Production and manufacturing equipment and method for cement soil with multiple adjustable proportions

Abstract

The invention provides multi-proportion adjustable cement soil production and manufacturing equipment and a manufacturing method, and relates to the technical field of cement soil production, the multi-proportion adjustable cement soil production and manufacturing equipment comprises a mixer, the top of the mixer is integrally connected with a rotary bearing table, and the bottom of the mixer is integrally connected with a discharging pipe; wherein the rotary bearing table is provided with a rotary ring, the top of the rotary ring is integrally connected with a ring plate, the outer edge of the ring plate is provided with a tooth block, the tooth block is meshed with a gear arranged on a motor, and the motor is arranged on the rotary bearing table so as to drive the rotary ring to rotate through the meshing effect between the gear and the tooth block. Compared with the prior art, the invention has the following beneficial effects: 1, the unique combined design of the stirring plates and the stirring blades is adopted, so that the space occupation of the stirring structure in the mixer is obviously reduced; 2, the production process is greatly simplified, and a plurality of intermediate links are omitted; and thirdly, in combination with the training, tuning and testing of the XGBoost model, the optimal raw material proportioning scheme under different engineering conditions can be accurately predicted.

Description

Technical Field

[0001] The invention relates to the technical field of cement soil production, and in particular to a cement soil production device and a production method capable of adjusting multiple mix ratios. Background Art

[0002] In the field of construction engineering, cement soil is widely used in many projects such as road base, building foundation treatment, dam reinforcement, etc. With the development of the construction industry, the performance requirements of cement soil in different projects have become more diversified and refined, which makes it crucial to accurately adjust the cement soil ratio to meet specific engineering needs.

[0003] There are many ways to mix cement soil. When producing cement soil, the cement soil will be mixed and mixed in advance. A mixer with an automatic batching system is required, which makes the mixer very large. After mixing, the cement soil needs to be transferred to a conveyor truck and then transferred from the conveyor truck to the pouring point. There are many intermediate processes and many equipments. This is because ordinary mixers do not have an automatic proportioning system, and the proportioning and mixing of cement soil raw materials cannot be effectively and accurately controlled on site. In addition, the agitator of ordinary mixers is too large, which affects the mixing amount. There is a discharge pipe at the bottom of the mixer. The cement soil is viscous, the mixing is incomplete, the outlet is narrow, and the discharge and pouring are slow. As a result, ordinary mixers have many inconveniences in making cement soil. In addition, the production method of cement soil mainly relies on manual experience to control the raw material ratio. This method is difficult to ensure that the quality of cement soil produced each time is stable and meets the design requirements. Due to the lack of precise proportioning control, the produced cement soil may fluctuate in key performance indicators such as strength, durability, and impermeability. Summary of the invention

[0004] In view of the deficiencies in the above-mentioned prior art, the technical problem to be solved by the present invention is to provide a cement soil production equipment that can accurately control multiple ratios, realize intelligent production monitoring and adjustment, use data to optimize processes, ensure the stability of cement soil quality, and meet the needs of different projects. Furthermore, a method for producing cement soil production equipment that can adjust multiple ratios can be provided that can respond to environmental changes in real time and improve production efficiency.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: the present invention provides a cement soil production equipment with multiple ratios that can be adjusted, including a mixer, the top of the mixer is integrally connected to a rotating bearing platform, and the bottom is integrally connected to a feeding pipe; Among them, a rotating ring is provided on the rotating bearing platform, the top of the rotating ring is integrally connected to a ring plate, the outer edge of the ring plate is provided with a tooth block, the tooth block is meshed with a gear provided on the motor, and the motor is provided on the rotating bearing platform to drive the rotating ring to rotate through the meshing action between the gear and the tooth block.

[0006] In a preferred embodiment, a stirring plate is fixedly installed around the axis on the inner edge of the ring plate, the lower end of the stirring plate is located at the connection between the mixer and the discharge pipe, and the lower ends of the two stirring plates are integrally connected with stirring blades, the stirring blades are located inside the discharge pipe, and the two sides are in contact with the inner side surface of the discharge pipe; A sealing plate is provided at the bottom end of the discharge pipe, the side of the sealing plate is fixed by a locking structure, a weighing device is fixedly installed at the bottom, the weighing device display screen is located on the side, and a rotating bearing platform is located outside the weighing device and a cylinder is fixedly installed.

[0007] In a preferred solution, a sensor network is deployed on the weighing device, and the sensors include a temperature sensor, a pressure sensor, a torque sensor, a flow sensor, a concentration sensor, a temperature and humidity sensor, a wind speed sensor, and an air pressure sensor; A ring groove is provided on the rotating bearing platform, arc grooves are provided on both sides of the ring groove around the axis, balls are installed at the arc groove positions, ball ring grooves are provided on both sides of the rotating ring, and the balls are arranged in rolling contact with the ball ring grooves to reduce the rotational friction.

[0008] In the preferred solution, the top of the mixer is a vertical cylinder structure, and the rest is an oblique contraction structure, and the stirring plate is arranged to fit the inner wall of the mixer so that the cement soil gathers and falls; A protective tube is provided on the ring plate, and the inner wall of the protective tube is flush with the inner wall of the top of the mixer to prevent water from splashing out during the mixing stage; The locking structure includes a locking groove, a fixed block, a first movable shaft and a locking handle. A locking groove is opened on the side of the sealing plate, and a fixed block is provided on the outer side of the bottom of the discharge pipe. The fixed block is connected to the locking handle through the first movable shaft, and the locking handle is inserted into the locking groove.

[0009] In a preferred embodiment, the length of the locking handle is 10 cm to 15 cm, the first movable axis is arranged at a position 3 cm to 5 cm away from the end of the locking handle, and when the locking handle is locked, a rotational clearance is left between the end of the locking handle and the bottom of the lock groove; A clamping assembly is integrally provided on the side of the sealing plate, a bending fixing frame is provided on the outer side of the feeding pipe, the end of the bending fixing frame is connected to the clamping assembly through a second movable shaft, the extension port of the cylinder is arranged flush with the bottom of the weighing device, and the cylinder touches the ground in the retracted state.

[0010] In a preferred embodiment, the present invention further provides a method for producing the cement soil production equipment capable of adjusting the above-mentioned various mix ratios, comprising the following steps: S1, On-site construction coordination and feedback stage: The construction site is equipped with intelligent terminal devices, and the on-site data is uploaded to the cloud server according to the terminal devices. The cloud server processes the received data; S2, data collection and collation stage: collect historical cement soil production data under different working conditions, pre-process the historical cement soil production data, and store the pre-processed data in a distributed database; S3, model training and ratio prediction stage: select the XGBoost model, and perform feature engineering on numerical features based on the distributed database, input the data processed by feature engineering into the XGBoost model for iterative training, obtain the implementation environment data under the current working conditions through sensors, and input it into the trained XGBoost model to obtain the optimal raw material ratio plan; S4, real-time monitoring stage of production process: setting up an automated sample collection device and a sensor network, obtaining real-time collected data according to the automated sample collection device and the sensor network, and transmitting the real-time collected data to a cloud server for processing; Among them, the real-time collected data includes real-time data of cement-soil ratio and sample test results; S5, intelligent adjustment and optimization stage: The dynamic time warping algorithm is used to compare the real-time collected data with the ideal value, and the adjustment decision is generated according to the PSO-SVR hybrid algorithm to automatically adjust the production parameter adjustment plan.

[0011] In a preferred solution, step S2 specifically includes the following steps: S21. Multi-channel collection: Collect cement soil production data of various construction projects, including raw materials, production environment, geological conditions and performance test data; Prepare and test cement soil samples to obtain experimental data; S22, data cleaning and preprocessing: use density-based spatial clustering algorithm to identify and remove data outliers; divide data points into core points, boundary points and noise points; The missing values ​​are handled by using the imputation method based on singular value decomposition; Assume the data matrix is ,right Perform singular value decomposition ; in, and is an orthogonal matrix, is a diagonal matrix; Reconstruct the data matrix based on singular values ​​to determine missing values; Normalize the numerical data and map it to the [0,1] interval. The formula is: ; in, is the original data, and are the maximum and minimum values ​​in the data; The cleaned and preprocessed data is stored in a distributed database in a structured manner, and a data index is established.

[0012] In a preferred solution, step S3 specifically includes the following steps: S31, Feature selection and engineering: Select cement soil numerical features based on the database, including raw material indicators, environmental factors and production process parameters; Among them, raw material indicators include cement strength grade, soil plasticity index, and additive dosage ratio; Environmental factors include temperature and humidity ranges; Production process parameters include mixing time and feeding sequence; The cement soil numerical features are converted into numerical vectors using one-hot encoding, and the numerical features are standardized by feature scaling: ; in, is the mean, is the standard deviation; and perform polynomial feature expansion; S32, model construction and training: the processed feature data is divided into training set, validation set and test set according to 7:1:2, and the cement soil performance indicators of the training set and validation set are marked, and the marked values ​​are the corresponding cement soil performance indicators.

[0013] Initialize the XGBoost model. In each iteration, the improved Huber loss function is used as the objective function. The formula is: ; in, is the true value, is the predicted value; Calculate the gradient of the current model loss function with respect to the predicted value and the second derivative , the formula is: ; ; A new decision tree fitting residual is constructed based on the gradient information and splitting criteria. Pre-pruning and post-pruning strategies are used during construction. Pre-pruning limits the growth of the decision tree by setting a threshold. Post-pruning prunes the tree according to the evaluation index after the decision tree is built to prevent overfitting. S33, Model evaluation and early stopping: After each iteration, the model performance is evaluated through the validation set and the root mean square error is calculated. , mean absolute error ; If the root mean square error and mean absolute error do not improve on the validation set for 10 consecutive rounds, stop training to prevent overfitting; S34, model tuning and testing: Use grid search or random search to traverse the hyperparameter combination training model, evaluate the performance on the validation set, select the optimal hyperparameter combination, and use the test set to test the tuned model to evaluate the generalization ability; S35, optimal ratio prediction: collect current project information, including project type, design strength requirements, and real-time environmental data of the construction site, and input them into the trained XGBoost model. After calculation and reasoning, the XGBoost model outputs the optimal raw material ratio under the specific conditions to obtain the prediction results. Based on the prediction results of the XGBoost model trained under multiple different initialization conditions, the final optimal ratio prediction value is obtained by weighted average. The formula is: ; in, is the final optimal raw material ratio prediction value, It is The matching results predicted by the model are It is The weight of the model, is the number of models.

[0014] In a preferred solution, step S4 specifically includes the following steps: S41, sensor network deployment and data collection: obtain real-time data of cement-soil ratio through the sensor network, and transmit the collected data to the cloud server in real time through the industrial Internet of Things; S42, sample collection and instant testing: collect multiple samples through an automated sample collection device, obtain sample test results, and transmit the sample test results to a cloud server in the form of digital signals; S43. Data integration and analysis: The cloud server integrates the real-time data of cement-soil ratio and sample test data according to time series and production batches, uses a sliding window algorithm to calculate the statistical characteristics of the data over a period of time, and adjusts the decision.

[0015] In a preferred solution, step S5 specifically includes the following steps: S51, deviation identification and analysis: using dynamic time warping algorithm to compare real-time collected data with ideal values, if it is lower than the threshold, the deviation is determined and analyzed; S52, PSO-SVR hybrid algorithm to adjust decision generation: using historical production data in a distributed database as input, quality indicators as output and dividing into training sets and test sets, selecting radial basis kernel function to build a support vector regression model, and using cross-validation to adjust penalty parameters and kernel function parameters; Determine the particle swarm size, dimension, initial position and speed, and set the PSO algorithm parameters; Iteratively calculate the fitness value of particles, update individual and global optimal positions, and optimize particles according to the speed and position update formula; When the maximum number of iterations is reached or the stopping condition is met, the adjustment plan corresponding to the global optimal position is output; S53, automatic adjustment execution: The cloud server sends instructions to the production equipment according to the adjustment plan to adjust the raw material delivery amount, stirring motor speed and time, and stirring temperature.

[0016] The present invention provides a cement soil production equipment and method with multiple ratios that can be adjusted. Compared with the prior art, it has the following beneficial effects: First, the present invention adopts a unique combination design of stirring plates and stirring blades, which significantly reduces the space occupied by the stirring structure inside the mixer, and is optimized for the high viscosity of cement soil, so that efficient stirring can be continuously performed during the material discharge process. The stirring plates and stirring blades not only promote the uniform mixing of materials, but also effectively scrape off the cement soil attached to the mixer and the inner wall of the discharge pipe to ensure that it falls smoothly and avoids residue. Compared with traditional mixing devices, the present invention eliminates dead corners in stirring and achieves more complete and complete material mixing. In addition, the large-diameter discharge pipe design ensures rapid material discharge, further improving production efficiency; Second, the invention also has the function of mixing cement soil on site and automatically lifting and guiding the material for pouring, which greatly simplifies the production process and eliminates multiple intermediate links. The equipment adopts the most basic but efficient weighing and proportioning method to make cement soil, combined with a simple and reliable locking and sealing mechanism, ensuring that the entire operation process is simple and efficient; Third, by combining the training, tuning and testing of the XGBoost model, it can accurately predict the optimal raw material ratio under different engineering conditions, ensuring that the produced cement soil meets the specific needs of various projects, greatly improving the adaptability of the product and avoiding engineering quality problems caused by improper ratios; Fourth, in the real-time monitoring part of the production process, with the help of an all-round sensor network to collect key data, combined with sample collection and real-time detection, and then using the sliding window algorithm to analyze the integrated data, it is possible to promptly discover abnormal fluctuations and trend changes in the production process, effectively ensuring the stability of the production process and ensuring that the quality of each batch of cement soil is highly consistent. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The present invention will be further described below in conjunction with the accompanying drawings and embodiments: Figure 1 It is a schematic diagram of the cross-sectional structure of the present invention; Figure 2 It is a three-dimensional structural schematic diagram of the present invention; Figure 3 The present invention is a schematic diagram of the three-dimensional structure Figure 4 It is the main structural diagram of the process of the present invention; Figure 5 It is a structural diagram of an electronic device of the present invention.

[0018] In the figure: mixer 1, rotating supporting platform 2, feeding pipe 3, rotating ring 4, ring plate 5, gear block 6, gear 7, motor 8, stirring plate 9, stirring blade 10, sealing plate 11, weighing device 12, cylinder 13, ball 14, protective tube 15, locking groove 16, fixing block 17, first movable shaft 18, locking handle 19, clamping assembly 20, bending fixing frame 21, second movable shaft 22. DETAILED DESCRIPTION

[0019] In order to better understand the purpose, structure and function of the present invention, the embodiments and features in the embodiments of the present application can be combined with each other without conflict. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0020] Example 1 like Figures 1 to 3 As shown, this embodiment provides a cement soil production equipment with multiple ratios that can be adjusted, including a mixer 1, the top of the mixer 1 is integrally connected to a rotating bearing platform 2, and the bottom is integrally connected to a feed pipe 3 for receiving and mixing raw materials of various ratios; Among them, a rotating ring 4 is movably installed on the rotating supporting platform 2, and the top of the rotating ring 4 is integrally connected to a ring plate 5. A gear block 6 is provided on the outer edge of the ring plate 5. The gear block 6 is meshed with a gear 7 fixed on a motor 8. The motor 8 is fixed on the rotating supporting platform 2 to form a power transmission system, so that when the motor 8 is started, the rotating ring 4 and the components above it can be driven to rotate through the meshing action between the gear 7 and the gear block 6.

[0021] The preferred solution is Figure 1 , 2 As shown, a stirring plate 9 is fixedly installed around the axis on the inner edge of the ring plate 5, and the lower end of the stirring plate 9 is located at the connection between the mixer 1 and the discharge pipe 3, and the lower ends of the two stirring plates 9 are integrally connected with stirring blades 10, and the stirring blades 10 are located inside the discharge pipe 3, and the two sides are arranged in contact with the inner side surface of the discharge pipe 3 to ensure that the materials can be fully mixed during the rotation process, and the materials attached to the inner wall of the mixer 1 and the discharge pipe 3 can be scraped off to prevent residue; A sealing plate 11 is movably installed at the bottom of the discharge pipe 3, and the side of the sealing plate 11 is fixed by a locking structure. A weighing device 12 is fixedly installed at the bottom, and the display screen of the weighing device 12 is located on the side. The rotating supporting platform 2 is located outside the weighing device 12 and a cylinder 13 is fixedly installed. The cylinder 13 provides a lifting force, so that the whole device can automatically lift the material for pouring after the mixing is completed, thereby simplifying the operation process.

[0022] The preferred solution is Figure 2 ,3 As shown, a sensor network is deployed on the weighing device 12, and the sensors include a temperature sensor, a humidity sensor, and a pressure sensor; A ring groove is provided on the rotating bearing platform 2, arc grooves are provided on both sides of the ring groove around the axis, balls 14 are installed at the arc groove positions, ball ring grooves are provided on both sides of the rotating ring 4, and the balls 14 are arranged in rolling contact with the ball ring grooves to reduce rotational friction, improve rotational efficiency, and extend the service life of the equipment.

[0023] The preferred solution is Figure 2 , 3 As shown, the top of the mixer 1 is a vertical cylinder structure, and the rest is an oblique contraction structure. The stirring plate 9 is arranged to fit the inner wall of the mixer 1 to facilitate the gathering and falling of cement soil, promote the fluidity and uniformity of the material in the mixer, and thus improve the mixing effect; A protective tube 15 is provided on the ring plate 5, and the inner wall of the protective tube 15 is flush with the inner wall of the top of the mixer 1 to prevent water from splashing out in the initial stage of mixing, keep the working environment clean, and avoid material waste; The locking structure includes a locking groove 16, a fixed block 17, a first movable shaft 18 and a locking handle 19. The locking groove 16 is opened on the side of the sealing plate 11, and the fixed block 17 is fixedly installed on the outer side of the bottom of the discharge pipe 3. The fixed block 17 is movably connected to the locking handle 19 through the first movable shaft 18. The locking handle 19 is inserted into the locking groove 16 for easy manual rotation of the lock, so as to realize the rapid opening and closing of the sealing plate 11 and ensure the safety and reliability of the discharge process.

[0024] The preferred solution is Figure 2 , 3 As shown, the length of the locking handle 19 is 10 cm to 15 cm, and the first movable shaft 18 is set at a position 3 cm to 5 cm away from the end of the locking handle 19. When the locking handle 19 is locked, a rotation gap is left between the end of the locking handle 19 and the bottom of the lock groove 16, which saves effort while ensuring the stability and durability of the locking structure; A clamping assembly 20 is integrally provided on the side of the sealing plate 11, and a bending fixing frame 21 is provided on the outer side of the discharge pipe 3. The end of the bending fixing frame 21 is connected to the clamping assembly 20 by a second movable shaft 22, ensuring that the sealing plate 11 rotates smoothly without jamming, which is convenient for operators to perform opening and closing operations, and at the same time ensures a tight fit between the discharge pipe 3 and the sealing plate 11. The extended port of the cylinder 13 is flush with the bottom of the weighing device 12, and the cylinder 13 touches the ground in a retracted state, ensuring that the equipment has good stability in a stationary state. When discharge is required, the extension of the cylinder 13 can effectively lift the entire device, providing sufficient space for discharge, thereby improving work efficiency.

[0025] Working principle: According to the type of cement soil to be poured, and through AI-assisted feeding to determine the proportion of each raw material, the equipment is placed directly above or near the pouring point for mixing nearby; Rotate the locking handle 19 and insert the locking handle 19 into the locking groove 16. The sealing plate 11 closes the bottom end of the feeding pipe 3. The cylinder 13 contracts and is grounded. When each raw material ratio is fed in, it is directly controlled accurately according to the display of the weighing device 12. After the feeding is completed, the motor 8 drives the gear 7 to rotate, and the gear 7 engages the tooth block 6 to drive the ring plate 5 to rotate. Because the ball 14 rolls, the rotating ring 4 and the ring plate 5 rotate smoothly with little friction. The ring plate 5 drives the stirring plate 9 and the stirring blade 10 in the mixer 1 and the feeding pipe. 3, cement soil is mixed and stirred. After the mixing is completed, the cylinder 13 is extended to lift the entire device, leaving space under the weighing device 12, and a pouring pipe with a collecting hood is placed under the feeding pipe 3. The locking handle 19 is opened, and the sealing plate 11 is turned over. After the cement soil falls, the material is poured. When pouring, the rotating state of the stirring plate 9 and the stirring blade 10 should be maintained. The cement soil on the inner wall of the mixer 1 and the feeding pipe 3 can be scraped together and fall along the stirring plate 9 and the stirring blade 10 to achieve sufficient feeding.

[0026] Example 2 Further illustrate with reference to Example 1, Figure 4 As shown, this embodiment also provides a method for manufacturing cement soil production equipment that can adjust the multiple ratios described in Example 1, which realizes the full process automation and intelligent management from data collection, model training to real-time monitoring of the production process and intelligent adjustment and optimization.

[0027] In this method, data drives model training, models guide the production process, real-time monitoring ensures production quality, intelligent adjustment and optimization improve production efficiency and product quality, and on-site construction feedback promotes continuous improvement of the system. It not only improves production flexibility and efficiency, but also ensures the consistency and reliability of product quality, and can adapt to production needs of different scales.

[0028] The specific steps include: S1. On-site construction coordination and feedback stage: The construction site is equipped with intelligent terminal devices, and the on-site data is uploaded to the cloud server according to the terminal devices. The cloud server processes the received data. The specific steps are as follows: S11. Intelligent terminal devices are equipped at the construction site, and construction workers receive production information sent by the cloud server, including mix details, quality inspection results, construction precautions, and feedback on the construction site conditions, including pouring progress and pouring problems.

[0029] S12. The terminal device uploads the field data to the cloud server, and the cloud runs XGBoost and PSO-SVR algorithms to analyze and process the data. The cloud server generates adjusted production parameters and construction optimization suggestions based on the field data and algorithm analysis results, including adjusting the cement-soil ratio according to the temperature and humidity of the construction site and adjusting the production plan according to the pouring progress. The adjusted production parameters and construction optimization suggestions are then pushed back to the terminal device to provide technical support.

[0030] S13. Construction personnel carry out construction operations according to the information provided by the terminal equipment, and use cement soil for pouring in strict accordance with production standards and construction specifications. During the construction process, the quality of the poured cement soil is tracked and tested, such as testing its strength and density through core sampling, and evaluating the performance of the cement soil in the actual project. The construction quality inspection data is fed back to the cloud server, and the cloud server further optimizes the production model and adjustment strategy based on the feedback data. For example, if it is found that the strength of a batch of cement soil at the construction site does not meet expectations, the system will analyze the reasons and adjust the subsequent production ratio or process parameters to form a closed-loop optimization management from production to construction, continuously improve the production quality and construction effect of cement soil, and ensure the quality and safety of the project.

[0031] S2, data collection and collation stage: collect historical cement soil production data under different working conditions, pre-process the historical cement soil production data, and store the pre-processed data in a distributed database. The specific steps are as follows: S21. Multi-channel data collection: Cooperate with construction companies and scientific research institutions to collect cement soil production data for various construction projects such as bridges, roads, and building foundations, covering raw materials such as cement chemical composition, soil mineral composition and physical properties, additive composition and efficacy, production environment, including construction site temperature and humidity, atmospheric pressure, geological conditions, and performance test data such as compressive strength, shear strength, and permeability coefficient.

[0032] At the same time, different combinations of variables such as mixing speed, mixing time, and curing conditions were set in the laboratory to make and test cement soil samples to obtain experimental data.

[0033] S22. Data cleaning and preprocessing: Use the density-based spatial clustering algorithm (DBSCAN) to identify and remove outliers in the data set caused by measurement errors or special anomalies. DBSCAN divides data points into core points, boundary points and noise points by defining concepts such as core points, density reachability and density connectivity. Noise points are outliers.

[0034] The missing values ​​are handled by using a singular value decomposition (SVD)-based imputation method. Assume that the data matrix is ,right Perform singular value decomposition ,in and is an orthogonal matrix, It is a diagonal matrix. The missing values ​​are estimated by reconstructing the data matrix by retaining some singular values, thereby improving data completeness.

[0035] Normalization: Normalize the numerical data and map it to the interval [0,1]. The formula is , is the original data, and It is the maximum and minimum value in the data, which can eliminate the difference in dimension and numerical range and improve the accuracy of subsequent analysis.

[0036] The cleaned and pre-processed data is stored in a distributed database in a structured manner, and data indexes are established to facilitate query and call, and regular maintenance and updates are performed to ensure data timeliness and integrity.

[0037] S3, model training and ratio prediction stage: select the XGBoost model, and perform feature engineering on numerical features based on the distributed database. Input the data processed by feature engineering into the XGBoost model for iterative training. Obtain the implementation environment data under the current working conditions through sensors and input it into the trained XGBoost model to obtain the optimal raw material ratio scheme. The specific steps are as follows: S31. Feature selection and engineering: Select the features of cement soil numerical characteristics from the database, such as raw material indicators such as cement strength grade, soil plasticity index, additive dosage ratio, environmental factors such as temperature and humidity range, and production process parameters such as mixing time and feeding sequence.

[0038] Convert categorical features into numerical vectors using one-hot encoding, and standardize feature scaling for numerical features ; in, is the mean, is the standard deviation; At the same time, polynomial feature expansion is performed to explore the potential relationships of features.

[0039] S32. Model construction and training: Divide the processed feature data into training set, validation set and test set according to 7:1:2, and mark the cement soil performance indicators of the training set and validation set. The marked values ​​are the corresponding cement soil performance indicators.

[0040] Initialize the XGBoost model and set the number of trees to , Maximum Depth , learning rate etc. initial parameters.

[0041] In each round of iteration, the improved Huber loss function is used as the objective function, and the formula is: ; in, is the true value, is the predicted value; Specifically, compared with the traditional mean square error loss function, the Huber loss function is more robust to outliers and can prevent the model from being overly affected by a few abnormal samples. It is a hyperparameter used to balance the influence of squared loss and absolute loss.

[0042] Calculate the gradient of the current model loss function with respect to the predicted value and the second derivative , the formula is: ; ; Specifically, it is used to build a new decision tree to fit the residual of the current model. In XGBoost, the structure and parameters of the new decision tree are determined based on the gradient and second-order derivative information.

[0043] When constructing a decision tree, an improved splitting criterion based on information gain ratio and Gini coefficient is used. The traditional information gain ratio formula is: ; in, It is a feature For the dataset The information gain of It is a dataset The information entropy and Gini coefficient formula are as follows: ; in, is the number of categories in the dataset, It is The proportion of class samples in the dataset.

[0044] The improved splitting criterion combines the advantages of both and introduces a balance factor ( ), the new splitting criterion formula is: ; Specifically, by selecting the features and splitting points that maximize the criterion value, we can more effectively select the features and splitting points that are most discriminative of the target variable, thus improving the classification and regression capabilities of the decision tree. The value of can optimize the splitting effect.

[0045] According to the gradient information and splitting criteria, a new decision tree is constructed. The goal of this decision tree is to fit the residual of the current model. When constructing a decision tree, a combination of pre-pruning and post-pruning strategies is adopted. Pre-pruning limits the growth of the decision tree by setting some thresholds such as the minimum number of samples and the maximum depth; post-pruning prunes the tree after the decision tree is built, based on evaluation indicators such as the error on the validation set, to prune those branches that do not improve the overall performance much, to prevent overfitting.

[0046] S33, Model evaluation and early stopping: After each round of iteration, the validation set is used to evaluate the model performance and calculate the root mean square error or mean absolute error If there is no improvement in RMSE or MAE on the validation set for 10 consecutive rounds, training is stopped to prevent overfitting.

[0047] S34. Model tuning and testing: Use grid search or random search methods to tune the model's hyperparameters, train the model under different hyperparameter combinations, and evaluate the performance on the validation set. By comparing the evaluation indicators under different combinations, select the hyperparameter combination with the best performance. Finally, use the test set to test the tuned model to evaluate the model's generalization ability on unseen data to ensure that the model can accurately predict the performance of cement soil under different engineering conditions.

[0048] S35. Optimal mix ratio prediction: Before actual production, collect detailed information about the current project, including project types such as large bridge construction and small building foundation treatment, design strength requirements such as compressive strength and shear strength indicators, and real-time environmental data of the construction site such as temperature, humidity, and geological conditions. Input this information into the trained XGBoost model. After calculation and reasoning, the model outputs the optimal raw material ratio under the specific conditions, providing accurate ratio guidance for production and ensuring that the cement soil produced meets the project requirements. At the same time, in order to improve the reliability of the prediction, the idea of ​​ensemble learning is adopted, combining the prediction results of the XGBoost model trained under multiple different initialization conditions, and obtaining the final optimal ratio prediction value by weighted average. The formula is: ; in, is the final optimal raw material ratio prediction value, It is The matching results predicted by the model are It is The weight of the model, is the number of models. Weights The weight is determined based on the performance of each model on the validation set. The better the performance, the higher the weight of the model. By integrating the prediction results of multiple models, the variance of the model can be reduced and the stability and accuracy of the prediction can be improved.

[0049] S4, real-time monitoring stage of production process: set up an automated sample collection device and a sensor network, obtain real-time collection data based on the automated sample collection device and the sensor network, and transmit the real-time collection data to the cloud server for processing.

[0050] Among them, the real-time collected data includes real-time data of cement-soil ratio and sample test results; S41. Sensor network deployment and data collection: Install a variety of sensors on cement soil production equipment, including temperature, pressure, torque, flow, concentration, temperature and humidity, wind speed, and air pressure sensors to build a comprehensive sensor network; Among them, temperature sensors, pressure sensors and torque sensors are installed in the mixer 1, the rotating support platform 2 and the feed pipe 3 respectively. The temperature sensor monitors the temperature change of the material during the stirring process in real time, the pressure sensor measures the pressure inside the stirring device, and the torque sensor reflects the torque of the stirring motor. These data can be used to evaluate the uniformity of stirring and the viscosity change of the material.

[0051] A flow sensor and a concentration sensor are installed on the weighing device 12 to measure the delivery flow and concentration of the raw materials in real time to ensure accurate delivery of the raw materials.

[0052] An environmental monitoring station is set up at the production site, equipped with temperature and humidity sensors, wind speed sensors and air pressure sensors to collect environmental meteorological data and monitor changes in the production environment.

[0053] The real-time data of cement-soil ratio is obtained, and the collected data is transmitted to the cloud server in real time through the Industrial Internet of Things (IIoT) technology.

[0054] S42. Sample collection and instant testing: After each mixing is completed, the automated sample collection device collects multiple samples from the mixed cement soil to ensure the representativeness of the samples. Use the equipped instant testing equipment, such as a portable X-ray fluorescence spectrometer (XRF), to detect the chemical composition of the cement soil and analyze the content of various elements in it; use an ultrasonic detector to measure the density of its internal structure and evaluate the uniformity of the cement soil; use a micro-pressure testing machine to test the early strength and obtain the mechanical properties of the cement soil in the early stage. These testing equipment transmits the test results to the cloud server in the form of digital signals, providing data support for real-time monitoring of cement soil quality.

[0055] S43. Data integration and analysis: The cloud server integrates the data collected by the sensor and the sample test data, and associates them according to the time series and production batches. Use the sliding window algorithm to analyze the real-time data, set the appropriate window size and step length, and calculate the statistical characteristics of the data over a period of time in real time, such as the mean, standard deviation, maximum value, minimum value, etc. By analyzing these statistical characteristics, abnormal fluctuations and trend changes in the data can be discovered in a timely manner, such as sudden changes in the concentration of raw materials, abnormal increases or decreases in the stirring temperature, etc., to provide a basis for subsequent adjustment decisions and ensure the stability of the production process and the consistency of product quality.

[0056] S5, intelligent adjustment and optimization stage: the dynamic time warping algorithm is used to compare the real-time collected data with the ideal value, and the adjustment decision is generated according to the PSO-SVR hybrid algorithm to automatically adjust the production parameter adjustment plan; S51, Deviation Identification and Analysis (DTW Algorithm): The cloud server compares the real-time collected data with the ideal values ​​set according to the predicted ratio and production standards.

[0057] The dynamic time warping (DTW) algorithm is used to calculate the similarity between the actual data sequence and the ideal data sequence.

[0058] The DTW algorithm finds the optimal matching path between two time series and calculates the sum of the distances of corresponding points on the path as a similarity measure.

[0059] When the similarity is lower than the set threshold, it is determined that a deviation has occurred. The types of deviations are further analyzed, including raw material ratio deviation, abnormal mixing temperature, insufficient mixing time, etc. Evaluate the degree of deviation, which is divided into mild, moderate and severe deviations; analyze the possible impact of deviations on cement soil quality, including reduced strength, reduced durability, deterioration of construction performance, etc., to provide detailed information for subsequent adjustment decisions.

[0060] S52, PSO-SVR hybrid algorithm adjustment decision generation SVR model construction: Use the production parameters of historical production data as input features, cement soil quality indicators as output labels, and divide the training set and test set. Select radial basis kernel function Build a support vector regression model and adjust the penalty parameters through training set training , kernel function parameters , cross-validation was used to evaluate the performance and select the optimal parameter combination, and a regression model of production parameters and cement soil quality was established to predict quality changes.

[0061] PSO algorithm initialization: determine the particle swarm size, particle dimension (i.e. the corresponding production parameter adjustment quantity), particle initial position and speed.

[0062] Set PSO algorithm parameters, including inertia weight , learning factor ( and ), maximum number of iterations.

[0063] PSO iterative optimization: The fitness value of each particle is calculated in each round of iteration, and the particle position (i.e., the production parameter adjustment plan) is substituted into the trained SVR model to predict the cement soil quality, and the fitness is calculated based on the difference between the predicted and target quality.

[0064] According to the fitness, the individual optimal position pbest of the particle and the global optimal position gbest are updated, and the speed update formula is: ; And the position update formula: ; in, , is a random number between (0,1); Update particle speed and position to make particles move in a better direction, and continuously search for better production parameter adjustment solutions.

[0065] Determination of the optimal adjustment plan: When the PSO algorithm reaches the maximum number of iterations or meets the stopping condition (i.e., the particle position changes little and the fitness value converges), the production parameter adjustment plan corresponding to the global optimal position is output as an adjustment decision for the current production process to ensure that the adjustment plan can effectively optimize the production process and improve the quality of cement soil.

[0066] S53, automatic adjustment execution: According to the optimized adjustment plan, the cloud server automatically sends control instructions to the production equipment. Further, by adjusting the valve opening on the raw material delivery pipeline, the delivery amount of raw materials is accurately controlled to ensure the accurate raw material ratio; the speed and mixing time of the mixing motor are adjusted to change the mixing process and improve the mixing effect; the heating or cooling device is controlled to adjust the temperature during the mixing process to meet the production process requirements. Through automatic adjustment execution, the quality of cement soil in subsequent production batches is ensured to meet the standard requirements, and intelligent control of the production process is realized.

[0067] Example 3 Further explanation is given in conjunction with Examples 1 and 2. Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include: A memory, a processor, and a computer program stored in the memory and executable on the processor.

[0068] When the processor executes the program, the method for manufacturing the cement soil production equipment can be adjusted with multiple ratios provided in the above embodiment.

[0069] Furthermore, the electronic device further comprises: Communication interface, used for communication between memory and processor. Memory is used to store computer programs that can be run on the processor.

[0070] The memory may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0071] If the memory, processor and communication interface are implemented independently, the communication interface, memory and processor can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. The processor may include one or more processing units, for example, the processor may include an application processor (AP), an application specific integrated circuit (ASIC), a modem processor, a graphics processor (CPU), an image signal processor (ISP), a controller, a memory, a video codec, a digital signal processor (DSP), a baseband processor, and / or a neural-network processing unit (NPU). Among them, different processing units may be independent devices or integrated into one or more processors. Among them, the controller may be a neural center and a command center. The controller may generate an operation control signal according to the instruction opcode and the timing signal to complete the control of fetching and executing instructions. A memory may also be provided in the processor for storing instructions and data. In some embodiments, the memory in the processor is a high-speed cache memory. The memory may store instructions or data that have just been used or circulated by the processor. If the processor needs to use the instruction or data again, it may be directly called from the memory. Repeated access is avoided, the waiting time of the processor is reduced, and the efficiency of the system is improved.

[0072] A visualization module is used to display images, videos, etc. The visualization module may include a display panel, which may be a liquid crystal display (LCD), an organic light-emitting diode (OLED), an active-matrix organic light-emitting diode or an active-matrix organic light-emitting diode (AMOLED), a flexible light-emitting diode (FLED), Miniled, MicroLed, Micro-oLed, a quantum dot light-emitting diode (QLED), etc.

[0073] Optionally, in a specific implementation, if the memory, processor and communication interface are integrated on a chip, the memory, processor and communication interface can communicate with each other through an internal interface.

[0074] The embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for manufacturing a cement soil production device that can adjust the above-mentioned multiple mix ratios.

[0075] The embodiment of the present application also provides a computer program product, which can run computer instructions. When the computer instructions are executed by a processor, a method for manufacturing a cement soil production device that can adjust multiple mix ratios as described above is implemented.

[0076] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or N wirings (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways as necessary and then storing it in a computer memory.

[0077] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A cement soil production device capable of adjusting multiple mix ratios, comprising a mixer (1), wherein the top of the mixer (1) is integrally connected to a rotating bearing platform (2), and the bottom of the mixer (1) is integrally connected to a feed pipe (3); It is characterized in that A rotating ring (4) is provided on the rotating bearing platform (2), the top of the rotating ring (4) is integrally connected to a ring plate (5), a tooth block (6) is provided on the outer edge of the ring plate (5), the tooth block (6) is meshed with a gear (7) provided on a motor (8), and the motor (8) is provided on the rotating bearing platform (2) so as to drive the rotating ring (4) to rotate through the meshing action between the gear (7) and the tooth block (6).

2. The cement soil production equipment according to claim 1, wherein the multiple mix ratios can be adjusted, A stirring plate (9) is fixedly mounted on the inner edge of the ring plate (5) around the axis, the lower end of the stirring plate (9) is located at the connection between the mixer (1) and the discharge pipe (3), and the lower ends of the two stirring plates (9) are integrally connected to a stirring blade (10), the stirring blade (10) is located inside the discharge pipe (3), and the two sides of the stirring blade (10) are arranged in contact with the inner side surface of the discharge pipe (3); A sealing plate (11) is provided at the bottom end of the feed tube (3), the side of the sealing plate (11) is fixed by a locking structure, a weighing device (12) is fixedly installed at the bottom, a display screen of the weighing device (12) is located on the side, and a cylinder (13) is fixedly installed on the rotating support platform (2) outside the weighing device (12).

3. The cement soil production equipment according to claim 2, wherein the various mix ratios can be adjusted, A sensor network is deployed on the weighing device (12), and the sensors include a temperature sensor, a pressure sensor, a torque sensor, a flow sensor, a concentration sensor, a temperature and humidity sensor, a wind speed sensor, and an air pressure sensor; The rotating bearing platform (2) is provided with an annular groove, arc grooves are provided on both sides of the annular groove around the axis, and balls (14) are installed at the arc groove positions. Ball annular grooves are provided on both sides of the rotating ring (4), and the balls (14) are arranged in rolling contact with the ball annular grooves to reduce rotational friction.

4. The equipment for producing soil cement with adjustable mix ratios according to claim 3, characterized in that: The top of the mixer (1) is a vertical cylindrical structure, and the rest of the structure is an oblique contraction structure. The stirring plate (9) is arranged to fit and contact the inner wall of the mixer (1) so that the cement soil can gather and fall. A protective tube (15) is provided on the ring plate (5), and the inner wall of the protective tube (15) is arranged flush with the inner wall of the top of the mixer (1) to prevent water from splashing out during the mixing stage; The lock structure comprises a lock slot (16), a fixing block (17), a first movable shaft (18) and a locking handle (19); the lock slot (16) is provided on the side of the sealing plate (11); a fixing block (17) is provided on the outer side of the bottom of the discharge tube (3); the fixing block (17) is connected to the locking handle (19) via the first movable shaft (18); and the locking handle (19) is inserted into the lock slot (16).

5. The cement soil production equipment according to claim 4, wherein the various mix ratios can be adjusted. The locking handle (19) is 10 cm to 15 cm long, the first movable shaft (18) is arranged at a position 3 cm to 5 cm away from the end of the locking handle (19), and when the locking handle (19) is in a locked state, a rotational clearance is left between the end of the locking handle (19) and the bottom of the locking groove (16); A clamping assembly (20) is integrally provided on the side of the sealing plate (11), a bending fixing frame (21) is provided on the outside of the feeding tube (3), an end of the bending fixing frame (21) is connected to the clamping assembly (20) via a second movable shaft (22), an extension opening of the cylinder (13) is arranged flush with the bottom of the weighing device (12), and the cylinder (13) touches the ground in a retracted state.

6. A method for producing a cement soil production device with multiple mix ratios adjustable as claimed in any one of claims 1 to 5, characterized in that: The following steps are involved: S1, On-site construction coordination and feedback stage: The construction site is equipped with intelligent terminal devices, and the on-site data is uploaded to the cloud server according to the terminal devices. The cloud server processes the received data; S2, data collection and collation stage: collect historical cement soil production data under different working conditions, pre-process the historical cement soil production data, and store the pre-processed data in a distributed database; S3, model training and ratio prediction stage: select the XGBoost model, and perform feature engineering on numerical features based on the distributed database, input the data processed by feature engineering into the XGBoost model for iterative training, obtain the implementation environment data under the current working conditions through sensors, and input it into the trained XGBoost model to obtain the optimal raw material ratio plan; S4, real-time monitoring stage of production process: setting up an automated sample collection device and a sensor network, obtaining real-time collected data according to the automated sample collection device and the sensor network, and transmitting the real-time collected data to a cloud server for processing; Among them, the real-time collected data includes real-time data of cement-soil ratio and sample test results; S5, intelligent adjustment and optimization stage: The dynamic time warping algorithm is used to compare the real-time collected data with the ideal value, and the adjustment decision is generated according to the PSO-SVR hybrid algorithm to automatically adjust the production parameter adjustment plan.

7. The method for producing the equipment for producing cement soil with multiple ratios adjustable according to claim 6, characterized in that: Step S2 specifically includes the following steps: S21. Multi-channel collection: Collect cement soil production data of various construction projects, including raw materials, production environment, geological conditions and performance test data; Prepare and test cement soil samples to obtain experimental data; S22. Data cleaning and preprocessing: Use density-based spatial clustering algorithms to identify and remove data outliers; Divide data points into core points, boundary points and noise points; The missing values ​​are handled by using the imputation method based on singular value decomposition; Assume the data matrix is ,right Perform singular value decomposition ; in, and is an orthogonal matrix, is a diagonal matrix; Reconstruct the data matrix based on singular values ​​to determine missing values; Normalize the numerical data and map it to the [0,1] interval. The formula is: ; in, is the original data, and are the maximum and minimum values ​​in the data; The cleaned and preprocessed data is stored in a distributed database in a structured manner, and a data index is established.

8. The method for producing the equipment for producing cement soil with adjustable multiple mix ratios according to claim 7, characterized in that: Step S3 specifically includes the following steps: S31, Feature selection and engineering: Select cement soil numerical features based on the database, including raw material indicators, environmental factors and production process parameters; Among them, raw material indicators include cement strength grade, soil plasticity index, and additive dosage ratio; Environmental factors include temperature and humidity ranges; Production process parameters include mixing time and feeding sequence; The cement soil numerical features are converted into numerical vectors using one-hot encoding, and the numerical features are standardized by feature scaling: ; in, is the mean, is the standard deviation; and perform polynomial feature expansion; S32, model construction and training: the processed feature data is divided into a training set, a validation set and a test set according to a ratio of 7:1:2, and the cement soil performance indicators of the training set and the validation set are marked, and the marked values ​​are the corresponding cement soil performance indicators; Initialize the XGBoost model. In each iteration, the improved Huber loss function is used as the objective function. The formula is: ； in, is the true value, is the predicted value; Calculate the gradient of the current model loss function with respect to the predicted value and the second derivative , the formula is: ； ； A new decision tree fitting residual is constructed based on the gradient information and splitting criteria. Pre-pruning and post-pruning strategies are used during construction. Pre-pruning limits the growth of the decision tree by setting a threshold. Post-pruning prunes the tree according to the evaluation index after the decision tree is built to prevent overfitting. S33, Model evaluation and early stopping: After each iteration, the model performance is evaluated through the validation set and the root mean square error is calculated. , mean absolute error ; If the root mean square error and mean absolute error do not improve on the validation set for 10 consecutive rounds, stop training to prevent overfitting; S34, model tuning and testing: Use grid search or random search to traverse the hyperparameter combination training model, evaluate the performance on the validation set, select the optimal hyperparameter combination, and use the test set to test the tuned model to evaluate the generalization ability; S35, optimal ratio prediction: collect current project information, including project type, design strength requirements, and real-time environmental data of the construction site, and input them into the trained XGBoost model. After calculation and reasoning, the XGBoost model outputs the optimal raw material ratio under the specific conditions to obtain the prediction results. Based on the prediction results of the XGBoost model trained under multiple different initialization conditions, the final optimal ratio prediction value is obtained by weighted average. The formula is: ； in, is the final optimal raw material ratio prediction value, It is The matching results predicted by the model are It is The weight of the model, is the number of models.

9. The method for producing the equipment for producing cement soil with adjustable multiple mix ratios according to claim 8, characterized in that: Step S4 specifically includes the following steps: S41, sensor network deployment and data collection: obtain real-time data of cement-soil ratio through the sensor network, and transmit the collected data to the cloud server in real time through the industrial Internet of Things; S42, sample collection and instant testing: collect multiple samples through an automated sample collection device, obtain sample test results, and transmit the sample test results to a cloud server in the form of digital signals; S43. Data integration and analysis: The cloud server integrates the real-time data of cement-soil ratio and sample test data according to time series and production batches, uses a sliding window algorithm to calculate the statistical characteristics of the data over a period of time, and adjusts the decision.

10. The method for producing the equipment for producing cement soil with multiple ratios adjustable according to claim 9, characterized in that: Step S5 specifically includes the following steps: S51, deviation identification and analysis: using dynamic time warping algorithm to compare real-time collected data with ideal values, if it is lower than the threshold, the deviation is determined and analyzed; S52, PSO-SVR hybrid algorithm to adjust decision generation: using historical production data in a distributed database as input, quality indicators as output and dividing into training sets and test sets, selecting radial basis kernel function to build a support vector regression model, and using cross-validation to adjust penalty parameters and kernel function parameters; Determine the particle swarm size, dimension, initial position and speed, and set the PSO algorithm parameters; Iteratively calculate the particle fitness value, update the individual and global optimal positions, and optimize the particles according to the speed and position update formula; When the maximum number of iterations is reached or the stopping condition is met, the adjustment plan corresponding to the global optimal position is output; S53, automatic adjustment execution: The cloud server sends instructions to the production equipment according to the adjustment plan to adjust the raw material delivery amount, stirring motor speed and time, and stirring temperature.

Images