Intelligent classification processing system and method for construction solid waste

Abstract

The application discloses a kind of intelligent classification processing system and method of construction solid waste, it is related to intelligent classification technical field, comprising: the multi-dimensional data of construction solid waste is collected in target area traversal, determine the data set to be classified;Determine a plurality of classification labels to construct intelligent classification model, the data set to be classified is synchronized to intelligent classification model and obtains classification decision;Simulate execution classification decision and carry out classification processing monitoring to construction solid waste, generate running feedback data;Intelligent classification model is optimized based on running feedback data, obtain intelligent optimization classification model, the data set to be classified is synchronized to intelligent optimization classification model, and obtain classification optimization decision;Intelligent classification processing is carried out to construction solid waste.The application solves the technical problems that existing technology has low efficiency and low accuracy when classifying and processing construction solid waste, to improve the accuracy and efficiency of the classification of construction solid waste.

Description

Technical Field

[0001] This invention relates to the field of intelligent sorting technology, specifically to an intelligent sorting and processing system and method for construction solid waste. Background Technology

[0002] In the modern construction industry, with the acceleration of urbanization and the continuous expansion of building scale, the amount of construction solid waste is also increasing rapidly. This construction solid waste mainly includes concrete, bricks, metals, plastics, wood, etc., and is diverse in type and complex in composition, often mixed together. If not properly handled, it not only occupies a large amount of land resources but also causes serious environmental pollution. Traditional construction solid waste treatment methods mainly rely on manual sorting and simple mechanical sorting, which are inefficient, inaccurate, and unable to meet the needs of large-scale waste treatment. Summary of the Invention

[0003] This application provides an intelligent classification and treatment system and method for construction solid waste, which addresses the technical problems of low efficiency and low accuracy in the classification and treatment of construction solid waste in existing technologies.

[0004] In view of the above problems, this application provides an intelligent sorting and treatment system and method for construction solid waste.

[0005] A first aspect of this application provides an intelligent sorting and processing system for construction solid waste, the system comprising: The system comprises the following modules: a data traversal module, which collects multidimensional data of construction solid waste from the target area to determine the dataset to be classified; a classification label determination module, which determines multiple classification labels, constructs an intelligent classification model using these labels, and synchronizes the dataset to be classified to the intelligent classification model to obtain a classification decision; a classification processing simulation module, which simulates the execution of the classification decision to classify and monitor the construction solid waste, generating operational feedback data; a classification optimization module, which optimizes the intelligent classification model based on the operational feedback data to obtain an intelligent optimized classification model, synchronizes the dataset to be classified to the intelligent optimized classification model, and obtains a classification optimization decision; and a classification processing module, which executes the classification optimization decision to perform intelligent classification processing of the construction solid waste.

[0006] A second aspect of this application provides an intelligent sorting and treatment method for construction solid waste, the method comprising: The process involves: collecting multidimensional data on construction solid waste from a target area to determine the dataset to be classified; identifying multiple classification labels and constructing an intelligent classification model using these labels; synchronizing the dataset to be classified to the intelligent classification model to obtain classification decisions; simulating the execution of the classification decisions to classify and process construction solid waste, generating operational feedback data; optimizing the intelligent classification model based on the operational feedback data to obtain an optimized intelligent classification model; synchronizing the dataset to be classified to the optimized intelligent classification model to obtain optimized classification decisions; and finally, executing the optimized classification decisions to intelligently classify and process construction solid waste.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: This application collects multidimensional data of construction solid waste from a target area to determine the dataset to be classified; identifies multiple classification labels and uses these labels to construct an intelligent classification model; synchronizes the dataset to be classified to the intelligent classification model to obtain classification decisions; simulates the execution of classification decisions to monitor the classification and processing of construction solid waste, generating operational feedback data; optimizes the intelligent classification model based on the operational feedback data to obtain an intelligent optimized classification model; synchronizes the dataset to be classified to the intelligent optimized classification model to obtain optimized classification decisions; and executes the optimized classification decisions to intelligently classify and process construction solid waste. This invention addresses the technical problems of low efficiency and low accuracy in the classification and processing of construction solid waste in existing technologies. Through multidimensional data collection, classification label construction, intelligent classification model training, simulated execution monitoring, feedback optimization, and final execution of optimized classification decisions, it achieves the technical effect of improving the accuracy and efficiency of construction solid waste classification. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 A schematic diagram of an intelligent sorting and processing system for construction solid waste provided in this application embodiment; Figure 2 This is a schematic diagram of a method for intelligent sorting and processing of construction solid waste provided in an embodiment of this application.

[0010] Figure labeling: Data traversal module 11, category label determination module 12, category processing simulation module 13, category optimization module 14, category processing module 15. Detailed Implementation

[0011] This application provides an intelligent classification and treatment system and method for construction solid waste, which addresses the technical problems of low efficiency and low accuracy in the classification and treatment of construction solid waste in existing technologies. By means of multi-dimensional data collection, classification label construction, intelligent classification model training, simulation execution monitoring, feedback optimization, and final classification optimization decision-making, the system achieves the technical effect of improving the accuracy and efficiency of construction solid waste classification.

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0013] It should be noted that any variation of the terms "comprising" and "having" is intended to cover non-exclusive inclusion, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such processes, methods, products, or devices.

[0014] Example 1, as Figure 1 As shown in the figure, this application provides an intelligent sorting and processing system for construction solid waste, the system comprising: The data traversal module 11 traverses the target area to collect multidimensional data of construction solid waste and determines the dataset to be classified.

[0015] In this embodiment, the data traversal module traverses the target area, that is, it uses sensors or other detection devices to systematically scan the target area to obtain multi-dimensional data including the spatial location, material type, volume, density, shape, and color of construction solid waste. The target area refers to the specific physical area where solid waste classification is required, such as a building demolition site or a construction waste dump.

[0016] Through the above process, multidimensional data of construction solid waste is collected from the target area, and the obtained multidimensional data is used as a dataset to be classified.

[0017] The classification label determination module 12 determines multiple classification labels, constructs an intelligent classification model using the multiple classification labels, and synchronizes the dataset to be classified to the intelligent classification model to obtain a classification decision.

[0018] In this embodiment, the dataset to be classified is first standardized to ensure comparability between data in each dimension. Then, multidimensional features are extracted from each data point in the dataset. After multidimensional feature extraction, the classification label determination module performs feature annotation and cluster analysis. Labels are assigned to each feature dimension using an annotation algorithm, and the labeled data is clustered using a cluster analysis algorithm to generate multiple classification labels.

[0019] Next, an intelligent classification model is built using multiple classification labels. During the model building process, each label corresponds to a reinforcement learner. Through continuous learning and training, each classification label is validated and optimized, ultimately generating an intelligent classification model capable of accurate classification based on multi-dimensional features. Finally, the dataset to be classified is synchronized to the model for classification decision-making.

[0020] Furthermore, in the system provided in the application embodiment, the classification label determination module 12 is also used for: Based on the dataset to be classified, data standardization is performed to obtain multiple standard data to be classified; multidimensional feature extraction is performed on the multiple standard data to be classified to determine multidimensional data features; feature annotation is performed on the multiple standard data to be classified according to the multidimensional data features to construct a multi-feature network; cluster analysis is performed on the multi-feature network to determine the multiple classification labels.

[0021] In this embodiment, the dataset to be classified is first preprocessed through data standardization. The Min-Max normalization method is used to scale the value of each feature to the range [0,1], eliminating dimensional differences between features and ensuring that each feature has a consistent numerical scale in subsequent processing. This process yields multiple standard datasets to be classified.

[0022] Next, multidimensional feature extraction is performed on multiple standard data to be classified. Principal component analysis is used to calculate the covariance matrix of the data and extract the principal components that can explain the largest variance of the data. After completion, a multidimensional feature dataset composed of multiple principal components is obtained.

[0023] Next, in the feature annotation stage, a Support Vector Machine (SVM) algorithm is employed. The goal of this stage is to assign a corresponding classification label to each data sample based on the extracted multidimensional features. The SVM is pre-trained and learns the classification boundaries of the samples to categorize new samples into the appropriate categories. For example, based on features such as volume, weight, and material, samples are labeled as "metal waste" or "concrete waste." After this stage, a labeled dataset with clear classification labels is generated. Following this, in the multi-feature network construction stage, a feature node network is built based on the correlations between features. Each multidimensional feature is considered a node, and these nodes are interconnected through their correlations, forming a multi-feature network. Specifically, each extracted feature is considered a node, and the connections between nodes are based on the correlations between features, such as measured by the Pearson correlation coefficient.

[0024] Finally, in the clustering analysis stage, the K-Means clustering algorithm is used. Based on the data in the multi-feature network, the K-Means algorithm iteratively adjusts the centroids of the clusters, grouping samples with similar features into the same cluster. Each cluster represents a specific classification label, such as "metal" or "plastic". Through this clustering process, multiple classification labels are generated.

[0025] Furthermore, in the system provided in the application embodiment, the classification label determination module 12 is also used for: A multi-label task environment is set up. Based on the multi-label task environment, ensemble reinforcement learning is performed by traversing the multiple classification labels to obtain multiple reinforcement learners. The multiple reinforcement learners correspond to the multiple classification labels. Supervised training is performed based on the multiple reinforcement learners to generate a first training result. The first training result is then verified to generate a data verification token. The multiple reinforcement learners are integrated according to the data verification token to construct the intelligent classification model.

[0026] In this embodiment, a multi-label classification framework, such as ML-KNN, is employed to handle complex multi-label classification tasks during the initial stage of setting up the multi-label task environment. Each construction waste sample may belong to multiple categories simultaneously, such as "metal" and "recyclable". Therefore, the classification task is first decomposed into multiple binary or multi-class problems, and a multi-label task environment is constructed. Within this environment, all label combinations are traversed, forming multiple independent sub-tasks that allow the system to process different labels separately.

[0027] Next, in the reinforcement learning phase, the classifier is trained using the Q-Learning algorithm. Q-Learning is a value-function-based reinforcement learning method that aims to maximize cumulative reward by learning an optimal policy. In this phase, a reinforcement learner is assigned to each labeled task. These learners use Q-Learning to learn how to make classification decisions in a multi-label task environment. Specifically, Q-Learning updates the Q-value by iteratively balancing exploration and exploitation; the Q-value represents the maximum expected reward obtainable by taking a certain action in a given state. Through multiple iterations, the learner's classification policy is continuously adjusted to optimize its performance across different labeled tasks. Ultimately, multiple reinforcement learners capable of executing optimal classification policies are trained.

[0028] After reinforcement learning is completed, the supervised training phase begins. This phase aims to further optimize the learner trained by Q-Learning using supervised learning methods. Supervised learning algorithms such as Support Vector Machines or Random Forests are used to compare the labeled data with the learner's initial classification results, and model parameters are adjusted to improve classification accuracy. In this phase, supervised learning utilizes the labeled training data to further optimize the output of Q-Learning, enabling the learner to perform well in higher-precision classification tasks. After this phase, the first training results are generated.

[0029] Next comes the validation and data validation token generation phase. In this phase, cross-validation is used to validate the performance of the first training result. Cross-validation divides the dataset into several folds, using one portion as the validation set and the remaining portion as the training set for multiple training and validation iterations to evaluate the model's generalization ability and robustness. Performance metrics such as accuracy and F1 score are calculated to evaluate the model's performance. If the model passes the preset performance criteria, a data validation token is generated, indicating that the model has passed validation and is eligible to participate in the integration process. After this phase, multiple data validation tokens are generated, validating the effectiveness of the reinforcement learner and confirming its eligibility for subsequent integration steps.

[0030] Finally, in the ensemble model building phase, ensemble learning techniques, such as Bagging, are employed to integrate multiple validated reinforcement learners. Bagging trains multiple independent learners and averages their predictions. During the ensemble process, the outputs of each learner are weighted and averaged according to their performance, thereby improving the robustness and accuracy of the final decision. Through these ensemble methods, the intelligent classification model is finally constructed.

[0031] Furthermore, in the system provided in the application embodiment, the classification label determination module 12 is also used for: Based on the multiple reinforcement learning machines, the system sequentially determines whether the dataset to be classified conforms to the target classification label; extracts the data to be classified that conforms to the target classification label, generates a positive reward instruction, and uses the positive reward instruction to identify the data to be classified that conforms to the target classification label to obtain a first classification result; extracts the data to be classified that does not conform to the target classification label, generates a negative reward instruction, and uses the negative reward instruction to identify the data to be classified that does not conform to the target classification label to obtain a second classification result; performs a weighted calculation based on the first classification result and the second classification result to generate multiple weight coefficients; and formulates the classification decision based on the multiple weight coefficients combined with the first classification result and the second classification result.

[0032] In this embodiment, multiple reinforcement learners are first used to determine whether each sample in the dataset to be classified conforms to the target classification label. Each reinforcement learner focuses on a specific label and makes a classification decision based on the characteristics of the sample. By having multiple reinforcement learners determine whether each sample conforms to the specific target label one by one, a preliminary classification result is generated after this process is completed.

[0033] Next, in the stage of generating positive reward instructions and identifying the first classification result, a positive reward instruction is generated for samples that match the target classification label. The positive reward is set to +1, indicating that the classification action should be reinforced for correctly classified samples. The correctly classified samples are identified as the first classification result and recorded as positive classification samples that match the target classification label. Simultaneously, in the stage of generating negative reward instructions and identifying the second classification result, a negative reward instruction is generated for samples that do not match the target classification label. The negative reward, as a feedback mechanism for correcting the classification strategy, is set to -1, informing the system that the classification action was incorrect and the classification strategy needs to be adjusted in the future. These misclassified samples are identified as second classification results, and these samples are recorded as negative classification samples that do not match the target label.

[0034] Then, a weighted calculation is performed based on the first and second classification results. Specifically, the first and second classification results are weighted according to their respective classification outcomes. The weights are determined based on their importance in the task requirements and the difficulty of the classification task. When certain classification labels are of high importance in the overall task, such as certain high-value waste categories, higher weights are assigned to these labels based on the task requirements. If certain classification labels are difficult to classify accurately, higher weights are assigned to these more difficult-to-classify labels. Based on these factors, multiple weight coefficients are generated using a weighted average algorithm.

[0035] Finally, in the stage of making a classification decision by combining weighted coefficients, a weighted voting mechanism is used to integrate the first and second classification results. Based on the previously generated weighted coefficients, a weighted vote is performed on the positive and negative classification results. The weighted voting mechanism allows the system to balance the influence of different classification results in the final classification decision, ensuring a more accurate classification decision. Through this weighted voting, the final classification decision is generated.

[0036] The classification and processing simulation module 13 simulates the execution of the classification decision to classify and monitor construction solid waste, and generates operational feedback data.

[0037] In this embodiment, the classification decision is first simulated, followed by a classification processing monitoring phase. During this process, the classification processing simulation module monitors the classification process in real time, records its performance, and captures various data, such as classification speed, accuracy, and waste allocation efficiency for each category. Operational feedback data is generated through this process.

[0038] Furthermore, in the system provided in the application embodiment, the classification processing simulation module 13 is also used for: Based on the dataset to be classified, the classification decision is simulated and a classification control instruction is generated; the sorting machinery is activated according to the classification control instruction, and multiple allocation areas are determined; the construction solid waste is classified and monitored according to the multiple allocation areas to obtain a classification monitoring dataset, which includes classification speed monitoring data, correct classification monitoring data, and incorrect classification monitoring data; the classification efficiency of the classification decision is calculated based on the classification speed monitoring data, the correct classification monitoring data, and the incorrect classification monitoring data, and a classification operation log is constructed; the classification operation log is added to the operation feedback data.

[0039] In this embodiment, classification decisions are first simulated based on the dataset to be classified to generate classification control instructions. Specifically, a virtual classification environment is first created to simulate the actual construction solid waste classification process. The classification processing simulation module simulates the execution of classification decisions based on the virtual data in the dataset to be classified. The intelligent classification model generates classification control instructions in the simulation environment based on the characteristics of the dataset to be classified, such as volume, weight, and material. These instructions specify the category of the waste and simulate the execution effect of the instructions.

[0040] Next, based on the classification control instructions generated by the simulation, the sorting machinery is further activated in the virtual environment. These sorting machinery models are used to simulate the operation of automated sorting equipment, classifying virtual construction solid waste according to the classification control instructions. The virtual waste is assigned to multiple allocation areas, which are the placement locations for the sorted construction waste. Each allocation area corresponds to a specific classification result; for example, metal waste is allocated to the metal area, and plastic waste is allocated to the plastic area. Then, in the simulation environment, the classification processing simulation module monitors the entire classification process through virtual sensors, recording various types of data to form a classification monitoring dataset. The classification monitoring dataset includes classification speed monitoring data, correct classification monitoring data, and incorrect classification monitoring data. Specifically, the classification speed monitoring data records the operating speed of the virtual sorting machinery; the correct classification monitoring data monitors the accuracy of classification; and the incorrect classification monitoring data records errors in classification.

[0041] Classification efficiency is calculated based on classification speed, accuracy, and error rate from the classification monitoring dataset. The calculation of classification efficiency considers both speed and accuracy, reflecting the overall performance of the intelligent classification model in the virtual environment. For example, high classification speed and accuracy result in high classification efficiency, while a high error rate will negatively impact efficiency. After efficiency calculation, a classification operation log is constructed to record the execution of classification control commands, the performance of the sorting machinery, and the classification efficiency results. Finally, the constructed classification operation log is added to the operation feedback data.

[0042] The classification optimization module 14 optimizes the intelligent classification model based on the running feedback data to obtain an intelligent optimized classification model, and synchronizes the dataset to be classified to the intelligent optimized classification model to obtain a classification optimization decision.

[0043] In this embodiment, the intelligent classification model is first optimized using runtime feedback data. Specifically, the runtime feedback data is analyzed to identify shortcomings of the intelligent classification model in actual operation, such as a high misclassification rate for certain categories or insufficient sorting speed. Based on this feedback, the classification optimization module employs an adaptive learning algorithm to automatically adjust the weights, parameters, and strategies of the intelligent classification model, thereby obtaining an optimized intelligent classification model.

[0044] The dataset to be classified is then synchronized to the intelligent optimization classification model, and the optimized model is used to infer the dataset and generate classification optimization decisions.

[0045] Furthermore, in the system provided in the application embodiment, the classification optimization module 14 is also used for: Data mining is performed on the operational feedback data based on the classified operational logs to obtain multiple classification association rules; the operational feedback data is analyzed according to the multiple classification association rules to obtain operational analysis results; classification evaluation is performed based on the operational analysis results to obtain operational classification evaluation results; anomalies are captured by traversing the operational classification evaluation results to generate abnormal classification information; a model adjustment strategy is formulated based on the abnormal classification information, and the intelligent classification model is retrained by executing the model adjustment strategy to obtain a second training result; the data validity of the second training result is verified, and the intelligent optimized classification model is generated when the verification is successful.

[0046] In this embodiment, an association rule mining algorithm, such as the Apriori algorithm, is first used to mine data from the classification operation logs. The classification operation logs record various data points related to the classification process, such as classification accuracy, classification error rate, and classification speed. Through the association rule mining algorithm, multiple classification association rules are extracted from the logs. These rules reveal the potential relationships between different classification behaviors. For example, a pattern with a high classification error rate for a certain type of waste under specific conditions is identified.

[0047] Next, the extracted classification association rules are used to analyze the operational feedback data. This feedback data includes classification performance after the classification task is completed, such as classification error rate, classification accuracy, and classification efficiency. By combining these classification association rules with the feedback data, potential bottlenecks or problems in the system during classification tasks are identified. For example, it may be found that under certain rules, the classification efficiency for specific categories is low or the classification error rate is high. Through this association rule-based analysis, the weaknesses of the intelligent classification model and directions for improvement are understood. After the analysis is completed, operational analysis results are generated, revealing the performance of the intelligent classification model in practical applications and areas requiring optimization.

[0048] Based on the operational analysis results, the current intelligent classification model is comprehensively evaluated using classification performance assessment tools, such as ROC curve analysis or confusion matrix analysis. These evaluation methods can demonstrate in detail the accuracy and error rate of the intelligent classification model in different classification tasks. For example, the confusion matrix shows the model's specific performance when classifying different waste categories, including the number of correct and incorrect classifications. Through evaluation, the strengths and weaknesses of the intelligent classification model are identified, and specific areas requiring further optimization are determined. After completing the evaluation, operational classification evaluation results are generated, providing a basis for subsequent model optimization.

[0049] During the classification evaluation process, anomaly detection algorithms, such as Isolation Forest or local anomaly factors, are used to traverse the classification evaluation results and capture anomalies that occur during classification. These anomalies include situations where the classification error rate increases significantly under specific conditions, or where the classification speed decreases significantly. These anomalies are captured, and anomaly classification information is generated, detailing the conditions under which the anomalies occur and their potential impact.

[0050] Next, targeted model tuning strategies are developed based on the anomaly classification information. These strategies involve optimizing model parameters, adjusting the learning rate, balancing class weights, or increasing the diversity of the training dataset. Then, the intelligent classification model is retrained using adaptive learning algorithms, such as stochastic gradient descent. Adaptive learning algorithms dynamically adjust model parameters to better handle complex classification tasks, reduce error rates, and improve classification efficiency. After retraining, a second training result is generated, showing that the model's performance on the classification task has improved after tuning.

[0051] Finally, the effectiveness of the retrained model is verified through cross-validation and test set evaluation. Cross-validation tests the model's performance under different conditions by splitting multiple datasets, ensuring that the model maintains good generalization ability when dealing with new data. Simultaneously, the test set is used to evaluate the model's performance in real-world applications, verifying whether it has achieved the expected classification accuracy and stability. Through these validation steps, the optimized intelligent classification model is finally confirmed and generated.

[0052] Furthermore, in the system provided in the application embodiment, the classification optimization module 14 is also used for: Anomaly tracing is performed based on the anomaly classification information to determine the anomaly data source, which has an anomaly source identifier; anomaly pattern is determined based on the anomaly source identifier, and adjustment analysis is performed according to the anomaly pattern to determine the parameter adjustment direction; the model adjustment strategy of the intelligent classification model is formulated based on the parameter adjustment direction and the anomaly data source.

[0053] In this embodiment, anomaly tracing is first performed based on anomaly classification information, using a causal inference algorithm to identify the specific data source causing the classification error. The causal inference algorithm analyzes the causal relationships between different features and variables in the model, gradually tracing the path of these anomalous data during the classification process, thereby determining which input data and features led to the classification anomaly. The input data causing the problem is referred to as the anomalous data source and marked with a specific anomaly source identifier. This identifier clearly indicates the cause of the anomaly, such as a specific combination of input variables or data characteristics under anomalous conditions.

[0054] Next, cluster analysis is performed using the DBSCAN clustering algorithm based on the anomaly source identifiers. DBSCAN identifies high-density areas within the anomalous data sources, i.e., anomalous data with similar characteristics and behaviors. Anomalous data are categorized into the same anomalous pattern, representing abnormal behavior caused by multiple data points under specific environmental or input conditions. For example, identifying a significant increase in the error rate of construction waste classification under certain temperature or humidity conditions allows clustering data with these common characteristics into an anomalous pattern.

[0055] After identifying anomalous patterns, the adjustment analysis phase begins, using LASSO regression analysis to analyze these patterns. LASSO regression incorporates L1 regularization to select and adjust the weights of model features. By analyzing anomalous patterns, it's determined which features play a major role in causing classification errors and further clarifies the direction of parameter adjustments. For example, the weights of certain features may need to be increased to ensure the model can better handle anomalous data sources; while the weights of other less important features need to be decreased to reduce their interference with the model's classification results.

[0056] Finally, combining the previously determined parameter adjustment direction and abnormal data sources, a model tuning strategy is formulated. The core of this strategy is to adjust the model's hyperparameters through optimization algorithms, such as genetic algorithms or random search, enabling the model to better handle these abnormal data sources. For example, genetic algorithms simulate the natural evolutionary process, continuously adjusting parameters to select the optimal solution. Through multiple iterations, the parameter combinations that maximize classification performance are found, thereby optimizing the overall model performance. The above process completes the formulation of the model tuning strategy for the intelligent classification model.

[0057] The classification processing module 15 executes the classification optimization decision to intelligently classify and process construction solid waste.

[0058] In this embodiment, the classification processing module receives classification optimization decisions and applies them to the actual waste classification task. At this time, the data on construction solid waste has already been preprocessed and classified and standardized according to the optimization decisions.

[0059] Next, the sorting and processing module utilizes intelligent sorting equipment, such as automated sorting machinery, to physically execute these decisions. The automated equipment allocates different categories of waste to corresponding processing areas based on pre-set sorting criteria and optimized parameters. For example, metal waste is identified and transported to the metal recycling area, while plastic waste is transported to the plastic sorting area. This process completes the intelligent sorting and processing of construction solid waste.

[0060] In summary, the embodiments of this application have at least the following technical effects: This application collects multidimensional data of construction solid waste from a target area to determine the dataset to be classified; identifies multiple classification labels and uses these labels to construct an intelligent classification model; synchronizes the dataset to be classified to the intelligent classification model to obtain classification decisions; simulates the execution of classification decisions to monitor the classification and processing of construction solid waste, generating operational feedback data; optimizes the intelligent classification model based on the operational feedback data to obtain an intelligent optimized classification model; synchronizes the dataset to be classified to the intelligent optimized classification model to obtain optimized classification decisions; and executes the optimized classification decisions to intelligently classify and process construction solid waste. This invention addresses the technical problems of low efficiency and low accuracy in the classification and processing of construction solid waste in existing technologies. Through multidimensional data collection, classification label construction, intelligent classification model training, simulated execution monitoring, feedback optimization, and final execution of optimized classification decisions, it achieves the technical effect of improving the accuracy and efficiency of construction solid waste classification.

[0061] Example 2, based on the same inventive concept as the intelligent sorting and processing system for construction solid waste in the foregoing examples, such as... Figure 2 As shown in the embodiment of this application, an intelligent sorting and treatment method for construction solid waste is provided, the method comprising: The process involves: collecting multidimensional data on construction solid waste from a target area to determine the dataset to be classified; identifying multiple classification labels and constructing an intelligent classification model using these labels; synchronizing the dataset to be classified to the intelligent classification model to obtain classification decisions; simulating the execution of the classification decisions to classify and process construction solid waste, generating operational feedback data; optimizing the intelligent classification model based on the operational feedback data to obtain an optimized intelligent classification model; synchronizing the dataset to be classified to the optimized intelligent classification model to obtain optimized classification decisions; and finally, executing the optimized classification decisions to intelligently classify and process construction solid waste.

[0062] Furthermore, regarding the multiple classification labels, the method also includes: Based on the dataset to be classified, data standardization is performed to obtain multiple standard data to be classified; multidimensional feature extraction is performed on the multiple standard data to be classified to determine multidimensional data features; feature annotation is performed on the multiple standard data to be classified according to the multidimensional data features to construct a multi-feature network; cluster analysis is performed on the multi-feature network to determine the multiple classification labels.

[0063] Furthermore, the method further includes constructing the intelligent classification model using the multiple classification labels: A multi-label task environment is set up. Based on the multi-label task environment, ensemble reinforcement learning is performed by traversing the multiple classification labels to obtain multiple reinforcement learners. The multiple reinforcement learners correspond to the multiple classification labels. Supervised training is performed based on the multiple reinforcement learners to generate a first training result. The first training result is then verified to generate a data verification token. The multiple reinforcement learners are integrated according to the data verification token to construct the intelligent classification model.

[0064] Furthermore, the method further includes synchronizing the dataset to be classified to the intelligent classification model to obtain a classification decision. Based on the multiple reinforcement learning machines, the system sequentially determines whether the dataset to be classified conforms to the target classification label; extracts the data to be classified that conforms to the target classification label, generates a positive reward instruction, and uses the positive reward instruction to identify the data to be classified that conforms to the target classification label to obtain a first classification result; extracts the data to be classified that does not conform to the target classification label, generates a negative reward instruction, and uses the negative reward instruction to identify the data to be classified that does not conform to the target classification label to obtain a second classification result; performs a weighted calculation based on the first classification result and the second classification result to generate multiple weight coefficients; and formulates the classification decision based on the multiple weight coefficients combined with the first classification result and the second classification result.

[0065] Furthermore, the method further includes simulating the execution of the classification decision to monitor the classification and treatment of construction solid waste and generating operational feedback data. Based on the dataset to be classified, the classification decision is simulated and a classification control instruction is generated; the sorting machinery is activated according to the classification control instruction, and multiple allocation areas are determined; the construction solid waste is classified and monitored according to the multiple allocation areas to obtain a classification monitoring dataset, which includes classification speed monitoring data, correct classification monitoring data, and incorrect classification monitoring data; the classification efficiency of the classification decision is calculated based on the classification speed monitoring data, the correct classification monitoring data, and the incorrect classification monitoring data, and a classification operation log is constructed; the classification operation log is added to the operation feedback data.

[0066] Furthermore, the intelligent classification model is optimized based on the operational feedback data to obtain an intelligent optimized classification model. The method further includes: Data mining is performed on the operational feedback data based on the classified operational logs to obtain multiple classification association rules; the operational feedback data is analyzed according to the multiple classification association rules to obtain operational analysis results; classification evaluation is performed based on the operational analysis results to obtain operational classification evaluation results; anomalies are captured by traversing the operational classification evaluation results to generate abnormal classification information; a model adjustment strategy is formulated based on the abnormal classification information, and the intelligent classification model is retrained by executing the model adjustment strategy to obtain a second training result; the data validity of the second training result is verified, and the intelligent optimized classification model is generated when the verification is successful.

[0067] Furthermore, based on the anomaly classification information, a model adjustment strategy is formulated, and the method further includes: Anomaly tracing is performed based on the anomaly classification information to determine the anomaly data source, which has an anomaly source identifier; anomaly pattern is determined based on the anomaly source identifier, and adjustment analysis is performed according to the anomaly pattern to determine the parameter adjustment direction; the model adjustment strategy of the intelligent classification model is formulated based on the parameter adjustment direction and the anomaly data source.

[0068] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0069] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0070] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. An intelligent sorting and processing system for construction solid waste, characterized in that, The system includes: The data traversal module traverses the target area to collect multidimensional data of construction solid waste and determines the dataset to be classified. The classification label determination module determines multiple classification labels, constructs an intelligent classification model using the multiple classification labels, and synchronizes the dataset to be classified to the intelligent classification model to obtain a classification decision. A classification and processing simulation module, which simulates the execution of the classification decision to classify and monitor construction solid waste, and generates operational feedback data. A classification optimization module optimizes the intelligent classification model based on the running feedback data to obtain an intelligent optimized classification model, and synchronizes the dataset to be classified to the intelligent optimized classification model to obtain a classification optimization decision; A classification processing module, which executes the classification optimization decision to intelligently classify and process construction solid waste; The classification label determination module is further used for: Based on the dataset to be classified, data standardization processing is performed to obtain multiple standard datasets to be classified. Multidimensional feature extraction is performed by traversing the multiple standard data to be classified to determine the multidimensional data features; Based on the multidimensional data features, feature annotation is performed on the multiple standard data to be classified to construct a multi-feature network; Cluster analysis is performed by traversing the multi-feature network to determine the multiple classification labels; The category label determination module is also used for: A multi-label task environment is set up. Based on the multi-label task environment, ensemble reinforcement learning is performed by traversing the multiple classification labels to obtain multiple reinforcement learners. The multiple reinforcement learners have a corresponding relationship with the multiple classification labels. Supervised training is performed based on the multiple reinforcement learners to generate a first training result, and the first training result is verified to generate a data verification token. The multiple reinforcement learners are integrated based on the data verification token to construct the intelligent classification model; The classification processing simulation module is further used for: Based on the dataset to be classified, the classification decision is simulated to generate classification control instructions; The sorting machine is activated according to the classification control command to determine multiple areas to be assigned. Construction solid waste is classified and monitored according to the multiple areas to be allocated, and a classification monitoring dataset is obtained. The classification monitoring dataset includes classification speed monitoring data, correct classification monitoring data, and incorrect classification monitoring data. Based on the classification speed monitoring data, the classification correctness monitoring data, and the classification error monitoring data, the classification efficiency of the classification decision is calculated, and a classification operation log is constructed. Add the categorized operation logs to the operation feedback data; The classification optimization module is further used for: Data mining is performed on the operational feedback data based on the classified operational logs to obtain multiple classification association rules; The operational feedback data is analyzed according to the multiple classification association rules to obtain operational analysis results; Based on the operational analysis results, a classification and evaluation are performed to obtain operational classification and evaluation results; The results of the operational classification evaluation are iterated to capture anomalies and generate anomaly classification information. Based on the anomaly classification information, a model adjustment strategy is formulated, and the intelligent classification model is retrained by executing the model adjustment strategy to obtain a second training result; The validity of the second training result is verified, and the intelligent optimized classification model is generated when the verification is successful.

2. The intelligent sorting and processing system for construction solid waste as described in claim 1, characterized in that, The system synchronizes the dataset to be classified to the intelligent classification model to obtain a classification decision. The system includes: Based on the multiple reinforcement learners, it is sequentially determined whether the dataset to be classified conforms to the target classification label; Extract the data to be classified that matches the target classification label, generate a positive reward instruction, and use the positive reward instruction to identify the data to be classified that matches the target classification label to obtain the first classification result; Extract the data to be classified that does not conform to the target classification label, generate a negative reward instruction, and use the negative reward instruction to identify the data to be classified that does not conform to the target classification label to obtain a second classification result; Based on the first classification result and the second classification result, a weighted calculation is performed to generate multiple weight coefficients; The classification decision is made based on the multiple weighting coefficients combined with the first classification result and the second classification result.

3. The intelligent sorting and processing system for construction solid waste as described in claim 1, characterized in that, The system includes a model adjustment strategy based on the anomaly classification information, comprising: Based on the anomaly classification information, anomaly source tracing is performed to determine the anomaly data source, and the anomaly data source has an anomaly source identifier. The anomaly pattern is determined based on the anomaly source identifier, and the adjustment analysis is performed according to the anomaly pattern to determine the direction of parameter adjustment; Based on the parameter adjustment direction and the abnormal data source, the model adjustment strategy of the intelligent classification model is formulated.

4. A method for intelligent sorting and processing of construction solid waste, characterized in that, The method is executed by an intelligent sorting and processing system for construction solid waste according to any one of claims 1 to 3, comprising: Collect multidimensional data of construction solid waste by traversing the target area to determine the dataset to be classified; Multiple classification labels are determined, an intelligent classification model is constructed using the multiple classification labels, and the dataset to be classified is synchronized to the intelligent classification model to obtain a classification decision. The classification decision is simulated to monitor the classification and treatment of construction solid waste, and operational feedback data is generated. The intelligent classification model is optimized based on the operational feedback data to obtain an intelligent optimized classification model. The dataset to be classified is then synchronized to the intelligent optimized classification model to obtain a classification optimization decision. The aforementioned classification optimization decision is used to intelligently classify and process construction solid waste.

Images