Method and system for preparing backfill material by recycling spoil
By establishing a database of waste soil type and using neural network models to analyze the impact of admixtures on waste soil parameters, the required admixture ratio is calculated, and the problem of inaccurate proportion of waste soil backfill materials in the prior art is solved, efficient and reliable use of waste soil resources is achieved, and the quality and stability of backfill materials are improved.
Patent Information
- Application Number
- CN202411826438.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-12
- Publication Date
- 2025-03-14
AI Technical Summary
The prior art is difficult to accurately control the performance of backfill materials in a variable waste environment, and it is impossible to accurately match the appropriate backfill material ratio, resulting in unstable material quality and the actual use effect is inconsistent with expectations.
By establishing a database of waste soil type and using machine learning algorithms for automated classification, combining artificial neural network models to analyze the impact of admixture components on waste soil parameters, calculate the required admixture ratio, and generate the specific proportion of each admixture component.
It significantly improves the utilization efficiency of waste soil resources, ensures the scientificity and reliability of the proportion of backfill materials, avoids artificial experience errors, improves the uniformity and quality of materials, and has adaptability to adapt to changes in the properties of waste soil.
Smart Images
Figure CN119624439A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of data processing technology, and in particular to a method and system for recycling waste soil to prepare backfill materials. Background Art
[0002] In construction and civil engineering, a large amount of waste soil is generated due to excavation or construction. These waste soils are usually directly abandoned due to their uneven physical and chemical properties, resulting in waste of land resources and environmental pressure. Although existing technologies have attempted to utilize waste soil resources, they still face the following problems: The existing sources of abandoned soil are diverse, and their specific parameters are not the same. Although some methods for treating abandoned soil are disclosed in the prior art, their admixtures are often only applicable to abandoned soil under certain specific conditions. If the various parameters of the abandoned soil change, the consistency of the quality of the backfill material prepared therefrom cannot be guaranteed. Therefore, technicians are often required to adjust the specific component proportions of the admixtures they use based on experience according to the changes in the abandoned soil parameters to ensure the quality of the finished backfill material. However, such operation will not only greatly increase the cost of abandoned soil treatment, but also the quality of the finished product is too dependent on the experience of the technicians, thereby leading to technical problems such as the actual use effect of the backfill material not being consistent with expectations. Summary of the invention
[0003] The purpose of this application is to provide a method and system for recycling abandoned soil and preparing backfill materials, which aims to achieve the recycling of abandoned soil resources and prepare backfill materials that meet construction requirements by classifying and analyzing parameters of different types of abandoned soil and combining the addition ratio of admixtures.
[0004] In view of the above problems, the present application provides a method and system for recycling waste soil to prepare backfill materials, and the specific scheme is as follows; A method for recovering spoil to prepare backfill material comprises: Obtain the test results of parameters of the discarded soil after screening; The parameter detection results of the spoil are compared with the spoil categories in the spoil type database to obtain the type of the spoil; According to the type and parameters of the abandoned soil, the influence of each component of the admixture on different parameters is analyzed, the influence model between the parameters and the admixture components is established, and the effect of each admixture is quantified; According to the model results, the admixture ratio required for the spoil is calculated, and the specific ratio of each component of the admixture is generated.
[0005] Furthermore, the method for establishing the spoil type database is as follows: Obtaining initial data, wherein the initial data includes spoil parameters, admixture ratios, and backfill material parameters of different types of spoil; Introducing production data, which is the test results of real-time monitoring of spoil parameters, admixture ratio records, and backfill material parameters during spoil recovery; Data update: compare the newly collected data with the spoil type, mix ratio and performance target values in the initial database to confirm whether the new data belongs to the existing type. If any deviation is found, it will be recorded as a new category or the existing category will be expanded; The spoil type database is updated regularly based on production data. When new spoil types appear in production or changes in the parameter range of specific spoil types are found, corresponding entries are added or updated.
[0006] Furthermore, the method for establishing the spoil type database further includes: Through machine learning algorithms, the characteristics of spoil types are automatically classified and identified, so that the database can automatically generate new classification labels or ratio recommendations.
[0007] Further, the method for establishing the influence model between parameters and admixture components is as follows: According to the initial data in the spoil type database, the basic parameters of each type of spoil are obtained; Obtain the admixture ratio and finished product parameters of backfill materials corresponding to each type of spoil; Clean and standardize the parameters of waste soil, admixture ratio and finished product parameter data; Based on artificial neural network, the type of spoil, admixture ratio and finished product parameters were used as the input variables of the model to construct a parameter-admixture relationship model to quantify the influence of each admixture on different parameters.
[0008] Furthermore, the method for establishing the influence model between parameters and admixture components also includes: Incorporate production data into the model to update its predictions and fine-tune the model to more accurately reflect the impact of admixtures on various parameters.
[0009] Furthermore, according to the model results, the method of calculating the admixture ratio required for the spoil and generating the specific ratio of each component of the admixture is as follows: Obtain the parameters and type information of the spoil; According to the requirements of backfill materials, set the target performance of the final material of the spoil; According to the neural network model, the influence degree of each admixture component on different spoil parameters is confirmed, and the admixture-parameter influence weight is obtained; The admixture ratio required for the spoil is calculated through the admixture-parameter influence weight and the spoil parameter information, and the specific ratio of each component of the admixture is generated.
[0010] A system for recycling spoil to prepare backfill material, comprising: A spoil parameter acquisition module: used to obtain the spoil parameter detection results after screening, and compare the spoil parameter detection results with the spoil categories in the spoil type database to obtain the type of spoil; Influence model module establishment module: used to analyze the influence of various components of admixture on different parameters according to the type and parameters of the spoil, establish the influence model between parameters and admixture components, and quantify the effect of each admixture; Admixture ratio calculation module: used to calculate the admixture ratio required for the spoil according to the model results, and generate the specific ratio of each component of the admixture.
[0011] One or more technical solutions provided in this application have at least the following technical effects or advantages: 1. This application uses a spoil type database and machine learning algorithm to achieve automatic classification of spoil, accurately identify the characteristics of different spoil types, and significantly improve the utilization efficiency of spoil resources.
[0012] 2. This application associates the parameters of the abandoned soil and the components of the admixture through a neural network model, and quantifies the effects of different admixtures on the characteristics of the abandoned soil in a data-driven manner, making the ratio of the backfill material of the abandoned soil more scientific and reliable.
[0013] 3. This application automatically calculates the precise proportions of the admixture components required for different types of spoil based on the parameter influence weights output by the neural network model, thus avoiding human experience errors and improving the uniformity and quality of backfill materials.
[0014] 4. This system collects the parameters of waste soil and finished products during the production process, regularly optimizes the influencing model, and forms a dynamic tuning mechanism, so that the system has adaptiveness and can adjust the ratio in time to adapt to the changes in the properties of waste soil and maintain stable material quality.
[0015] 5. This application incorporates real-time monitoring data into the spoil type database, enabling the system to automatically generate new classification labels or ratio recommendations based on production data, thereby enhancing the flexibility and adaptability of spoil recovery.
[0016] In summary, this method significantly improves the practicality and economy of waste soil recycling. It not only meets the demand for backfill materials in construction, reduces the waste of resources and environmental burden caused by waste soil disposal, but also effectively reduces its recycling cost.
[0017] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.
[0019] Figure 1 A schematic diagram of a process for preparing backfill material by recycling spoil in this application; Figure 2 This is a schematic diagram of the structure of a system for recycling waste soil and preparing backfill materials in this application; Description of the reference numerals: spoil parameter acquisition module 11, influence model module establishment module 12, admixture ratio calculation module 13. DETAILED DESCRIPTION
[0020] The present application provides a method and system for preparing backfill materials by recycling abandoned soil, which solves the problems in the prior art that the influence of admixtures on the characteristics of abandoned soil cannot be accurately quantified, the performance of backfill materials cannot be accurately controlled in a changeable abandoned soil environment, and the suitable backfill material ratio cannot be accurately matched during the utilization of abandoned soil, making it difficult to maintain the quality of the prepared materials stable during production, resulting in the actual use effect of the backfill materials being inconsistent with expectations, especially when dealing with complex and diverse abandoned soil components, which will lead to the technical problem that the quality of the manufactured backfill materials is uneven, and even substandard quality occurs.
[0021] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.
[0022] Example 1, please refer to the attached Figure 1The present application provides a method and system for recycling waste soil to prepare backfill materials, wherein the production control method of the compound additive comprises: S100 obtains the test results of parameters of the sieved waste soil; That is, by sampling and testing the abandoned soil, its basic characteristic parameter range (such as moisture content, particle size distribution, density, etc.) is obtained, and the data can be obtained through signal transmission or manual input.
[0023] S200 compares the spoil parameter detection result with the spoil category in the spoil type database to obtain the type of spoil; Compare the detected spoil parameters with the data in the spoil type database; Confirm the type of spoil, if it does not match the existing type, mark it as “new type; S300 analyzes the influence of each component of admixture on different parameters according to the type and parameters of the spoil, establishes the influence model between parameters and admixture components, and quantifies the effect of each admixture; That is, by cleaning and standardizing the initial data and production data; Using artificial neural network model, with spoil type, admixture ratio and target finished product parameters as input variables, a model of the influence of admixture on backfill material parameters was constructed; Continuously tune the model based on feedback from production data to improve its prediction accuracy of the relationship between admixtures and parameters; S400 calculates the admixture ratio required for the waste soil according to the model results, and generates the specific ratio of each component of the admixture.
[0024] That is, by inputting spoil parameters and type information; Set target finished product parameters (i.e. backfill material requirements); The admixture-parameter influence weights obtained by the neural network model are used to calculate the proportions of each admixture component in combination with the actual parameters of the spoil; Output the proportioning scheme and generate specific proportioning data of each component of the admixture; S500 optimizes and adjusts the admixture ratio calculation model through actual data feedback during the production process; That is, collect the parameters of the spoil, admixture ratio and final product parameters generated during the production process; Compare the parameters of the generated backfill material finished product with the target parameters and analyze the errors; Based on error feedback, the weights and structure of the neural network model are adjusted to continuously optimize the model's prediction accuracy.
[0025] Wherein: the method for establishing the spoil type database is as follows: Initial data are obtained, wherein the initial data include spoil parameters, admixture ratios and backfill material parameters of different types of spoil.
[0026] Specifically, the initial data and the method for obtaining the initial data are as follows: Obtain spoil parameters: By sampling and testing various types of spoil, obtain the basic characteristic parameter range (such as moisture content, particle size distribution, mineral composition, etc.). The parameters of each type of spoil should be as comprehensive as possible to cover common types of spoil; Admixture ratio: For each type of spoil, record the type of admixture required and its specific ratio, including the addition ratio of different admixtures (such as cement, lime, plasticizer, etc.); Backfill material parameters: Determine the key performance parameters (such as fluidity, strength, density, etc.) of backfill materials prepared with various types of spoil as raw materials to ensure that they meet the requirements of backfill materials; Data storage: After the above initial data is sorted out, it is stored in the initial record of the database according to the relationship between the type of spoil, admixture ratio and finished product parameters, forming a basic database of spoil types; The above data can be entered based on the experience of technicians and data records from previous recycling processes to generate a basic database of spoil types; Introducing production data, which is the test results of real-time monitoring of spoil parameters, admixture ratio records, and backfill material parameters during spoil recovery; During the waste soil recovery process, the parameters of the waste soil, such as moisture content, particle size distribution, etc., are tracked and monitored in real time and stored in the historical database. Record the specific proportion of admixtures in each preparation process and store them in association with spoil parameters for easy traceability; Measure the performance parameters of the prepared backfill material and record them in the database as feedback data of the actual results; Data update: compare the newly collected data with the spoil type, mix ratio and performance target values in the initial database to confirm whether the new data belongs to the existing type. If any deviation is found, it will be recorded as a new category or the existing category will be expanded; The spoil type database is updated regularly based on production data. When new spoil types appear in production or changes in the parameter range of specific spoil types are found, corresponding entries are added or updated.
[0027] Through machine learning algorithms, the characteristics of spoil types are automatically classified and identified, so that the database can automatically generate new classification labels or mix ratio suggestions; Specifically: First, a large number of sample data of different types of spoil, as well as preliminary admixture ratio schemes and parameters of the final backfill materials (such as fluidity, compressive strength, etc.) are collected to form a data set; Clean up missing values or outliers in the data set, fill in with the mean or delete outliers when necessary; Standardize each feature to ensure that the dimensions of different features are consistent, which is conducive to the stability of the model; Divide the data set into training set and test set, usually in a ratio of 80%:20% or 70%:30%; Use algorithms suitable for classification tasks (such as decision trees or random forests) to automatically classify and identify the characteristics of the spoil types.
[0028] Using the feature importance output of classification models (such as random forests), identify which features have the greatest impact on the classification of spoil types, providing a reference for the subsequent automatic generation of mix ratios; For new spoil sample data, after inputting into the model, the model will automatically identify the category of spoil or predict the category to which the sample should belong; If new features are discovered (e.g., newly added physical or chemical properties), unsupervised clustering methods (e.g., DBSCAN) can be used to detect potential new classes or incorporated into existing classification models through adaptive updates; After the unsupervised cluster analysis, if the generated new category obviously has unique characteristics, set the category as the new spoil label (e.g., “category B1”, “category B2”); Based on the classification model, a regression model can be trained for each category of data to predict the optimal admixture ratio; Utilize the optimal ratio of historical data and combine it with optimization algorithms (such as particle swarm optimization PSO, genetic algorithm, etc.) to further refine the ratio suggestions for different categories. According to the material performance requirements, adjust the ratio of admixture components and finally output the ratio scheme that matches the category; After each classification and mix ratio recommendation of spoil samples, test verification is carried out and the results are fed back to the database; If a certain type of spoil is found to have large deviations in performance, the model can be continuously optimized through feedback data to continuously enhance the accuracy of classification and proportion generation.
[0029] In practical applications, this method mainly compares the spoil parameter detection results with the spoil types in the database to determine its type; The inventors found that for some relatively fixed and less changing usage scenarios, this method can also be used to directly generate corresponding admixture ratio suggestions, but for some more complex usage scenarios, the effect of using an artificial neural network to establish a model will be more stable; The finished product performance test data is fed back to the database to adjust or optimize the admixture ratio records for the waste soil type, forming an automated optimization cycle.
[0030] The method for establishing the influence model between parameters and admixture components is as follows: S310 obtains basic parameters of each type of spoil according to initial data in the spoil type database; That is, according to the initial data in the spoil type database, the basic parameters of each type of spoil (such as moisture content, particle distribution, density, etc.) are obtained; Extract the above parameter information into the model input data set through SQL query or database API interface to form a basic data table or data structure (such as Pandas DataFrame); For parameter values that may be missing in the data, interpolation, mean filling or deletion of missing data are used to process them; Process extreme data values (such as values that deviate from the mean by several times) to avoid their interference with the model; Convert parameter data to standard data types (such as floating point) to ensure data consistency and avoid errors when inputting into the model; S320 obtains the admixture ratio and finished product parameters of backfill materials corresponding to each type of spoil; That is, query the admixture ratio data in the database to ensure that each type of waste soil corresponds to a set of admixture ratios, such as the ratio of admixture A, admixture B, admixture C, etc. (which can be expressed as a percentage or a specific mass value); Obtain the corresponding finished backfill material parameters, such as fluidity, strength, density, etc.
[0031] Integrate admixture ratio, finished product parameter data and basic spoil parameters to generate a complete data input form for subsequent data cleaning and standardization steps; S330 cleans and standardizes the parameters of spoil, admixture ratio and finished product parameters to ensure data consistency and model validity; That is, irrelevant data such as duplicate spoil data and redundant mix ratio information are eliminated to ensure that each record in the data set is unique; Check the data format to ensure that the data types of each column are consistent. For example, the moisture content needs to be unified as a percentage, the density needs to be unified as kg / m³, etc.; Convert text and symbolic data into numerical data to ensure consistency in model processing; Perform mean normalization or Z-score standardization on the values of finished product parameters to reduce the impact of numerical deviation on the model; Check whether the distribution of each parameter in the data set is uniform. If necessary, data augmentation can be performed on some parameter data (such as using the SMOTE algorithm to generate synthetic samples) to balance the data set. S340 is based on artificial neural network and takes spoil type, admixture ratio and finished product parameters as the input variables of the model to construct a parameter-admixture relationship model to quantify the influence of each admixture on different parameters.
[0032] Specifically, this step includes: Define the model structure: Input layer: Set the number of input layer nodes to be consistent with the number of model input variables (spoil type, admixture ratio, and all fields of finished product parameters).
[0033] Hidden layer: Set several hidden layers according to the data scale, usually including 1-3 layers, with 32-128 nodes in each layer; ReLU activation function can be used.
[0034] Output layer: Set the number of nodes equal to the number of output variables, which are the influence weights of each admixture component. The output layer activation function can be linear or other suitable functions.
[0035] Divide the dataset into training set, validation set and test set, with a common ratio of 7:2:1 to ensure the generalization ability of the model; Choose a loss function (such as mean squared error MSE) and an optimization algorithm (such as Adam optimizer); Train the model in the training set, evaluate the model effect in the validation set, and adjust the model parameters (such as learning rate and regularization parameters) based on the validation results; The training process adopts batch training mode; the number of training iterations (epochs) can be adjusted according to the changes in the validation loss to prevent overfitting; Use the test set to perform a final evaluation on the trained model, observe its prediction error and accuracy, and adjust the model architecture or hyperparameters to improve accuracy; The final model output is the influence weight of each admixture on the finished product parameters, which is used for subsequent ratio calculations.
[0036] S350 incorporates production data into the model to update its predictions and fine-tune the model to more accurately reflect the impact of admixtures on various parameters; Specifically, the method is as follows: During the production process, the actual parameters of the spoil, admixture ratio and finished product parameters are collected and the new data are stored in the database; Extract production data from the database at certain time intervals or frequency intervals and organize them into a format consistent with the initial data of the model; Perform the same cleaning and standardization as the original data to ensure data consistency; Use new data as incremental data and fine-tune the existing model to update parameters and weights; Verify the updated model and test the accuracy of model predictions through a new round of validation sets; Adjust the hyperparameters of the neural network, such as learning rate, number of hidden layer nodes, regularization strength, etc., based on the model evaluation results to ensure the accuracy of the model on new data; The optimized model is redeployed into the system to enable it to process new spoil parameters and admixture ratio data, and to improve the prediction accuracy so that it can better reflect the impact of admixtures on spoil parameters.
[0037] S400 calculates the admixture ratio required for the spoil according to the model results, and generates the specific ratio of each component of the admixture as follows: S401 obtains parameters and type information of spoil; That is, key parameters of spoil, such as moisture content, particle size distribution, density, etc., are obtained through sensors or laboratory testing. These parameters will be used as inputs to the model; By querying the spoil type database, the acquired spoil parameters are compared with the spoil types in the database to identify the type of spoil; The parameter detection results and type identification results are saved together in the database for subsequent calls.
[0038] S402 sets the target performance of the final material of the spoil according to the requirements of the backfill material; That is, define the target performance of the backfill material according to the project requirements. For example, the target performance may include; Fluidity: For example, the slump should be within a certain range to facilitate construction; Strength: Ensure that the backfill material has sufficient load-bearing capacity after solidification; Density: ensure the density and durability of the material; Shrinkage rate: ensure that no cracks will occur after the material solidifies; Parameter standardization: converting target properties into numerical form to facilitate subsequent calculations. For example, fluidity can be quantified as a slump range, strength in MPa, and density in kg / m³; The numerical values of the target performance are stored in a database or a model input structure for later use in calculating the admixture ratio.
[0039] S403 confirms the influence of each admixture component on different spoil parameters according to the neural network model, and obtains the admixture-parameter influence weight; Input the parameter information of the spoil, the type of spoil and the target performance into the neural network model, and start to calculate the influence of each admixture component on the target performance; Use the pre-trained neural network model to output the influence of each admixture component on the target performance (i.e., influence weight); Its model can quantify the contribution of each admixture to target properties such as fluidity, strength, and density through weighted regression or normalized output weights; The model may conclude that admixture A has a weight of 0.3 for fluidity and 0.2 for strength, admixture B has a weight of 0.1 for fluidity and 0.5 for strength, and so on; Verify the obtained impact weights to ensure that the sum of the weights is consistent with the target performance requirements so that the proportion calculation can be performed; If deviations occur, fine-tuning can be performed, such as through model tuning or weight normalization; The calculated admixture-parameter influence weights are stored in the model output or in an intermediate data structure to provide a basis for subsequent steps.
[0040] S404 calculates the admixture ratio required for the spoil by using the admixture-parameter influence weight and the spoil parameter information, and generates the specific ratio of each component of the admixture; The weighted formula can be used to calculate the proportion of each admixture to ensure that the target performance requirements are met. The formula is as follows; P i =∑(W ij ⋅T j ) Where: P i is the recommended proportion of the ith admixture; W ij : The influence weight of the ith admixture on the jth spoil parameter (extracted from the neural network); T j : Target performance requirement of the jth spoil parameter.
[0041] Convert the calculated ratio of each admixture into mass or volume ratio to ensure easy measurement in actual operation; Substitute the generated admixture ratio into the neural network model to simulate the performance parameters of the generated backfill material; If the simulation results meet the set target performance, the ratio is confirmed to be accurate; if the deviation is large, the impact weight needs to be adjusted and the ratio needs to be recalculated until the target is achieved; Generate the final data table of the distribution ratio of each group of admixtures, store the data in the database or output system, and generate the corresponding formula file for use in the production line.
[0042] Embodiment 2: A system for recovering waste soil and preparing backfill materials, comprising: The spoil parameter acquisition module 11 is used to obtain the spoil parameter detection results after screening, and compare the spoil parameter detection results with the spoil categories in the spoil type database to confirm the type of the spoil. If it does not conform to the existing type, it is marked as "new type; The influence model module establishment module 12 is used to analyze the influence of each component of the admixture on different parameters according to the type of spoil and the parameters of the spoil, establish the influence model between the parameters and the admixture components, and quantify the effect of each admixture.
[0043] The specific steps are as follows: Clean and standardize initial and production data; Using artificial neural network model, with spoil parameter type, admixture ratio and target finished product parameters as input variables, a model of the influence of admixture on backfill material parameters was constructed. Continuously tune the model based on feedback from production data to improve its prediction accuracy of the relationship between admixtures and parameters; The admixture ratio calculation module 13 is used to calculate the admixture ratio required for the waste soil according to the model results, and generate the specific ratio of each component of the admixture.
[0044] The specific steps are as follows: Input spoil parameters and type information; Set target finished product parameters (requirements for backfill material); The admixture-parameter influence weights obtained by the neural network model are used to calculate the proportions of each admixture component in combination with the actual parameters of the spoil; Output the proportioning scheme and generate specific proportioning data of each component of the admixture; Data feedback and optimization module 14: used to optimize and adjust the admixture ratio calculation model through actual data feedback in the production process.
[0045] The specific steps are as follows: Collect the parameters of waste soil, admixture ratio and final product parameters generated during the production process; Compare the parameters of the generated backfill material finished product with the target parameters and analyze the errors; According to the error feedback, the weight and structure of the neural network model are adjusted to continuously optimize the prediction accuracy of the model; The system also includes a spoil type database management module, which is used to establish and maintain a spoil type database, store different spoil types and their related admixture ratios and finished product parameters, and provide data support for spoil parameters and admixture ratio models; That is, input the spoil parameters of different spoil types, initial admixture ratio and finished product parameters of backfill materials; During the waste soil recovery process, the waste soil parameters, admixture ratio, and finished product parameters are monitored in real time and recorded; Newly collected data are compared with existing database entries. If the parameters deviate from the range of existing categories, the system automatically generates new classification labels or expands existing category entries; Automatically classify the characteristics of spoil types through machine learning algorithms to generate new classification labels or admixture ratio recommendations; Any of the above methods or steps can be stored as computer instructions or programs in various types of computer memories, and the computer instructions or programs can be recognized by various types of computer processors to implement any of the above methods or steps; Secondly, the above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
[0046] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.
Claims
1. A method for recovering abandoned soil to prepare backfill material, characterized in that: include: Obtain the test results of parameters of the discarded soil after screening; The parameter detection results of the spoil are compared with the spoil categories in the spoil type database to obtain the type of the spoil; According to the type and parameters of the abandoned soil, the influence of each component of the admixture on different parameters is analyzed, the influence model between the parameters and the admixture components is established, and the effect of each admixture is quantified; According to the model results, the admixture ratio required for the spoil is calculated, and the specific ratio of each component of the admixture is generated.
2. The method for preparing backfill material by recycling spoil according to claim 1, characterized in that: The method for establishing the spoil type database is as follows: Obtaining initial data, wherein the initial data includes spoil parameters, admixture ratios, and backfill material parameters of different types of spoil; Introducing production data, which is the test results of real-time monitoring of spoil parameters, admixture ratio records, and backfill material parameters during spoil recovery; Data update: compare the newly collected data with the spoil types, proportions and performance target values in the initial database to confirm whether the new data belongs to the existing type. If any deviation is found, it will be recorded as a new category or an expansion of the existing category; The spoil type database is updated regularly based on production data. When new spoil types appear in production or changes in the parameter range of specific spoil types are found, corresponding entries are added or updated.
3. The method for preparing backfill material by recycling spoil according to claim 1, characterized in that: The method for establishing the spoil type database further includes: Through machine learning algorithms, the characteristics of spoil types are automatically classified and identified, so that the database can automatically generate new classification labels or ratio recommendations.
4. The method for preparing backfill material by recycling spoil according to claim 1, characterized in that: The method for establishing the influence model between parameters and admixture components is as follows: According to the initial data in the spoil type database, the basic parameters of each type of spoil are obtained; Obtain the admixture ratio and finished product parameters of backfill materials corresponding to each type of spoil; Clean and standardize the parameters of waste soil, admixture ratio and finished product parameter data; Based on artificial neural network, the type of spoil, admixture ratio and finished product parameters were used as the input variables of the model to construct a parameter-admixture relationship model to quantify the influence of each admixture on different parameters.
5. The method for preparing backfill material by recycling spoil according to claim 4, characterized in that: The method for establishing the influence model between parameters and admixture components also includes: Incorporate production data into the model to update its predictions and fine-tune the model to more accurately reflect the impact of admixtures on various parameters; The method for preparing backfill material by recycling spoil according to claim 4 is characterized in that the method of calculating the admixture ratio required for the spoil according to the model results and generating the specific ratio of each component of the admixture is as follows: Obtain the parameters and type information of the spoil; According to the requirements of backfill materials, set the target performance of the final material of the spoil; According to the neural network model, the influence degree of each admixture component on different spoil parameters is confirmed, and the admixture-parameter influence weight is obtained; The admixture ratio required for the spoil is calculated through the admixture-parameter influence weight and the spoil parameter information, and the specific ratio of each component of the admixture is generated.
6. A system for recycling spoil to prepare backfill material, the system being used to execute the method according to any one of claims 1 to 6, characterized in that: include: A spoil parameter acquisition module: used to obtain the spoil parameter detection results after screening, and compare the spoil parameter detection results with the spoil categories in the spoil type database to obtain the type of spoil; Influence model module establishment module: used to analyze the influence of various components of admixture on different parameters according to the type and parameters of the spoil, establish the influence model between parameters and admixture components, and quantify the effect of each admixture; Admixture ratio calculation module: used to calculate the admixture ratio required for the spoil according to the model results, and generate the specific ratio of each component of the admixture.