimization function to obtain the updated optimal construction parameter combination;

[0042] During the construction process, real-time monitor the environmental temperature change and the sand layer migration situation. When the change exceeds the preset range, trigger the dynamic adjustment of the multi-objective optimization function.

[0043] In the second aspect of the embodiments of the present invention,

[0044] Provide an electronic device, including:

[0045] A processor;

[0046] A memory for storing instructions executable by the processor;

[0047] Wherein, the processor is configured to call the instructions stored in the memory to execute the method described above.

[0048] In the third aspect of the embodiments of the present invention,

[0049] Provide a computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the method described above is implemented.

[0050] In the present invention, multiple groups of strain sensors are arranged along the depth direction of the test pile, and the distribution law of the uplift bearing capacity of the pile-soil interface is established by combining the soil moisture content data, comprehensively reflecting the contribution characteristics of the soil layer to the uplift bearing capacity of the pile foundation under different depths and different moisture contents. By using the depth influence function, the moisture content influence function and their coupling correction coefficients, the distribution characteristics of the shear stress at the pile-soil interface are accurately characterized, and the prediction accuracy of the bearing performance of the pile foundation is improved.

[0051] Based on the distribution law, the present invention establishes a multi-objective optimization function including the stability, redundancy and construction efficiency of the uplift bearing capacity, dynamically adjusts the weight coefficient through the environmental temperature and sand layer migration, and realizes the intelligent optimization of the construction parameters; the multi-objective optimization function is solved based on the improved particle swarm algorithm, and the convergence performance of the algorithm is improved by using the dynamic inertia weight and the adaptive learning factor, and the optimal construction parameter combination satisfying multiple constraint conditions is obtained.

[0052] During the construction process, the present invention establishes a dynamic fine-tuning mechanism for the construction parameters through the uplift bearing capacity review and time series analysis, effectively coping with the influence brought by environmental changes. The whole method realizes the closed-loop optimization of the pile foundation construction parameters, significantly improves the stability of the uplift bearing capacity of the pile foundation, makes the redundancy control more reasonable, and at the same time ensures a high construction efficiency, providing effective technical support for the design and construction of the desert photovoltaic support pile foundation. Brief Description of the Drawings

[0053] Figure 1 It is a schematic flow chart of the method for detecting the uplift bearing capacity and optimizing the construction parameters of the desert photovoltaic support pile foundation in the embodiment of the present invention;

[0054] Figure 2 It is a comparison chart of the prediction accuracy of the depth-moisture content coupling effect;

[0055] Figure 3 Flow chart of the dynamic adjustment of the weight coefficient by environmental factors. Detailed Embodiments

[0056] To make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0057] The technical solutions of the present invention will be described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0058] Figure 1 This is a schematic flow chart of the method for detecting the uplift bearing capacity of the pile foundation of the desert photovoltaic support and optimizing the construction parameters in the embodiments of the present invention. As Figure 1 shown, the method includes:

[0059] Dividing the construction sub-areas according to the soil parameter information of the desert target area, and conducting test pile detection in each construction sub-area, including: arranging multiple groups of strain sensors along the depth direction of the test pile, conducting layered loading detection on each test pile, and obtaining the strain data at each depth; combining the soil moisture content data collected in real time during the detection process, calculating the contribution value of the soil layer to the uplift bearing capacity of the pile foundation under different depths and different moisture contents, and establishing the distribution law of the uplift bearing capacity at the pile-soil interface;

[0060] Based on the distribution law, taking the pile foundation diameter, pile foundation burial depth, and pile foundation surface treatment method as variables, establishing a multi-objective optimization function including the uplift bearing capacity stability target, uplift bearing capacity redundancy target, and construction efficiency target, and the weight coefficients of the multi-objective optimization function are dynamically adjusted with the ambient temperature and sand layer migration; using the particle swarm algorithm to solve the multi-objective optimization function to obtain the optimal construction parameter combination for each construction sub-area;

[0061] Using the optimal construction parameter combination for pile foundation construction; during the construction process, selecting pile foundations for uplift bearing capacity review, and feeding back the review results to the multi-objective optimization function to dynamically fine-tune the construction parameters.

[0062] In an alternative embodiment,

[0063] Establishing the distribution law of the uplift bearing capacity at the pile-soil interface based on sensor data includes:

[0064] Obtaining strain data through staged loading detection includes gradually increasing the load in percentage of the design ultimate load, and continuously loading each stage of load until the strain change rate measured by the strain sensor is less than the first preset threshold; calculating the axial force of each depth section of the pile body based on the strain data, and calculating the shear stress at the pile-soil interface according to the axial force difference between adjacent depth sections;

[0065] Establishing the functional relationship between the shear stress at the pile-soil interface and the depth and moisture content, the functional relationship includes a depth influence function, a moisture content influence function, and a coupling correction coefficient, the depth influence function uses a power function form to characterize the influence law of depth on the interface shear stress, the moisture content influence function uses a piecewise linear function, and the coupling correction coefficient characterizes the interaction between depth and moisture content;

[0066] Calculate the contribution rate of the soil layer per unit depth to the total uplift bearing capacity of the pile foundation using the said functional relationship. The contribution rate is expressed as the percentage of the product of the shear stress at the pile-soil interface at this depth and the pile circumference area to the total uplift force at the pile top. Divide the pile foundation burial depth range into shallow layer section, middle layer section and deep layer section, calculate the cumulative contribution values at different depth sections under different moisture content conditions, and determine the distribution law of the uplift bearing capacity at the pile-soil interface.

[0067] Exemplarily, according to the soil parameter information of the desert area (such as terrain slope, moisture content and particle size distribution), use the K-means clustering algorithm to divide the site into different construction sub-areas, and arrange multiple groups of strain sensors along the depth direction of the test pile. When performing hierarchical loading tests to obtain strain data, connect the top of the test pile to the hydraulic loading system. The designed ultimate load is 20 kN, and the loading increases step by step according to the percentage of the designed ultimate load, which are 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% and 100% respectively, that is, 2 kN, 4 kN, 6 kN, 8 kN, 10 kN, 12 kN, 14 kN, 16 kN, 18 kN and 20 kN respectively. Each level of load is continuously loaded, and the readings of the strain sensors at each depth are recorded. When the strain change rate is less than the first preset threshold of 0.01% / min, it is considered that the strain under this level of load has stabilized, and the next level of load is loaded. At the same time, arrange soil moisture sensors at different depths around the test pile, and install them at depths of 0.5 m, 1.5 m and 2.5 m respectively to collect the moisture content data of the soil layers at different depths in real time. During the test, record the environmental temperature, wind force and the changes on the sand layer surface.

[0068] Calculate the axial force of each depth section of the pile body based on the strain data. The strain values measured by the strain sensors are converted into the axial force at the corresponding depth by multiplying the elastic modulus of steel and the cross-sectional area of the pile body according to Hooke's law. For example, at a depth of 1.2 m, the four strain values measured by the strain sensors are 120 με, 116 με, 124 με and 118 με respectively, and the average value is 119.5 με. The elastic modulus of steel is 206 GPa, and the cross-sectional area of the pile body is 3699 mm². The axial force at this depth is calculated to be 9.1 kN. Calculate the shear stress at the pile-soil interface according to the difference in axial force between adjacent depth sections. Take the pile section between depths of 0.9 m and 1.2 m as an example. The axial force at a depth of 0.9 m is 11.3 kN, and the axial force at a depth of 1.2 m is 9.1 kN. Then the axial force difference on this pile section is 2.2 kN. The length of this pile section is 0.3 m, and the pile circumference is 628 mm. The average shear stress at the pile-soil interface of this pile section is calculated to be 11.7 kPa. Calculate the shear stress at the pile-soil interface at each depth according to this method.

[0069] Establish the functional relationship between the shear stress at the pile-soil interface, depth, and water content. Take the normalized depth (the ratio of the actual depth to the pile length) as the abscissa and the shear stress as the ordinate to plot a scatter diagram. Determine the parameters of the depth influence function through data fitting. In this example, the depth influence function adopts the power function form, and the fitted parameters are a = 1.35, b = 0.45, and c = 0.2.

[0070] Repeat the above tests under different water content conditions to obtain the relationship data between water content and shear stress. Determine the parameters of the water content influence function through comparative analysis. When the water content is lower than the in-situ value, the shear stress increases significantly with the increase of water content, and the growth slope is 15; when the water content is higher than the in-situ value, the shear stress decreases with the increase of water content, and the reduction slope is -8. Therefore, the water content influence function adopts a piecewise linear function with an obvious inflection point at the in-situ water content. The interaction between depth and water content is characterized by a coupling correction coefficient.

[0071] Use the established functional relationship to calculate the contribution rate of the soil layer per unit depth to the total uplift bearing capacity of the pile foundation. Take the case of a pile length of 3m and a water content of 3% as an example. Multiply the micro-element area around the pile by the shear stress at the corresponding depth and then divide by the total uplift force at the pile top. During specific calculations, the pile length can be divided into multiple micro-segments, each with a length of 0.1m. Calculate the contribution of each segment to the total uplift force as the contribution rate at that depth. For example, at a depth of 1.2m, the shear stress calculated through the functional relationship is 12.5 kPa, the pile circumference at this point is 628 mm, the corresponding micro-element length is 0.1m, the uplift force provided by this micro-element is 0.785 kN, and the total uplift force at the pile top is 20 kN. Then the contribution rate at this depth is 3.9%. Calculate the contribution rates at each depth according to this method and draw the curve of the contribution rate changing with depth.

[0072] The depth range of the pile foundation is divided into a shallow layer section (0 - 0.3 times the pile length), a middle layer section (0.3 - 0.7 times the pile length), and a deep layer section (0.7 - 1.0 times the pile length). Taking a 3 - m long pile as an example, the shallow layer section is 0 - 0.9 m, the middle layer section is 0.9 - 2.1 m, and the deep layer section is 2.1 - 3.0 m. Calculate the cumulative contribution values of each depth section under different water content conditions. For example, under the condition of a water content of 3%, the contribution rate of the shallow layer section is 32%, the contribution rate of the middle layer section is 46%, and the contribution rate of the deep layer section is 22%. Further analyze the contribution characteristics of different depth sections to the uplift bearing capacity. By introducing two characteristic parameters, the middle - shallow layer contribution ratio (the ratio of the contribution rate of the middle layer section to the contribution rate of the shallow layer section) and the deep - middle layer contribution ratio (the ratio of the contribution rate of the deep layer section to the contribution rate of the middle layer section), the distribution characteristics of the uplift bearing capacity of the pile foundation are characterized. For example, when the water content is 3%, the middle - shallow layer contribution ratio is 1.44, and the deep - middle layer contribution ratio is 0.48. By comparing the distribution laws under different water contents, different pile diameters, and different surface treatment methods, a complete database of the distribution law of the uplift bearing capacity of the pile - soil interface can be established, providing a basis for the optimization of subsequent construction parameters.

[0073] The present invention obtains the distribution law of the uplift bearing capacity of the pile - soil interface along the depth through sensor data, establishes a depth influence function, a water content influence function, and their coupling mechanism, realizes the quantitative characterization and sectional analysis of the contribution rate; overcomes the limitations of traditional empirical formulas, closely combines design parameters with actual mechanical behaviors, significantly improves the accuracy and reliability of pile foundation design, and provides data support for the optimized design of desert photovoltaic supports.

[0074] Establishing the functional relationship between the shear stress of the pile - soil interface and depth and water content includes:

[0075] Construct the depth influence function in the form of a power function. Taking the shear stress of the pile - soil interface at the ground surface as the reference shear stress, fit the ratio of the interface shear stress at different depths to the reference shear stress by the least - squares method to determine the initial exponent of the power function; determine the final exponent of the power function from the linear combination of the initial exponent of the power function, the coefficient of earth pressure at rest, and the relative density;

[0076] Construct the water content influence function as a piece - wise linear function. Taking the in - situ water content of the target area before construction as the demarcation point, use the first slope when the water content is less than the in - situ water content, and use the second slope when the water content is greater than or equal to the in - situ water content to respectively characterize the influence of different water content intervals on the interface shear stress;

[0077] Construct the product of the depth exponent and the moisture content deviation as the basic coupling term, construct the product of the depth power term and the moisture content as the dynamic coupling term, and construct the product of the depth demarcation function and the moisture content as the interlayer coupling term, where the depth demarcation function takes the value of 1 at the mutation of soil layer properties and 0 at other positions, and superimpose the basic coupling term, the dynamic coupling term and the interlayer coupling term to form a coupling correction coefficient.

[0078] Exemplarily, construct a depth influence function. Taking test pile A as an example, under the action of a 20kN uplift force, the shear stress at the ground surface (depth 0m) is 5.8kPa, 7.6kPa at a depth of 0.3m, 9.1kPa at a depth of 0.6m, 10.3kPa at a depth of 0.9m, 11.2kPa at a depth of 1.2m, 11.8kPa at a depth of 1.5m, 12.1kPa at a depth of 1.8m, 12.0kPa at a depth of 2.1m, 11.7kPa at a depth of 2.4m, 11.1kPa at a depth of 2.7m, and 10.2kPa at a depth of 3.0m.

[0079] Take the shear stress of 5.8kPa at the ground surface as the reference shear stress, and calculate the ratio of the shear stress at other depths to the reference shear stress. For example, the ratio at a depth of 0.3m is 1.31, 1.57 at a depth of 0.6m, and so on. Take the normalized depth (the ratio of the actual depth to the pile length) as the abscissa and the shear stress ratio as the ordinate, and plot a scatter diagram. Represent the depth influence function in the form of a power function, and fit the scatter data by the least squares method. Use computer numerical analysis software, input the depth and shear stress ratio data, and set the power function form as "a times x to the power of b", where x is the normalized depth, and a and b are undetermined coefficients. Through iterative calculation to minimize the sum of squared errors, the initial fitting result is obtained: a = 1.32, b = 0.43. At this time, the initial power function exponent b = 0.43 is the initial exponent of the power function.

[0080] Considering the influence of soil properties on the depth influence function, introduce the coefficient of earth pressure at rest and the relative density to correct the power function exponent. Take multiple soil samples in the test area to measure the relative density, which are 0.65, 0.72, 0.78, 0.82 and 0.76 respectively, and the average value is 0.75. Through comparative tests under different relative density conditions, analyze the influence law of relative density on the power function exponent, and determine that the relative density influence coefficient is 0.08. This means that when the relative density increases by 1.0, the power function exponent increases by 0.08. Since the measured relative density is 0.75, its contribution to the final power function exponent is 0.08 times 0.75, which is equal to 0.06.

[0081] According to soil mechanics theory, the coefficient of earth pressure at rest \(K_0 = 0.42\). Through multiple groups of comparative tests and analyses, the influence coefficient of the coefficient of earth pressure at rest on the power function exponent is determined to be \(0.15\). A linear relationship between the power function exponent, the coefficient of earth pressure at rest, and the relative density is established: the final power function exponent is equal to the initial exponent \(0.43\), plus the influence of the coefficient of earth pressure at rest \(0.15\times0.42\) (equal to \(0.063\)), plus the influence of the relative density \(0.06\). The calculated final power function exponent is \(0.43 + 0.063+0.06 = 0.553\), and rounding it gives \(0.55\).

[0082] Thus, the depth influence function is "1.32 times the 0.55th power of the normalized depth". To verify this function, calculate the shear stress ratio at a depth of \(1.5\ m\) (normalized depth of \(0.5\)) as \(1.32\times0.5^{0.55}\), which is equal to \(1.32\times0.682\), approximately \(2.00\). The corresponding shear stress is the reference shear stress \(5.8\ kPa\times2.00\), equal to \(11.6\ kPa\), which is close to the measured value of \(11.8\ kPa\), with a relative error of \(1.7\%\).

[0083] Construct the water content influence function. Collect soil samples at different points in the test area to measure the in-situ water content, and calculate the average value to be \(2.5\%\). Design a water content gradient test, and repeat the above pull-out test under the conditions of water contents of \(1\%\), \(2\%\), \(2.5\%\), \(3\%\), \(4\%\), and \(5\%\) respectively to obtain the shear stress data at the same depth under different water content conditions.

[0084] Taking the depth of \(1.5\ m\) as an example, the shear stress is \(7.3\ kPa\) when the water content is \(1\%\), \(10.1\ kPa\) when the water content is \(2\%\), \(11.8\ kPa\) when the water content is \(2.5\%\), \(10.9\ kPa\) when the water content is \(3\%\), \(8.7\ kPa\) when the water content is \(4\%\), and \(7.0\ kPa\) when the water content is \(5\%\). Taking the shear stress of \(11.8\ kPa\) at the in-situ water content of \(2.5\%\) as the reference, calculate the ratio of the shear stress under other water content conditions to the reference shear stress.

[0085] The water content influence function is constructed as a piecewise linear function, with the in-situ water content of 2.5% as the demarcation point. For the data points with water content less than 2.5%, linear fitting is carried out, and the first slope is obtained as 0.55, indicating that for every 1% decrease in water content, the shear stress ratio decreases by 0.55. For the data points with water content greater than or equal to 2.5%, linear fitting is carried out, and the second slope is obtained as -0.38, indicating that for every 1% increase in water content, the shear stress ratio decreases by 0.38. Therefore, the water content influence function is "1 plus 0.55 times (water content minus 2.5%)" when the water content is less than 2.5%, and "1 plus -0.38 times (water content minus 2.5%)" when the water content is greater than or equal to 2.5%. To verify this function, calculate the shear stress ratio at a water content of 4% as 1 plus -0.38 times (4 - 2.5), which is equal to 1 - 0.57, approximately 0.73. The corresponding shear stress is the reference shear stress of 11.8 kPa multiplied by 0.73, equal to 8.61 kPa, which is close to the measured value of 8.7 kPa, with a relative error of 1.0%.

[0086] Finally, a coupling correction coefficient is constructed. There is an interactive effect between depth and water content, and correction needs to be carried out through the coupling term. First, a basic coupling term is constructed, and the product of the depth exponent 0.55 and the water content deviation (actual water content minus in-situ water content) is used as the coefficient of the basic coupling term. By comparing and analyzing the test data under different combinations of depth and water content, the basic coupling coefficient is determined to be -0.12. A dynamic coupling term is constructed, and the product of the power term of depth (the 0.55th power of the normalized depth) and the water content is used as the coefficient of the dynamic coupling term. Analysis shows that the dynamic coupling coefficient is 0.04. An interlayer coupling term is constructed at the soil layer property mutation. Through soil sample analysis, it is found that there is an obvious soil layer change at a depth of 1.8 m in the test area. The upper layer is fine sand, and the lower layer is silty fine sand. A depth demarcation function is defined, which takes a value of 1 at a depth of 1.8 m and 0 at other positions. The product of the depth demarcation function and the water content is used as the coefficient of the interlayer coupling term, which is determined to be 0.15. The above three coupling terms are superimposed to form the coupling correction coefficient. For example, under the conditions of a depth of 1.5 m (normalized depth 0.5) and a water content of 4%, the basic coupling term is -0.12 times 0.55 times (4 - 2.5), equal to -0.099; the dynamic coupling term is 0.04 times 0.5 to the 0.55th power times 4, equal to 0.109; since this point is not at the soil layer mutation, the interlayer coupling term is 0; the coupling correction coefficient is -0.099 plus 0.109 plus 0, equal to 0.01. Combining the depth influence function, the water content influence function, and the coupling correction coefficient, the functional relationship between the shear stress at the pile-soil interface and depth and water content can be obtained.

[0087] Figure 2It is a comparison chart of the prediction accuracy of the depth-water content coupling effect. In the chart, the blue bar chart represents the relative error of the method of the present invention, and the red bar chart represents the relative error of the method without coupling correction. It can be clearly seen from the chart that under most test conditions, the prediction error of the method of the present invention is significantly lower than that of the model without coupling correction. Especially under the combined conditions of a depth of 0.3 m and a water content of 1.0%, the relative error of the method of the present invention is only 2.2%, while the relative error of the model without coupling correction is as high as 13.3%; under the conditions of a depth of 1.5 m and a water content of 1.0%, the relative error of the method of the present invention is 1.4%, and that of the model without coupling correction is 12.3%; under the conditions of a depth of 2.7 m and a water content of 4.0%, the relative error of the method of the present invention is 1.3%, and that of the model without coupling correction is as high as 7.6%. The average error of the present invention (including coupling correction) is only 1.55%, while the average error of the model without coupling correction is as high as 10.07%. This result fully proves the importance of considering the depth-water content coupling effect in the present invention for improving the prediction accuracy of the pile-soil interface shear stress, and verifies the technical value of the coupling correction in the design of the desert photovoltaic support pile foundation. The present invention constructs a depth influence function, a water content influence function and their coupling correction coefficients, establishes an accurate functional relationship between the pile-soil interface shear stress and the depth and water content, overcomes the limitation of the traditional method that ignores the particularity of the desert environment, realizes the accurate description of the mechanical behavior of the pile-soil interface, provides a reliable basis for optimizing the design and construction parameters of the photovoltaic support pile foundation, and improves the system reliability and economy.

[0088] In an alternative embodiment,

[0089] Establishing a multi-objective optimization function and solving to obtain the optimal construction parameter combination includes:

[0090] The anti-pull bearing capacity stability target is determined by the dispersion degree of the anti-pull bearing capacity under different depths and different water content conditions, the anti-pull bearing capacity surplus target is determined by the deviation between the actual anti-pull bearing capacity and the required anti-pull bearing capacity, and the construction efficiency target is determined by the ratio of the pile foundation construction time to the reference construction time;

[0091] Dynamically adjusting the weight coefficients of the multi-objective optimization function based on the environmental temperature and sand layer migration. The environmental temperature is described by an annual cycle sine function, and the sand layer migration determines the adjustment value of the weight coefficient by the coupling effect of the wind-induced migration rate and the migration duration, combined with the non-uniformity of the influence of the wind direction angle on the pile-soil interface;

[0092] The particle swarm optimization algorithm is used to solve the multi-objective optimization function. The particle swarm optimization algorithm is improved by adjusting the inertia weight and learning factors through the environmental temperature and sand layer migration dynamics, and the particle positions including the pile foundation diameter, pile foundation depth, and surface treatment method are iteratively updated to obtain the optimal construction parameter combination that meets the uplift bearing capacity constraint and geometric constraint conditions.

[0093] Exemplarily, the uplift bearing capacity stability target is characterized by calculating the coefficient of variation of the uplift bearing capacity under different depths and different water contents. In specific implementation, representative depth points (such as 0.5m, 1.0m, 1.5m, 2.0m, 3.0m) and typical water content points in the desert area (such as 0.5%, 1.0%, 2.0%, 3.0%, 4.0%) are selected, the uplift bearing capacity under each combined condition is calculated, and then the ratio of the standard deviation to the average value of all samples is calculated as the coefficient of variation. For example, in a certain desert construction sub-region, the uplift bearing capacity at a depth of 2.0m and a water content of 1.0% is 15.2kN. By calculating the average value of the uplift bearing capacity under 25 condition combinations as 14.8kN and the standard deviation as 2.3kN, the coefficient of variation is 0.155. The stability target value is 1 minus the coefficient of variation, and the closer it is to 1, the higher the stability.

[0094] The uplift bearing capacity redundancy target is determined by the difference ratio between the actual uplift bearing capacity and the designed required uplift bearing capacity. For example, when the actual uplift bearing capacity is 18.5kN and the designed requirement is 15kN, the redundancy is (18.5 - 15) / 15 = 23.3%. Considering the safety redundancy requirement, the ideal redundancy range is set to 15% - 30%. When the redundancy is lower than 15%, the redundancy target value decreases linearly; when the redundancy is within the range of 15% - 30%, the redundancy target value is 1; when the redundancy exceeds 30%, the redundancy target value decreases exponentially with the increase of the redundancy to avoid resource waste.

[0095] The construction efficiency target is determined by the ratio of the pile foundation construction time to the reference construction time. The reference construction time is set to 60 minutes, corresponding to the construction conditions of a standard pile foundation diameter of 200mm, a depth of 2.5m, and no special surface treatment. When the actual construction time is 45 minutes, the time ratio is 45 / 60 = 0.75, and the efficiency target value is 1 / 0.75 = 1.33. When the actual construction time is 80 minutes, the time ratio is 80 / 60 = 1.33, and the efficiency target value is 1 / 1.33 = 0.75. Considering the above three targets, a multi-objective optimization function is constructed in the form of a weighted sum and the initial weights are set.

[0096] The weight coefficients of the multi-objective optimization function are dynamically adjusted based on the environmental temperature and sand layer migration. The particle swarm optimization algorithm is used to solve the multi-objective optimization function to obtain the optimal solution.

[0097] The present invention realizes the balance between the uplift bearing capacity of the desert photovoltaic support pile foundation and the construction efficiency through multi-objective optimization. The dynamic adjustment mechanism of environmental temperature and sand layer migration improves the adaptability of the scheme. The improvement of the particle swarm algorithm ensures the solution efficiency and accuracy, providing technical guarantee for the safe and reliable operation of the desert photovoltaic support.

[0098] In an alternative embodiment,

[0099] The dynamic adjustment of the weight coefficients of the multi-objective optimization function based on environmental temperature and sand layer migration includes:

[0100] Establish an environmental temperature fluctuation equation, which uses an annual cycle sine function to describe the desert environmental temperature change. Calculate the contact stress at the pile-soil interface based on the environmental temperature fluctuation equation, and use the ratio of the contact stress at the pile-soil interface to the reference contact stress as the temperature influence coefficient. Establish a temperature weight correction coefficient using a sine function including a first harmonic term and a second harmonic term;

[0101] Establish a wind-induced migration rate calculation equation, and determine the instantaneous migration rate using the wind force magnitude and wind direction angle; use the integral value of the instantaneous migration rate over the migration duration as the soil stress change amount; based on the soil stress change amount, establish a migration influence coefficient by combining the non-uniform influence of the soil around the pile with the wind direction angle;

[0102] Multiply the temperature weight correction coefficient by the migration influence coefficient and introduce a coupling enhancement term based on the soil stress change amount to establish an environmental coupling coefficient. Dynamically adjust the weight coefficients of the multi-objective optimization function based on the environmental coupling coefficient, where the weight coefficient of the uplift bearing capacity stability target is positively correlated with the ratio of the contact stress at the pile-soil interface, the weight coefficient of the uplift bearing capacity redundancy target is positively correlated with the soil stress change amount, and the weight coefficient of the construction efficiency target is negatively correlated with the environmental coupling coefficient; normalize the weight coefficients.

[0103] Exemplarily, such as Figure 3As shown in the flowchart of dynamic adjustment of weight coefficients by environmental factors, an environmental temperature fluctuation equation is established, and the annual cycle sine function is used to describe the temperature change in the desert environment. Taking a certain desert area as an example, the annual average temperature is 28°C, the annual maximum temperature is 55°C, the annual minimum temperature is 0°C, and the temperature amplitude is 27.5°C. The temperature fluctuation equation is expressed by the sine function, with a period of 365 days, an amplitude of 27.5°C, and an average value of 28°C. For example, on the 100th day of the year (early April), the calculated temperature value is 28 + 27.5×sine value (based on the angle between the current day and the reference day), and the calculated value is approximately 40°C. To enhance the accuracy of the temperature fluctuation equation, the first harmonic term and the second harmonic term are introduced, and the adjustment coefficients are 0.15 and 0.08 respectively. The period of the first harmonic term is 182.5 days, and the period of the second harmonic term is 91.25 days. By introducing these harmonic terms, the characteristics that the temperature changes relatively slowly in spring and autumn and relatively quickly in summer and winter in the desert area can be more accurately simulated.

[0104] Calculate the contact stress at the pile-soil interface based on the environmental temperature fluctuation equation. When the temperature rises, the thermal expansion coefficient of the pile material (usually metal or composite material) (about 1.2×10 -5 / °C) is greater than that of the sandy soil (about 0.5×10 -5 / °C), resulting in a greater expansion of the pile relative to the soil when the temperature rises, increasing the contact stress at the pile-soil interface. Taking 20°C as the reference temperature, when the temperature rises to 50°C, the temperature increment is 30°C, and the difference in thermal expansion coefficients between the pile material and the sandy soil is 0.7×10 -5 / °C. The calculated radial strain caused by the thermal expansion difference is 0.021%. Referring to the stiffness characteristics of the pile body, the increase in contact stress can be estimated to be about 12%.

[0105] Take the ratio of the contact stress at the pile-soil interface to the reference contact stress (the contact stress at 20°C) as the temperature influence coefficient. For example, when the temperature is 50°C, the contact stress ratio is 1.12, that is, the temperature influence coefficient is 1.12; when the temperature is 0°C, the contact stress ratio is about 0.92, and the temperature influence coefficient is 0.92. A sine function containing the first harmonic term and the second harmonic term is used to establish the temperature weight correction coefficient. The amplitude of the basic sine term is 0.15, the amplitude of the first harmonic term is 0.05, and the amplitude of the second harmonic term is 0.02. By superimposing these three terms, the temperature weight correction coefficient is obtained, and its value range at different times within the year is about 0.85 - 1.15.

[0106] An equation for calculating the wind-induced migration rate is established, and the instantaneous migration rate is determined by the wind force magnitude and the wind direction angle. When the wind speed exceeds the threshold wind speed (usually 4 m / s), sand grains start to move; when the wind speed is 8 m / s, the sand grain migration rate is approximately 2 mm / hour; when the wind speed is 12 m / s, the migration rate is approximately 8 mm / hour; when the wind speed is 16 m / s, the migration rate is approximately 20 mm / hour. The wind direction angle has a regulatory effect on the migration rate. Assuming the windward side is 0°, when the wind direction angle is within the range of 0° ± 30°, the migration rate reaches the maximum value; when the wind direction angle is 90° ± 30° or 270° ± 30°, the migration rate decreases to 0.4 times the maximum value; when the wind direction angle is 180° ± 30°, the migration rate decreases to 0.2 times the maximum value. For example, when the wind speed is 12 m / s and the wind direction angle is 20°, the migration rate is 8 mm / hour; while when the wind direction angle is 100°, the migration rate decreases to 3.2 mm / hour. The integral value of the instantaneous migration rate over the migration duration is taken as the change in soil body stress. For example, in a certain desert area, the wind speed is 12 m / s and the wind direction angle is 20° for 6 consecutive hours, and the calculated total migration amount is 48 mm, which is further converted into the change in soil body stress. Based on the change in soil body stress, a migration influence coefficient is established in combination with the non-uniform influence of the wind direction angle on the soil around the pile. The non-uniformity coefficient depends on the wind direction stability. When the wind direction is concentrated within a 30° range, the non-uniformity coefficient is 0.8; when the wind direction is distributed within a 60° range, the non-uniformity coefficient is 0.9; when the wind direction is distributed over a range of more than 90°, the non-uniformity coefficient is 1.0. The migration influence coefficient is equal to 1 plus the product of the stress change amount and the non-uniformity coefficient. For example, when the stress change amount is -10% and the non-uniformity coefficient is 0.8, the migration influence coefficient is 1 + (-10%) × 0.8 = 0.92.

[0107] Multiply the temperature weight correction coefficient by the migration influence coefficient and introduce a coupling enhancement term based on the change in soil body stress to establish an environmental coupling coefficient. The coupling enhancement term is designed as the product of 0.05 and the 0.5 power of the absolute value of the change in soil body stress. For example, when the temperature weight correction coefficient is 1.14, the migration influence coefficient is 0.92, and the change in soil body stress is -10%, the calculated coupling enhancement term is , and the environmental coupling coefficient is 1.14 × 0.92 × (1 + 0.016) = 1.065.

[0108] Dynamically adjust the weight coefficients of the multi-objective optimization function based on the environmental coupling coefficient. The initial weight of the uplift bearing capacity stability objective is set to 0.4, which is positively correlated with the ratio of the pile-soil interface contact stress, and the adjustment coefficient is 0.2; the initial weight of the redundancy objective is set to 0.35, which is positively correlated with the absolute value of the change in soil stress, and the adjustment coefficient is 0.15; the initial weight of the efficiency objective is set to 0.25, which is negatively correlated with the environmental coupling coefficient, and the adjustment coefficient is -0.1. For example, when the environmental coupling coefficient is 1.065, the ratio of the pile-soil interface contact stress is 1.12, and the change in soil stress is -10%, the weight of the stability objective is adjusted to 0.4 + 0.2×(1.12 - 1) = 0.424; the weight of the redundancy objective is adjusted to 0.35 + 0.15×0.1 = 0.365; the weight of the efficiency objective is adjusted to 0.25 - 0.1×(1.065 - 1) = 0.2435. Normalize the adjusted weight coefficients to obtain the final weights of 0.41, 0.35, and 0.24 respectively.

[0109] In the changing desert environment, the traditional photovoltaic support pile foundation design method usually adopts a multi-objective optimization function with fixed weights, and the weight coefficients are no longer adjusted once determined. This static optimization method has obvious limitations in practical applications. The improvement of the present invention over the prior art lies in establishing a dynamic correlation mechanism between environmental factors and weight coefficients, introducing a temperature fluctuation equation containing harmonic terms to more accurately describe the desert temperature change characteristics; constructing a wind-induced migration rate calculation equation and considering the non-uniform influence of the wind direction angle on the soil around the pile; most importantly, innovatively proposing the concept of environmental coupling coefficient, and comprehensively capturing the interactive influence of environmental factors by multiplying the temperature weight correction coefficient by the migration influence coefficient and introducing a coupling enhancement term. The present invention can solve the problem that the existing optimization method lacks response to environmental changes, enabling the weight coefficients to be adaptively adjusted according to environmental dynamic changes, so that the optimization results are more in line with the actual engineering requirements. The present invention realizes the dynamic adaptive adjustment of the weights of the multi-objective optimization function by accurately capturing the influence of desert environmental temperature fluctuations and sand layer migration on the pile foundation performance, improves the stability and safety margin of the optimization results in extreme environments, and ensures the reliable operation of desert photovoltaic support pile foundations under different climate conditions throughout the year.

[0110] In an alternative embodiment,

[0111] Use the particle swarm algorithm to solve the multi-objective optimization function, and obtain the optimal construction parameter combination including:

[0112] Establish a dynamic inertia weight calculation formula based on the environmental coupling coefficient. The dynamic inertia weight calculation formula adopts the exponential difference form of the maximum inertia weight and the minimum inertia weight, and adjusts the change rate of the inertia weight through the environmental coupling coefficient;

[0113] Establish calculation formulas for the individual optimal solution learning factor and the global optimal solution learning factor in the form of exponential functions, and perform adaptive adjustment through the ratio of the difference between the current particle fitness and the maximum and minimum fitnesses of the population;

[0114] Construct a default penalty function, the default penalty function includes a uplift bearing capacity constraint term and a geometric constraint term, the uplift bearing capacity constraint term adopts the form of the square of the difference between the actual uplift bearing capacity and the required uplift bearing capacity, and the geometric constraint term adopts the form of the square of the difference between the pile diameter and the pile embedment depth and their constraint boundaries;

[0115] Encode the particles including the pile diameter, the pile embedment depth, and the surface treatment method, iteratively update the particle swarm using the dynamic inertia weight and the adaptive learning factor, combine the default penalty function with the multi-objective optimization function to calculate the particle fitness, and iteratively optimize to obtain the optimal construction parameter combination that meets the constraint conditions.

[0116] Exemplarily, establish a calculation formula for the dynamic inertia weight based on the environmental coupling coefficient. The setting range of the inertia weight is from 0.4 to 0.9, where 0.9 is the maximum inertia weight, which is beneficial for global search; 0.4 is the minimum inertia weight, which is beneficial for local development. The dynamic inertia weight is expressed in the form of the exponential difference between the maximum inertia weight and the minimum inertia weight, specifically as the maximum inertia weight minus the product of the difference between the maximum inertia weight and the minimum inertia weight and the square of the deviation of the environmental coupling coefficient from 1. For example, when the environmental coupling coefficient is 1.065, the dynamic inertia weight is calculated as 0.9 minus (0.9 minus 0.4) multiplied by the square of (1.065 minus 1), resulting in 0.9 minus 0.5 multiplied by 0.004225, and the calculation result is 0.9 minus 0.002113, finally 0.898. When the environmental coupling coefficient is 0.92, the dynamic inertia weight is calculated as 0.9 minus 0.5 multiplied by 0.0064, resulting in 0.9 minus 0.0032, finally 0.897. In this way, the more extreme the environmental conditions (the greater the deviation of the environmental coupling coefficient from 1), the closer the inertia weight is to 0.9, enhancing the global search ability; the milder the environmental conditions (the closer the environmental coupling coefficient is to 1), the faster the inertia weight approaches 0.4, enhancing the local development ability.

[0117] To further optimize the algorithm performance, calculation formulas for the individual optimal solution learning factor and the global optimal solution learning factor in the form of exponential functions are established. The learning factor controls the degree to which the particle learns from the individual historical optimal position and the population global optimal position. The initial value of the individual optimal solution learning factor is set to 1.5, and the initial value of the global optimal solution learning factor is set to 2.0. The two learning factors are adaptively adjusted by the ratio of the difference between the current particle fitness and the maximum and minimum fitness of the population. The specific adjustment method is as follows: Define the relative fitness position as the current particle fitness minus the population minimum fitness, and then divide by the population maximum fitness minus the population minimum fitness. For example, if the current particle fitness is 0.75, the population maximum fitness is 0.92, and the minimum fitness is 0.51, then the relative fitness position is (0.75 - 0.51) / (0.92 - 0.51) = 0.24 / 0.41 = 0.585. The individual optimal solution learning factor decreases as the relative fitness position increases, and uses the exponential function with the initial value multiplied by the exponential of negative 5 times the relative fitness position with the natural logarithm base. Taking the above example, the individual learning factor is calculated as 1.5 multiplied by the exponential function with the natural logarithm base and the exponential of -2.926 (negative 5 times 0.585), that is, 1.5 multiplied by the exponential of -2.926 with the natural logarithm base, resulting in 1.5 multiplied by 0.054, and the calculation result is 0.081. The global optimal solution learning factor increases as the relative fitness position increases, and uses the exponential function with the initial value multiplied by the exponential of 2 times the relative fitness position with the natural logarithm base. Taking the above example, the global learning factor is calculated as 2.0 multiplied by the exponential function with the natural logarithm base and the exponential of 1.17 (2 times 0.585), that is, 2.0 multiplied by the exponential of 1.17 with the natural logarithm base, resulting in 2.0 multiplied by 3.22, and the calculation result is 6.44.

[0118] To ensure that the optimization results meet the engineering constraints, a penalty function for violation is constructed. The penalty function for violation includes a pull-out bearing capacity constraint term and a geometric constraint term. The pull-out bearing capacity constraint term takes the form of the square of the difference between the actual pull-out bearing capacity and the required pull-out bearing capacity. For example, when the actual pull-out bearing capacity is 12.5 kN and the required pull-out bearing capacity is 15 kN, the pull-out bearing capacity constraint term is the square of (12.5 - 15), which is 6.25. When the actual pull-out bearing capacity is greater than or equal to the required pull-out bearing capacity, this constraint term is 0. The geometric constraint term takes the form of the square of the difference between the pile diameter and pile embedment depth and their constraint boundaries. The pile diameter constraint is from 150 mm to 350 mm, and the pile embedment depth constraint is from 2.0 m to 4.0 m. For example, when the pile diameter is 130 mm, which is below the minimum constraint boundary of 150 mm, the diameter constraint term is the square of (130 - 150), which is 400; when the pile embedment depth is 4.2 m, which exceeds the maximum constraint boundary of 4.0 m, the embedment depth constraint term is the square of (4.2 - 4.0), which is 0.04. When the parameter values are within the constraint range, the corresponding constraint terms are 0. The penalty function for violation is the sum of each constraint term multiplied by a penalty coefficient, and the penalty coefficient is set to 10. In the above example, the value of the penalty function for violation is (6.25 + 400 + 0.04) multiplied by 10, which is 4062.9.

[0119] In practical applications, particles including pile diameter, pile embedment depth, and surface treatment method are encoded. The pile diameter and pile embedment depth adopt real-number encoding, with ranges of 150 mm to 350 mm and 2.0 m to 4.0 m respectively; the surface treatment method adopts discrete encoding, where 0 represents smooth, 1 represents rough, 2 represents spiral ribs, and 3 represents annular ribs. For example, the particle encoding [220, 3.2, 2] means the pile diameter is 220 mm, the pile embedment depth is 3.2 m, and the surface treatment method is spiral ribs.

[0120] The population size is initialized to 50, and the initial particle positions and velocities are randomly generated. The particle positions are uniformly randomly distributed within the parameter constraint range, and the particle velocities are randomly distributed between -10% and 10% of the corresponding parameter range length. For example, the velocity range of the pile diameter is from -20 mm to 20 mm, and the velocity range of the pile embedment depth is from -0.2 m to 0.2 m.

[0121] The particle swarm is iteratively updated using the aforementioned dynamic inertia weight and adaptive learning factor. In each iteration, first, the multi-objective optimization function value of each particle is calculated. For example, if a particle is encoded as [220, 3.2, 2], the calculated stability index of the uplift bearing capacity is 0.92, the redundancy is 24%, and the efficiency index is 1.15; the weights are 0.41, 0.35, and 0.24 respectively; then the multi-objective optimization function value is 0.92 multiplied by 0.41 plus 1.0 multiplied by 0.35 plus 1.15 multiplied by 0.24, and the calculation result is 0.3772 + 0.35 + 0.276 = 1.003. Next, the penalty function value for violation is calculated. If the actual uplift bearing capacity corresponding to this particle is 18.6 kN, which is higher than the required 15 kN, the uplift bearing capacity constraint term is 0; the pile diameter and pile embedment depth are both within the constraint range, and the geometric constraint term is also 0; then the penalty function value for violation is 0. The particle fitness is calculated by combining the multi-objective optimization function value and the penalty function value for violation, that is, 1.003 - 0 = 1.003. If a particle violates the rules significantly, such as the actual uplift bearing capacity is only 8 kN, then the penalty function value for violation is the square of (8 - 15) multiplied by 10, that is, 490, and the fitness of this particle is the multi-objective optimization function value minus 490, which may be negative, indicating an unacceptable solution. The particle velocity and position are updated according to the dynamic inertia weight and adaptive learning factor. For example, when the dynamic inertia weight is 0.897, the individual learning factor is 0.081, and the global learning factor is 6.44, the pile diameter velocity of a certain particle is updated from 5 mm to 5 mm multiplied by 0.897 plus 0.081 multiplied by the difference between the individual optimum and the current position plus 6.44 multiplied by the difference between the global optimum and the current position. The updated position is the current position plus the updated velocity. For discrete variables such as the surface treatment method, the probability rounding method is used to determine the updated value.

[0122] The maximum number of iterations is set to 200. When the maximum number of iterations is reached or the improvement amplitude of the optimal solution is less than 0.001 for 30 consecutive iterations, the algorithm terminates. The final output is the optimal construction parameter combination that meets the constraint conditions. For example, the pile diameter is 240 mm, the pile embedment depth is 3.5 m, the surface treatment method is spiral ribs, the corresponding uplift bearing capacity is 20.3 kN, the stability index is 0.95, the redundancy is 35%, the efficiency index is 1.08, and the comprehensive score is 1.205.

[0123] When dealing with the multi-objective optimization problem in the desert environment, the traditional particle swarm optimization algorithm exhibits disadvantages such as slow convergence speed, easy to fall into local optimum, and inflexible handling of constraint conditions. The present invention establishes a dynamic inertia weight calculation formula based on the environmental coupling coefficient, enabling the inertia weight to be adaptively adjusted according to environmental conditions, enhancing the global search ability when the environment is harsh and the local development ability when the environment is mild; designs an adaptive learning factor in the form of an exponential function, dynamically adjusts the proportion of individual learning and global learning through the relative fitness position of particles, enhancing the search diversity and convergence efficiency of the algorithm; constructs a default penalty function that comprehensively considers the uplift bearing capacity and geometric constraints, effectively guiding the search direction through the penalty term in the form of a square to ensure that the optimization result meets the engineering constraints.

[0124] By introducing the inertia weight and adaptive learning factor dynamically adjusted by environmental factors, the present invention significantly improves the solution efficiency and accuracy of the particle swarm optimization algorithm in the complex desert environment. The default penalty mechanism ensures that the optimization result meets the engineering constraints, realizes the precise optimization of the construction parameters of the desert photovoltaic support pile foundation, and provides a scientific and reliable decision-making basis for engineering practice.

[0125] In an optional implementation manner,

[0126] Adopting the optimal construction parameter combination for pile foundation construction and performing dynamic fine-tuning includes:

[0127] Performing pile foundation construction in each construction sub-area according to the optimal construction parameter combination, and performing uplift bearing capacity detection after every preset number of pile foundation constructions are completed to obtain actual uplift bearing capacity data;

[0128] Using the time series analysis method to process the actual uplift bearing capacity data at multiple detection time points in the same construction sub-area to obtain the time-varying characteristics of the uplift bearing capacity;

[0129] Comparing the actual uplift bearing capacity data with the theoretical calculated value. When the deviation rate of the uplift bearing capacity exceeds the second preset threshold, correcting the weight coefficient in the multi-objective optimization function based on the time-varying characteristics, and re-performing optimization calculation using the corrected multi-objective optimization function to obtain an updated optimal construction parameter combination;

[0130] During the construction process, real-time monitor the changes in environmental temperature and the migration of sand layers. When their changes exceed the preset range, trigger the dynamic adjustment of the multi-objective optimization function.

[0131] Exemplarily, pile foundation construction is carried out according to the optimal construction parameter combination in each construction sub - area. Taking a certain desert photovoltaic project as an example, the site is divided into four construction sub - areas: the northeast area, the southeast area, the southwest area, and the northwest area. The optimal construction parameter combinations are determined respectively according to the differences in soil characteristics. For example, the optimal parameter combination in the northeast area is a pile foundation diameter of 220 mm, a pile foundation burial depth of 3.5 m, and the surface treatment method is spiral ribs; the optimal parameter combination in the southeast area is a pile foundation diameter of 240 mm, a pile foundation burial depth of 3.2 m, and the surface treatment method is annular ribs; the optimal parameter combination in the southwest area is a pile foundation diameter of 200 mm, a pile foundation burial depth of 3.8 m, and the surface treatment method is spiral ribs; the optimal parameter combination in the northwest area is a pile foundation diameter of 250 mm, a pile foundation burial depth of 3.0 m, and the surface treatment method is rough.

[0132] During the construction process, a preset detection frequency is set. After every 50 pile foundation constructions are completed, 3 representative pile foundations are selected for uplift bearing capacity detection to obtain the actual uplift bearing capacity data. For example, the uplift bearing capacities of the 3 pile foundations detected in the first batch in the northeast area are 19.2 kN, 18.7 kN, and 20.1 kN respectively, and the average value is 19.3 kN; the average uplift bearing capacity detected in the first batch in the southeast area is 21.5 kN; in the southwest area it is 18.8 kN; and in the northwest area it is 20.7 kN.

[0133] As the construction progresses, the detection data is continuously accumulated. For example, the average uplift bearing capacity detected in the second batch in the northeast area is 18.5 kN, in the third batch it is 17.8 kN, and in the fourth batch it is 17.2 kN. These data are processed using time - series analysis methods to obtain the time - varying characteristics of the uplift bearing capacity. The time - series analysis method includes the combined application of three techniques: the moving average method, the exponential smoothing method, and linear regression analysis. The moving average method uses 3 batches as the window width to calculate the moving average value and eliminate the influence of random fluctuations. For example, the moving average value of the first to the third batches in the northeast area is (19.3 + 18.5 + 17.8) / 3 = 18.5 kN, and the moving average value of the second to the fourth batches is (18.5 + 17.8 + 17.2) / 3 = 17.8 kN. The exponential smoothing method uses a smoothing coefficient of 0.3 and combines the historical smoothed value and the current measured value to calculate a new smoothed value. For example, if the previous smoothed value is 19.0 kN and the current measured value is 17.2 kN, then the new smoothed value is 19.0×(1 - 0.3)+17.2×0.3 = 18.5 kN. Linear regression analysis uses the least - squares method to fit the changing trend of the uplift bearing capacity over time. For example, the data of four batches in the northeast area are fitted to obtain a slope of - 0.7 kN / batch, indicating that the uplift bearing capacity shows a downward trend.

[0134] Based on the results of the three analysis methods, the time-varying characteristic parameters of the uplift bearing capacity are extracted, including the short-term change rate, the medium-term trend coefficient, and the long-term stability coefficient. The short-term change rate is defined as the relative change percentage of the test values of two adjacent batches. For example, the short-term change rate from the third batch to the fourth batch in the Northeast region is (17.2 - 17.8) / 17.8×100% = -3.4%. The medium-term trend coefficient is the ratio of the slope of the linear regression of four consecutive batches to the initial value. For example, the medium-term trend coefficient in the Northeast region is -0.7 / 19.3×100% = -3.6% / batch. The long-term stability coefficient is determined through the convergence analysis of the exponential smoothing method, and its value range is from 0 to 1. The closer it is to 1, the better the long-term stability. The long-term stability coefficient in the Northeast region is calculated to be 0.85, indicating that although there is a downward trend, it generally remains relatively stable.

[0135] The actual uplift bearing capacity data is compared with the theoretical calculated value, and the deviation rate of the uplift bearing capacity is calculated. The theoretical calculated value is based on the multi-objective optimization function. For example, the theoretical calculated value in the Northeast region is 20.5 kN. After the fourth batch of tests, the actual average value is 17.2 kN, and the deviation rate is (17.2 - 20.5) / 20.5×100% = -16.1%. The second preset threshold is set at ±15%. Here, the deviation rate exceeds the threshold, triggering the parameter adjustment mechanism.

[0136] Based on the time-varying characteristics of the uplift bearing capacity, the weight coefficients in the multi-objective optimization function are corrected. The time-varying characteristics show that the uplift bearing capacity in the Northeast region shows a downward trend, indicating potential problems with stability, and the weight of the stability objective needs to be increased. The correction formula adopts a comprehensive adjustment based on the time-varying characteristic parameters. Among them, the short-term change rate affects the urgency of the weight adjustment, the medium-term trend coefficient affects the direction and amplitude of the adjustment, and the long-term stability coefficient affects the upper limit of the adjustment. The specific adjustment strategy is as follows: when the absolute value of the short-term change rate exceeds 3%, the weight is adjusted immediately; when the medium-term trend coefficient is negative and its absolute value exceeds 2% / batch, the weight of the stability objective is increased; when the long-term stability coefficient is lower than 0.9, the adjustment amplitude is increased. In the example of the Northeast region, the original weight of the stability objective is 0.4. According to the adjustment strategy, the increment is calculated to be 0.08, and the adjusted weight is 0.48; the original weight of the redundancy objective is 0.35, the calculated reduction is 0.05, and the adjusted value is 0.3; the original weight of the efficiency objective is 0.25, the calculated reduction is 0.03, and the adjusted value is 0.22. The adjusted weight coefficients are normalized, and the final weights are 0.48, 0.3, and 0.22 respectively.

[0137] Re-optimize using the corrected multi-objective optimization function to obtain an updated optimal combination of construction parameters. For example, the updated optimal parameter combination in the Northeast region is a pile foundation diameter of 240 mm (increased by 20 mm), a pile foundation burial depth of 3.7 m (increased by 0.2 m), and the surface treatment method remains spiral ribs. The theoretical uplift bearing capacity of the new parameter combination is 22.8 kN, which is 11.2% higher than the original plan, and it is expected to make up for the performance decline found in the actual inspection.

[0138] During the construction process, monitor the changes in environmental temperature and the migration of sand layers in real time. When the environmental parameters change beyond the preset range, trigger the dynamic adjustment of the multi-objective optimization function. For example, on a certain day, it is detected that the sand layer migration in the Southwest region reaches 30 mm / week, far exceeding the expected level. It is speculated that the monthly cumulative migration will exceed 120 mm, triggering the dynamic adjustment mechanism. At this time, re-evaluate the environmental coupling coefficient based on the migration impact model, which drops from the original 0.95 to 0.82, indicating that the environmental change significantly increases the instability risk. Accordingly, adjust the weight of the stability objective in the multi-objective optimization function from 0.42 to 0.52, the weight of the redundancy objective from 0.33 to 0.38, and the weight of the efficiency objective from 0.25 to 0.1. Recalculate the optimal construction parameters based on the adjusted optimization function. The pile foundation burial depth in the Southwest region increases from 3.8 m to 4.0 m (reaching the upper limit), the pile foundation diameter increases from 200 mm to 230 mm, and the surface treatment method changes from spiral ribs to annular ribs to cope with the increased sand layer migration risk.

[0139] Through real-time detection and feedback mechanisms, combined with time series analysis methods to capture the time-varying characteristics of the uplift bearing capacity, the present invention realizes the dynamic fine-tuning of construction parameters, effectively responds to the complex changes in the desert environment, significantly improves the reliability and stability of photovoltaic support pile foundations under extreme conditions, and provides technical guarantee for the safe construction of large-scale desert photovoltaic projects.

[0140] In the second aspect of the embodiments of the present invention,

[0141] Provide an electronic device, including:

[0142] A processor;

[0143] A memory for storing instructions executable by the processor;

[0144] Wherein, the processor is configured to call the instructions stored in the memory to execute the method described above.

[0145] In the third aspect of the embodiments of the present invention,

[0146] Provide a computer-readable storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the method described above is implemented.

[0147] The present invention may be a method, an apparatus, a system, and / or a computer program product. The computer program product may include a computer-readable storage medium having thereon computer-readable program instructions for performing various aspects of the present invention.

[0148] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for detecting the uplift bearing capacity of desert photovoltaic support pile foundations and optimizing construction parameters, characterized in that, Including: Dividing the construction sub - areas according to the soil parameter information of the desert target area, and conducting pile testing in each construction sub - area, including: arranging multiple groups of strain sensors along the depth direction of the test pile body, conducting hierarchical loading tests on each test pile, and obtaining the strain data at each depth; combining the soil moisture content data collected in real - time during the testing process, calculating the contribution value of the soil layer to the uplift bearing capacity of the pile foundation under different depths and different moisture content conditions, and establishing the distribution law of the uplift bearing capacity at the pile - soil interface; Based on the above - mentioned distribution law, taking the pile foundation diameter, pile foundation burial depth, and pile foundation surface treatment method as variables, establishing a multi - objective optimization function including the uplift bearing capacity stability target, uplift bearing capacity redundancy target, and construction efficiency target; the weight coefficients of the multi - objective optimization function are dynamically adjusted according to the environmental temperature and sand layer migration, including: establishing an environmental temperature fluctuation equation, which uses an annual - cycle sine function to describe the desert environmental temperature change, calculating the contact stress at the pile - soil interface based on the environmental temperature fluctuation equation, taking the ratio of the contact stress at the pile - soil interface to the reference contact stress as the temperature influence coefficient, and establishing a temperature weight correction coefficient using a sine function including the first - harmonic term and the second - harmonic term; establishing a wind - induced migration rate calculation equation, determining the instantaneous migration rate using the wind force magnitude and wind direction angle; taking the integral value of the instantaneous migration rate along the migration duration as the soil stress change amount; based on the soil stress change amount, establishing a migration influence coefficient by combining the non - uniform influence of the wind direction angle on the soil around the pile; multiplying the temperature weight correction coefficient by the migration influence coefficient and introducing a coupling enhancement term based on the soil stress change amount to establish an environmental coupling coefficient, and dynamically adjusting the weight coefficients of the multi - objective optimization function based on the environmental coupling coefficient, where the weight coefficient of the uplift bearing capacity stability target is positively correlated with the ratio of the contact stress at the pile - soil interface, the weight coefficient of the uplift bearing capacity redundancy target is positively correlated with the soil stress change amount, and the weight coefficient of the construction efficiency target is negatively correlated with the environmental coupling coefficient; normalizing the weight coefficients; using the particle swarm algorithm to solve the multi - objective optimization function to obtain the optimal construction parameter combination for each construction sub - area; Carrying out pile foundation construction using the optimal construction parameter combination; during the construction process, selecting pile foundations for uplift bearing capacity review, and feeding back the review results to the multi - objective optimization function to dynamically fine - tune the construction parameters.

2. The method according to claim 1, characterized in that, Establishing the distribution law of the uplift bearing capacity at the pile - soil interface based on sensor data includes: The hierarchical loading test to obtain strain data includes loading step - by - step in percentage of the design ultimate load, and continuously loading each level of load until the strain change rate measured by the strain sensor is less than the first preset threshold; calculating the axial force of each depth section of the pile body based on the strain data, and calculating the shear stress at the pile - soil interface according to the axial force difference between adjacent depth sections; Establish the functional relationship between the shear stress at the pile-soil interface, depth, and water content. The functional relationship includes a depth influence function, a water content influence function, and a coupling correction coefficient. The depth influence function is represented by a power function to characterize the influence law of depth on the interface shear stress. The water content influence function is a piecewise linear function. The coupling correction coefficient characterizes the interaction between depth and water content; Use the functional relationship to calculate the contribution rate of the soil layer per unit depth to the total uplift bearing capacity of the pile foundation. The contribution rate is expressed as the percentage of the product of the shear stress at the pile-soil interface at this depth and the pile circumference area to the total uplift force at the pile top. Divide the pile foundation burial depth range into shallow layer section, middle layer section, and deep layer section, calculate the cumulative contribution values at different depth sections under different water content conditions, and determine the distribution law of the uplift bearing capacity at the pile-soil interface.

3. The method according to claim 2, wherein Establishing the functional relationship between the shear stress at the pile-soil interface, depth, and water content includes: Construct the depth influence function in the form of a power function. Taking the shear stress at the pile-soil interface at the ground surface as the reference shear stress, fit the ratio of the interface shear stress at different depths to the reference shear stress by the least squares method to determine the initial exponent of the power function. Determine the final power function exponent by the linear combination of the initial exponent of the power function, the coefficient of earth pressure at rest, and the relative density; Construct the water content influence function as a piecewise linear function. Taking the in-situ water content of the target area before construction as the demarcation point, use the first slope when the water content is less than the in-situ water content, and use the second slope when the water content is greater than or equal to the in-situ water content, respectively characterizing the influence of different water content intervals on the interface shear stress; Construct the product of the depth exponent and the water content deviation as the basic coupling term, construct the product of the depth power term and the water content as the dynamic coupling term, and construct the product of the depth demarcation function and the water content as the interlayer coupling term. The depth demarcation function takes the value of 1 at the mutation of soil layer properties and 0 at other positions. Superimpose the basic coupling term, the dynamic coupling term, and the interlayer coupling term to form the coupling correction coefficient.

4. The method according to claim 1, wherein Establish a multi-objective optimization function and solve to obtain the optimal construction parameter combination, including: The uplift bearing capacity stability target is determined by the dispersion degree of the uplift bearing capacity under different depths and different water content conditions. The uplift bearing capacity redundancy target is determined by the deviation between the actual uplift bearing capacity and the required uplift bearing capacity. The construction efficiency target is determined by the ratio of the pile foundation construction time to the reference construction time; Dynamically adjust the weight coefficients of the multi-objective optimization function based on environmental temperature and sand layer migration. The environmental temperature is described by an annual cycle sine function. The sand layer migration determines the adjustment value of the weight coefficient by the coupling effect of the wind-induced migration rate and the migration duration, combined with the non-uniformity of the influence of the wind direction angle on the pile-soil interface; Use the particle swarm algorithm to solve the multi-objective optimization function. Improve the particle swarm algorithm by dynamically adjusting the inertia weight and learning factor through environmental temperature and sand layer migration. Iteratively update the particle positions including the pile foundation diameter, pile foundation burial depth, and surface treatment method to obtain the optimal construction parameter combination that satisfies the uplift bearing capacity constraint and geometric constraint conditions.

5. The method according to claim 4, wherein The particle swarm optimization algorithm is used to solve the multi-objective optimization function, and the optimal construction parameter combination obtained includes: A dynamic inertia weight calculation formula is established based on the environmental coupling coefficient. The dynamic inertia weight calculation formula adopts the exponential difference form of the maximum inertia weight and the minimum inertia weight, and the change rate of the inertia weight is adjusted through the environmental coupling coefficient; Calculation formulas for the individual optimal solution learning factor and the global optimal solution learning factor in the form of exponential functions are established, and adaptive adjustment is performed through the ratio of the difference between the current particle fitness and the maximum fitness and the minimum fitness of the population; A default penalty function is constructed. The default penalty function includes a pullout bearing capacity constraint term and a geometric constraint term. The pullout bearing capacity constraint term adopts the square form of the difference between the actual pullout bearing capacity and the required pullout bearing capacity, and the geometric constraint term adopts the square form of the difference between the pile foundation diameter and the pile foundation burial depth and their constraint boundaries; Particles including the pile foundation diameter, the pile foundation burial depth, and the surface treatment method are encoded. The particle swarm is iteratively updated using the dynamic inertia weight and the adaptive learning factor. The default penalty function and the multi-objective optimization function are combined to calculate the particle fitness, and the optimal construction parameter combination that meets the constraint conditions is obtained through iterative optimization.

6. The method according to claim 1, characterized in that, Using the optimal construction parameter combination for pile foundation construction and performing dynamic fine-tuning includes: Pile foundation construction is carried out in each construction sub-area according to the optimal construction parameter combination. After every preset number of pile foundation constructions are completed, pullout bearing capacity tests are carried out to obtain actual pullout bearing capacity data; The time series analysis method is used to process the actual pullout bearing capacity data at multiple detection time points in the same construction sub-area to obtain the time-varying characteristics of the pullout bearing capacity; The actual pullout bearing capacity data is compared with the theoretical calculated value. When the deviation rate of the pullout bearing capacity exceeds the second preset threshold, the weight coefficient in the multi-objective optimization function is corrected based on the time-varying characteristics, and the updated optimal construction parameter combination is obtained by re-optimization calculation using the corrected multi-objective optimization function; During the construction process, the environmental temperature change and the sand layer migration situation are monitored in real time. When their changes exceed the preset range, the dynamic adjustment of the multi-objective optimization function is triggered.

7. An electronic device, characterized in that, Including: A processor; A memory for storing instructions executable by the processor; Wherein, the processor is configured to call the instructions stored in the memory to execute the method according to any one of claims 1 to 6.

8. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, the method according to any one of claims 1 to 6 is implemented.

Images