Real-time monitoring and optimization method and system for resource utilization of construction waste

Abstract

The present invention relates to a real-time monitoring and optimization method and system for resource utilization of construction waste. The method: collect data on the vibration frequency, material image and weight data of a crusher, a screening machine and a sorting device through a multi-source sensor network to obtain an original parameter data set; perform feature extraction to obtain a standardized feature data set; establish a crushing dynamics model, a screening efficiency model and a sorting process model, couple the material balance equation and the energy balance equation, use the least squares method and genetic algorithm to identify the model parameters, and obtain a process mathematical model; input the current process parameter combination and historical quality data into an ε-constrained optimizer to obtain a Pareto optimal solution for resource conversion rate and energy consumption; calculate the control amount of the actuator through a model predictive controller to generate an optimal process parameter combination. The implementation of the present invention realizes real-time optimization and adjustment of process parameters for resource utilization of construction waste.

Description

Technical Field

[0001] The present invention relates to the technical field of process optimization, and particularly to a real-time monitoring and optimization method and system for the resource utilization of construction waste. Background Art

[0002] Construction waste is one of the main solid wastes generated during the urban development process. Its composition is complex and the output is huge. If not properly treated, it will cause environmental pollution and resource waste. At present, the resource utilization of construction waste has become an important way to solve this problem. However, in the actual treatment process, due to the uncertainty of the material composition and properties, it is difficult to accurately control the process parameters of the treatment equipment.

[0003] Traditional construction waste treatment methods mainly rely on manual experience to adjust process parameters, lacking real-time monitoring and optimization means, and unable to respond to changes in material characteristics in a timely manner, resulting in low resource conversion efficiency and energy waste. At the same time, there are complex coupling relationships between multiple processes in the treatment process, and the operating parameters of each process affect each other, making it difficult to find the optimal combination of process parameters. In addition, existing construction waste treatment equipment generally has problems such as single monitoring means and insufficient data analysis ability, and is unable to conduct comprehensive status monitoring and quality control on the treatment process, resulting in large fluctuations in product quality and unstable resource utilization effects. Summary of the Invention

[0004] The main object of the present invention is to provide a real-time monitoring and optimization method and system for the resource utilization of construction waste, and the present invention realizes the real-time optimization adjustment of the process parameters for the resource utilization of construction waste.

[0005] To achieve the above object, the present invention provides a real-time monitoring and optimization method for the resource utilization of construction waste, including the following steps:

[0006] Perform resource treatment on construction waste based on the current process parameter combination, and collect data on the vibration frequency, material image, and weight data of the crusher, screen, and separation device through a multi-source sensor network to obtain an original parameter data set;

[0007] Input the original parameter data set into a multi-dimensional feature extraction model for feature extraction and feature standardization processing to obtain a standardized feature data set;

[0008] Establish a crushing dynamics model, a screening efficiency model, and a separation process model according to the standardized feature data set, couple them through a material balance equation and an energy balance equation, and use the least squares method and a genetic algorithm for model parameter identification to obtain a process mathematical model;

[0009] Perform autoregressive analysis based on the mathematical model of the process, and input the current process parameter combination and historical quality data into an ε-constraint optimizer to obtain the Pareto optimal solutions of resource conversion rate and energy consumption;

[0010] Generate a control strategy based on the Pareto optimal solutions, calculate the control quantity of the actuator through a model predictive controller, and perform process parameter adjustment by combining feedforward compensation and feedback compensation to generate an optimal process parameter combination.

[0011] The present invention also provides a real-time monitoring and optimization system for the resource utilization of construction waste, including:

[0012] An acquisition module, which is used to perform resource utilization processing on construction waste based on the current process parameter combination, and collect vibration frequency, material image, and weight data of crushers, screeners, and sorting devices through a multi-source sensing network to obtain an original parameter data set;

[0013] A feature extraction module, which is used to input the original parameter data set into a multi-dimensional feature extraction model for feature extraction and feature standardization processing to obtain a standardized feature data set;

[0014] A establishment module, which is used to establish a crushing dynamics model, a screening efficiency model, and a sorting process model according to the standardized feature data set, couple them through a material balance equation and an energy balance equation, and perform model parameter identification by using the least square method and the genetic algorithm to obtain a mathematical model of the process;

[0015] An analysis module, which is used to perform autoregressive analysis based on the mathematical model of the process, and input the current process parameter combination and historical quality data into an ε-constraint optimizer to obtain the Pareto optimal solutions of resource conversion rate and energy consumption;

[0016] A generation module, which is used to generate a control strategy based on the Pareto optimal solutions, calculate the control quantity of the actuator through a model predictive controller, and perform process parameter adjustment by combining feedforward compensation and feedback compensation to generate an optimal process parameter combination.

[0017] The present invention also provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of the method described in any one of the above are implemented.

[0018] The present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method described in any one of the above are implemented.

[0019] In summary, the technical solution provided by the present invention realizes the comprehensive monitoring of key processes such as crushing, screening, and sorting through a multi-source sensing network, obtains multi-dimensional data such as equipment vibration frequency, material images, and weights, and solves the problem of incomplete information acquisition in traditional single monitoring methods; uses a multi-dimensional feature extraction model to process the original parameter data, combines the multi-head attention mechanism and the graph convolutional network to extract deep features, and overcomes the defect of insufficient processing ability of traditional feature extraction methods for multi-source heterogeneous data; establishes a process mathematical model including a crushing dynamics model, a screening efficiency model, and a sorting process model, and realizes multi-process coupling modeling through the material balance equation and the energy balance equation; based on autoregressive analysis and the ε-constraint optimization method, realizes the multi-objective optimization of resource conversion rate and energy consumption, obtains the Pareto optimal solution set, and solves the problem that traditional single-objective optimization cannot balance multiple performance indicators; adopts a control strategy of combining a model predictive controller with feedforward compensation and feedback compensation to realize the real-time optimization adjustment of process parameters, and overcomes the disadvantage of lagging response of traditional control methods to changes in material characteristics. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a schematic diagram of the steps of a real-time monitoring and optimization method for the resource utilization of construction waste in an embodiment of the present invention;

[0021] Figure 2 is a block diagram of the structure of a real-time monitoring and optimization system for the resource utilization of construction waste in an embodiment of the present invention;

[0022] Figure 3 is a schematic block diagram of the structure of a computer device in an embodiment of the present invention.

[0023] The implementation, functional features, and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0024] In order to make the object, technical solution, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0025] Referring to Figure 1 , this embodiment provides a real-time monitoring and optimization method for the resource utilization of construction waste, including the following steps:

[0026] S1, perform resource treatment on construction waste based on the current process parameter combination, and collect vibration frequency, material image, and weight data of the crusher, screen, and sorting device through a multi-source sensing network to obtain an original parameter data set;

[0027] Among them, resource treatment of construction waste is carried out based on the current process parameter combination. During the treatment process, vibration sensors are installed at the main bearing seats of crushers, screening machines and sorting devices to collect the vibration signals of the equipment in real time, obtain the vibration frequency data of the equipment, and directly reflect the operating state of the equipment and the load change situation. At the same time, in order to comprehensively obtain the dynamic characteristics of the material, high-precision image sensors are installed at key positions of the material conveying channel, and the material is image-collected in a non-contact manner to generate material morphology data that can describe the particle size, shape and distribution characteristics of the material. This data can intuitively reflect the crushing and screening effects of the material and assist in identifying abnormal characteristics existing in the material. In order to improve the understanding of the flow characteristics of the material, weight sensors are installed on the conveyor belt to measure the weight data of the material in real time, calculate the flow rate information of the material, and characterize the dynamic changes during the material conveying and treatment process. Based on the equipment vibration frequency data, material morphology data and material flow rate data, a sensor calibration curve database is established. Through experimental calibration of the output characteristics of the sensor, the response curve and deviation parameters of the sensor are calculated and stored to generate a calibration parameter matrix. This matrix can quantify the deviation relationship between the sensor output signal and the actual physical quantity, ensuring the accuracy and reliability during the data acquisition process. After the sensor calibration is completed, the collected equipment vibration frequency data, material morphology data and material flow rate data are respectively subjected to compensation calculation and data integration using the calibration parameter matrix. The compensation calculation process corrects the deviation of the sensor output based on the calibration parameters to make it closer to the actual value; data integration is to match and synthesize the compensated data of multi-source sensors according to the time axis and the material flow path to form an original parameter data set containing rich physical information.

[0028] S2. Input the original parameter data set into a multi-dimensional feature extraction model for feature extraction and feature standardization processing to obtain a standardized feature data set;

[0029] Specifically, the device vibration frequency data in the original parameter dataset is input into the first graph convolutional layer of the multi-dimensional feature extraction model. This first graph convolutional layer performs a convolution operation on the data using 8 convolutional kernels of size 5×5 to extract key features in the vibration data and generate a vibration feature mapping matrix. This matrix can capture the complex dynamic characteristics during the operation of the device, including frequency changes and amplitude patterns, etc. At the same time, the material morphology data in the original parameter dataset is input into the second graph convolutional layer of the multi-dimensional feature extraction model. This layer contains 16 convolutional kernels of size 3×3. By extracting and enhancing the local features of the material image data, an image feature mapping matrix is generated, revealing the fine features of the material particle size, shape, and distribution, and helping to understand the morphology characteristics of the material and the potential processing effects. The material flow rate data in the original parameter dataset is input into the third graph convolutional layer, which is equipped with 32 convolutional kernels of size 7×7 to capture the time series features and trend changes of the material flow rate data. Through the convolution operation of this layer, a weight feature mapping matrix is generated to reflect the dynamic changes and flow states of the material flow rate. The vibration feature mapping matrix, the image feature mapping matrix, and the weight feature mapping matrix are input into the feature fusion layer in the multi-dimensional feature extraction model. In this layer, by adopting the multi-head attention mechanism to calculate the weights of different modality features, a multi-modal feature fusion matrix is formed. The introduction of the multi-head attention mechanism can dynamically allocate the weights of different features to highlight the contributions of each modality feature in specific tasks, and at the same time effectively reduce the risk of feature redundancy and information loss. The multi-modal feature fusion matrix is input into a deep neural network composed of three fully connected layers for processing. Among them, the first fully connected layer contains 512 neurons, the second contains 256 neurons, and the third contains 128 neurons, and each layer uses the LeakyReLU activation function to enhance the non-linear representation ability. In this process, the model can extract higher-level abstract information from the fused features to generate a target feature vector. To improve the applicability and consistency of the data, the target feature vector is normalized to eliminate the differences in scale and unit for subsequent analysis. At the same time, to enhance the ability to capture different frequency domain features, multi-scale wavelet decomposition is used to decompose and reconstruct the target feature vector to generate the final normalized feature dataset.

[0030] S3. Establish a crushing dynamics model, a screening efficiency model, and a separation process model based on the normalized feature dataset. Couple them through the material balance equation and the energy balance equation, and use the least squares method and the genetic algorithm for model parameter identification to obtain the mathematical model of the process.

[0031] It should be noted that time series analysis is performed on the data of the crushing process in the standardized characteristic dataset. Based on the material particle size distribution function and the crushing power function, a kinetic equation describing the crushing process is constructed. In this process, the Bond power calculation formula is used to evaluate the crushing energy consumption per unit time, accurately quantify the energy consumption characteristics of the crushing equipment, and generate a crushing kinetic model, which can reflect the relationship between the change of material particle size and the crushing energy consumption. Probability analysis is performed on the data of the screening process in the standardized characteristic dataset. The passing rate of the material through the sieve is calculated through the Rosin-Rammler particle size distribution function, and combined with the material stacking thickness function, a kinetic equation describing the screening process is established. This screening efficiency model reveals the comprehensive influence of the sieve opening rate, vibration frequency, and material flow rate on the screening effect. At the same time, trajectory analysis is performed on the data of the sorting process in the standardized characteristic dataset. Based on the multi-body system motion equation, the motion trajectory of the material in the sorting device is calculated. Combining the material characteristics and device parameters, a sorting kinetic equation is constructed through the separation function to generate a sorting process model. This model can quantify the sorting accuracy and evaluate the separation effect of materials with different particle sizes or densities. The output parameters of the crushing kinetic model are combined with the input parameters of the screening efficiency model to establish a material balance equation. By calculating the material flow rate of each process and using the iterative method to solve the material balance constraint relationship, a material balance equation set is formed. This process can ensure the consistency of the material flow rate between different processes and ensure the clear physical meaning of the process model. According to the energy consumption parameters of the crushing kinetic model, the screening efficiency model, and the sorting process model, an energy balance equation is constructed, the energy transfer efficiency of each process is calculated, and the matrix method is used to solve the energy balance constraint relationship to obtain an energy balance equation set. The introduction of the energy balance equation can effectively describe the energy distribution and conversion efficiency in the process flow and provide a basis for optimizing energy consumption. The material balance equation set and the energy balance equation set are integrated into an objective function, and optimization variables are set, including the crushing ratio, screening efficiency, and sorting accuracy, etc. The model parameters are optimized through the genetic algorithm. In the optimization process, the genetic algorithm performs a global search on the objective function by simulating the mechanisms of natural selection and genetic variation to obtain the optimal solution of the model parameters. On this basis, the least squares method is used to further fit the optimized solution of the model parameters to ensure the accuracy and stability of parameter identification. Based on the results of model parameter identification, the parameters of the crushing kinetic model, the screening efficiency model, and the sorting process model are updated. Combining the updated models, the coupling relationship between process parameters is calculated through the material balance equation set and the energy balance equation set to generate a mathematical model of the process. This model can comprehensively describe the dynamic behavior and mutual influence of each process in the process of construction waste resource utilization.

[0032] S4. Perform autoregressive analysis based on the mathematical model of the process, and input the current process parameter combination and historical quality data into the ε-constraint optimizer to obtain the Pareto optimal solution of the resource conversion rate and energy consumption;

[0033] Specifically, based on the mathematical model of the process, the objective functions of resource conversion rate and energy consumption are calculated, and these two objective functions are weighted and combined to construct a multi-objective optimization function. In this process, the objective function of resource conversion rate evaluates the process effect by measuring the resource recovery efficiency in the process of recycling construction waste, while the objective function of energy consumption evaluates the system energy efficiency by calculating the energy consumption per unit resource conversion in the process. The multi-objective optimization function with weighted combination dynamically adjusts the weights of different optimization objectives to achieve a balance between resource efficiency and energy consumption. After constructing the optimization function, a p-order autoregressive model is established based on the current process parameter combination to analyze the time series characteristics of the process parameters. The order of the autoregressive model is evaluated and optimized using the Akaike information criterion to determine the optimal autoregressive model structure. After determining the model structure, the historical quality data is input into the autoregressive model, and a complete autoregressive prediction model is generated by calculating the autoregressive coefficients and white noise variances. This prediction model can use the current process parameter combination and historical data to perform rolling predictions on future process parameter changes, obtaining a series of process parameter prediction sequences. The process parameter prediction sequences are input into the ε-constraint optimizer, and the constraint intervals for resource conversion rate and energy consumption are set. The setting of the constraint intervals is based on actual process requirements and performance standards to ensure that the optimization results meet both the quality requirements of resource utilization and the energy consumption control objectives. During the optimization process, the ε-constraint optimizer comprehensively scans the process parameter space through the grid search method to construct the corresponding constraint condition matrix. This matrix contains the optimization information of each parameter combination under different constraint conditions. The constraint condition matrix is segmented and optimized, and the non-dominated sorting method is used to sort the segmented optimal solution sets. The non-dominated sorting effectively decomposes the better solutions in the solution set and selects elite individuals based on the regions with larger crowding degrees to generate the non-dominated solution set. The solutions in the non-dominated solution set represent the optimization results between resource conversion rate and energy consumption under different trade-off conditions, ensuring that the optimization solutions can achieve a balance among multiple objectives. After obtaining the non-dominated solution set, in order to improve the accuracy and stability of the optimization solutions, a simulated annealing strategy is used to perform local search optimization on the neighborhood of the solutions. The simulated annealing strategy simulates the temperature change in the physical annealing process and accepts worse solutions with a certain probability to jump out of the local optimal trap and improve the global search ability. During the process of optimizing in the neighborhood of the solutions, it gradually converges to the Pareto optimal solutions of resource conversion rate and energy consumption.

[0034] Input the process parameter prediction sequence into the ε-constraint optimizer for normality test to evaluate the distribution characteristics of the parameters. Through the normality test of the process parameters, determine whether the prediction sequence conforms to the normal distribution or other distribution forms. Based on the obtained parameter distribution characteristics, normalize the process parameter prediction sequence, map different parameters to a unified standardized range, eliminate the differences in data scales, and generate a standardized prediction sequence. On this basis, use the standardized prediction sequence to calculate the constraint boundaries. By analyzing the resource conversion rate target and combining the characteristics of the process flow, calculate and determine the constraint interval of the resource conversion rate; at the same time, conduct energy consumption analysis on the standardized prediction sequence, and calculate the constraint interval of energy consumption by quantifying the energy usage characteristics of each process. The determination of these two constraint intervals provides clear upper and lower boundaries for the optimization process, ensuring that the physical meaning of the optimization solution is reasonable and meets the requirements of practical applications. After the constraint intervals are determined, generate grid points based on the resource conversion rate constraint interval and the energy consumption constraint interval, and discretize the continuous constraint space into a finite set of discrete constraint points. The generation of the discrete constraint point set needs to evenly cover the entire constraint space, so as to provide a comprehensive search range for subsequent optimization. Input the discrete constraint point set into the ε vector generator to generate the ε-constraint vector by equidistant sampling. The generation of the ε-constraint vector is based on the distribution between points in the discrete constraint point set, and ensures that the optimizer can explore all possible solutions in the constraint space with a uniform step size by equidistant sampling. Perform a combined operation on the generated ε-constraint vector and the discrete constraint point set. By matching each constraint vector with the corresponding constraint point, form a complete constraint condition matrix. This constraint condition matrix not only contains the constraint boundary information of the resource conversion rate and energy consumption, but also reflects the optimization potential of the process parameter prediction sequence under different constraint conditions. Through this matrix, the ε-constraint optimizer can efficiently perform the optimization search in the constraint space, ensuring that the optimization result not only meets the boundary requirements of the resource conversion rate and energy consumption, but also achieves the global optimum in the objective function.

[0035] S5. Generate a control strategy based on the Pareto optimal solution, calculate the control amount of the actuator through the model predictive controller, and combine feedforward compensation and feedback compensation to adjust the process parameters to generate the optimal process parameter combination.

[0036] Among them, the Pareto optimal solution is input into the model predictive controller. Inside the model predictive controller, a quadratic objective function is adopted to construct the control performance index, and the objective function is used to balance the relationship between the resource conversion rate and energy consumption, while minimizing the control error and control increment. By analyzing the dynamic characteristics of the step response, the control increment sequence of the system response is determined, and this sequence reflects the adjustment strategy of the actuator at different time steps. Based on the control increment sequence, a feedforward compensation model is constructed. By analyzing the influence of the system input change on the output performance, feedforward compensation calculations are performed on the key process parameters of the crusher, screen, and sorting device. These parameters include the rotation speed of the crusher, the vibration frequency of the screen, and the sorting ratio of the sorting device. The parameter matrix obtained through the feedforward compensation calculation can pre-correct the potential disturbance effect, improve the system's ability to suppress input disturbances, and generate a feedforward compensation parameter matrix. The feedforward compensation parameter matrix is input into the feedback compensator to dynamically correct the deviation between the actual output and the desired output of the system. By collecting the device operation data in real time and comparing it with the output predicted by the model, the feedback compensator calculates the feedback compensation amount for each actuator to correct the deviation of the actual system, and finally obtains a feedback compensation parameter matrix. The introduction of feedback compensation significantly improves the stability and robustness of the system, ensuring that the process adjustment can adapt to complex dynamic environments. After the feedforward compensation parameter matrix and the feedback compensation parameter matrix are generated, the two are organically combined through compensation control calculation to generate a compensation control parameter matrix. This matrix synthesizes the feedforward and feedback information and can coordinate the predictability and real-time nature of the control strategy. Based on the compensation control parameter matrix, reasonable constraint conditions are set for the rotation speed of the crusher, the vibration frequency of the screen, and the sorting ratio parameter of the sorting device to optimize the adjustment range. These constraint conditions construct a parameter optimization interval to ensure that the finally generated process parameter combination not only meets the actual operation requirements but also conforms to the device performance limitations. By combining the compensation control parameter matrix with the parameter optimization interval, the system parameters are further adjusted through an optimization solution method, and a constrained optimization algorithm is used to search for the optimal solution in the objective function. During this process, by balancing the priorities of the resource conversion rate and energy consumption, the key process parameters of the crusher, screen, and sorting device are optimized, and finally the optimal process parameter combination is obtained. This parameter combination can minimize energy consumption to the greatest extent while meeting the process efficiency requirements, ensuring that the construction waste resource utilization process achieves optimal performance.

[0037] In one example, construction waste resource treatment is carried out based on the current process parameter combination, and data such as the vibration frequency, material image, and weight data of the crusher, screen, and sorting device are collected through a multi-source sensing network to obtain an original parameter data set, including:

[0038] Conduct resource treatment of construction waste based on the current process parameter combination. Install vibration sensors at the main bearing seats of crushers, screening machines, and sorting devices to collect vibration signals and obtain equipment vibration frequency data;

[0039] Install image sensors at the material conveying channels to collect material image data, and install weight sensors on the conveyor belts to collect material weight data and obtain material flow data;

[0040] Based on the equipment vibration frequency data, material morphology data, and material flow data, establish a sensor calibration curve database and perform calibration parameter calculations to obtain a calibration parameter matrix;

[0041] Based on the calibration parameter matrix, perform compensation calculations and data integration on the equipment vibration frequency data, material morphology data, and material flow data respectively to obtain an original parameter dataset.

[0042] In this example, vibration sensors are installed at the main bearing seats of crushers, screening machines, and sorting devices to collect vibration signals during equipment operation. These signals can reflect the working state of the equipment and the impact characteristics of the material, generating equipment vibration frequency data. The vibration signals are analyzed in the time domain and frequency domain to obtain key features. For example, through the fast Fourier transform, the time series signal is transformed into a frequency domain signal , and its calculation formula is:

[0043] ;

[0044] where represents the frequency domain signal, is time, is frequency, is the imaginary unit. Through the fast Fourier transform, the main frequency components in the equipment vibration signals are extracted to analyze the stability and abnormal states of equipment operation. At the same time, high-precision image sensors are installed at the material conveying channels to perform real-time image acquisition of the conveyed materials, generating material morphology data. The image data extracts the particle size and shape characteristics of the materials through edge detection algorithms. For example, based on the Canny edge detection algorithm, the edge intensity of the material morphology is calculated through gradient calculation, and its formula is:

[0045] ;

[0046] where is the image gray value, and They are the horizontal and vertical coordinates of the image respectively. By statistically analyzing the edge features, the particle size distribution and shape factor are calculated. To obtain the material flow rate data, a weight sensor is installed on the conveyor belt. According to Hooke's law, the weight sensor obtains the real-time weight of the material by measuring the change in the force exerted by the material on the elastic element of the sensor. If the sensor output voltage signal is , then the flow rate is calculated through the sensor sensitivity and the material velocity as follows:

[0047] ;

[0048] where is the sensitivity coefficient of the sensor. By measuring the conveyor belt speed and the sensor output signal in real time, the instantaneous flow rate of the material is accurately calculated. After collecting the vibration frequency, image, and weight data, a sensor calibration curve database is established based on these data. The calibration process measures the deviation relationship between the actual output and the theoretical value of the sensor through experiments and constructs a calibration model. For example, assuming the output of the vibration sensor is , and the actual vibration acceleration is , then the calibration curve is represented by a linear model as:

[0049] ;

[0050] where is the gain coefficient, and is the offset. By fitting the experimental data using the least squares method, the calibration parameters and are obtained. Similarly, the image and weight sensors are calibrated to obtain a complete calibration parameter matrix , and each element of the matrix corresponds to a set of calibration parameters for a sensor. Based on the calibration parameter matrix, compensation calculations and data integration are performed on the collected data. For example, the compensation of the vibration frequency data is achieved through the following formula:

[0051] ;

[0052] where is the compensated vibration acceleration, and is the uncompensated sensor output data. Similarly, the particle size distribution in the image data and the flow rate in the weight data are compensated to generate a dataset with higher consistency. By integrating the compensated vibration frequency data, material morphology data, and material flow rate data, a complete original parameter dataset is generated. For example, using a data fusion algorithm to combine these multi-modal data to construct a high-dimensional feature vector :

[0053] ;

[0054] Among them, is the vibration feature, is the morphology feature, is the flow rate feature. After this fusion, the original parameter dataset is obtained.

[0055] In one example, the original parameter dataset is input into a multi-dimensional feature extraction model for feature extraction and feature standardization processing, and a standardized feature dataset is obtained, including:

[0056] The device vibration frequency data in the original parameter dataset is input into the first graph convolutional layer in the multi-dimensional feature extraction model. The first graph convolutional layer contains 8 convolutional kernels of 5×5, and a vibration feature mapping matrix is obtained;

[0057] The material morphology data in the original parameter dataset is input into the second graph convolutional layer in the multi-dimensional feature extraction model. The second graph convolutional layer contains 16 convolutional kernels of 3×3, and an image feature mapping matrix is obtained;

[0058] The material flow rate data in the original parameter dataset is input into the third graph convolutional layer in the multi-dimensional feature extraction model. The third graph convolutional layer contains 32 convolutional kernels of 7×7, and a weight feature mapping matrix is obtained;

[0059] The vibration feature mapping matrix, the image feature mapping matrix, and the weight feature mapping matrix are input into the feature fusion layer in the multi-dimensional feature extraction model. The feature weights are calculated through the multi-head attention mechanism, and a multi-modal feature fusion matrix is obtained;

[0060] The multi-modal feature fusion matrix is input into the three-layer fully connected layer in the multi-dimensional feature extraction model. The number of neurons in the first fully connected layer of the three-layer fully connected layer is 512, the number of neurons in the second fully connected layer is 256, and the number of neurons in the third fully connected layer is 128. Each layer uses the LeakyReLU activation function to obtain the target feature vector;

[0061] The target feature vector is standardized and multi-scale wavelet decomposed, and the standardized feature dataset is reconstructed.

[0062] In this example, the device vibration frequency data in the original parameter dataset is input into the first graph convolutional layer in the multi-dimensional feature extraction model. The graph convolutional layer is designed to contain 8 convolutional kernels of size , and its function is to extract the spatial features in the vibration data through the local receptive field. Suppose the vibration data is represented in the form of a two-dimensional matrix , where is the time step, is a frequency component, and the output feature map matrix of the convolution operation is expressed as:

[0063] ;

[0064] where is the weight of the th convolution kernel, is the bias term, is the activation function (such as ReLU). Through this convolution operation, features related to the dynamic characteristics of equipment operation are extracted from vibration data, such as frequency fluctuation patterns and vibration intensity. The material morphology data in the original parameter dataset is input into the second graph convolutional layer in the multi-dimensional feature extraction model. This convolutional layer contains 16 convolutional kernels of size for extracting important features such as edges and textures in the material image. Let the material morphology data be a two-dimensional image matrix , and the formula for the image feature map matrix obtained after the convolution operation is:

[0065] ;

[0066] where and are respectively the weight and bias term of the th convolution kernel, is also the activation function. This feature map matrix can capture the material particle size distribution and shape information. At the same time, the material flow rate data in the original parameter dataset is input into the third graph convolutional layer. This layer is designed with 32 convolutional kernels of size specifically for extracting spatio-temporal features in the flow rate data. Let the flow rate data be a matrix , where is the time step, is the number of sensors, and the weight feature map matrix generated by the convolution operation is expressed as:

[0067] ;

[0068] where is the weight of the th convolution kernel, is the bias term, is the activation function. Through this processing method, the periodic fluctuation characteristics of the flow rate and the conveying dynamic mode are extracted. The vibration feature map matrix , the image feature map matrix and the weight feature map matrix Input into the feature fusion layer, calculate the weights of each modality feature through the multi-head attention mechanism, and generate a multi-modal feature fusion matrix. Let the input of the feature fusion layer be and the output be the fusion matrix . The calculation formula of the multi-head attention mechanism is:

[0069] ;

[0070] ;

[0071] where , , are the weight matrices of query, key, and value respectively, is the dimension of the key, is the number of attention heads. Through the attention mechanism, the importance of each modality feature is dynamically allocated. Input the multi-modal feature fusion matrix into a three-layer fully connected layer network. The number of neurons in the first layer of the fully connected layer is 512, the second layer is 256, and the third layer is 128. Each layer uses the LeakyReLU activation function, and its non-linear transformation formula is:

[0072] ;

[0073] where is the negative slope coefficient, usually taking a value of 0.01. After the non-linear mapping of the fully connected layer, generate the target feature vector , which contains the comprehensive features of all modalities. Perform normalization processing on the target feature vector to eliminate the numerical scale difference, and perform frequency domain decomposition on the features through multi-scale wavelet decomposition. Let the input of the wavelet decomposition be , and the output be the decomposition coefficient and the approximation coefficient . The decomposition formula is:

[0074] ;

[0075] where is the wavelet function, is the scaling function, is the decomposition level. By reconstructing the decomposed features, generate a normalized feature dataset for subsequent model training and optimization.

[0076] In one example, establish a fragmentation dynamics model, a screening efficiency model, and a separation process model based on the normalized feature dataset. Couple through the material balance equation and the energy balance equation, and use the least squares method and the genetic algorithm for model parameter identification to obtain the mathematical model of the process, including:

[0077] Perform time series analysis on the data of the crushing process in the standardized characteristic dataset, establish a crushing kinetics equation based on the material particle size distribution function and the crushing power function, calculate the crushing energy consumption per unit time using the Bond power calculation formula, and obtain the crushing kinetics model;

[0078] Perform probability analysis on the data of the screening process in the standardized characteristic dataset, calculate the material passing rate based on the Rosin-Rammler particle size distribution function, and establish a screening kinetics equation in combination with the material stacking thickness function to obtain the screening efficiency model;

[0079] Perform trajectory analysis on the data of the separation process in the standardized characteristic dataset, calculate the material separation trajectory based on the multi-body system motion equation, construct a separation kinetics equation through the separation degree function, and obtain the separation process model;

[0080] Establish a material balance equation based on the output parameters of the crushing kinetics model and the input parameters of the screening efficiency model, calculate the material flow rate of each process, and use the iterative method to solve the material balance constraint relationship to obtain the material balance equations;

[0081] Establish an energy balance equation based on the energy consumption parameters of the crushing kinetics model, the screening efficiency model, and the separation process model, calculate the energy transfer efficiency of each process, and use the matrix method to solve the energy balance constraint relationship to obtain the energy balance equations;

[0082] Construct an objective function from the material balance equations and the energy balance equations, set the optimization variables including the crushing ratio, screening efficiency, and separation accuracy, and perform parameter optimization through the genetic algorithm to obtain the optimized solution of the model parameters;

[0083] Use the least squares method to perform parameter fitting on the optimized solution of the model parameters to obtain the parameter identification result of the model, and update the parameters of the crushing kinetics model, the screening efficiency model, and the separation process model based on the parameter identification result of the model. Calculate the coupling relationship between the process parameters through the material balance equations and the energy balance equations to obtain the mathematical model of the process.

[0084] In this example, time series analysis is performed on the data of the crushing process involved in the standardized characteristic dataset to establish a crushing kinetics model. The crushing kinetics model describes the energy consumption and particle size change characteristics of the material during the crushing process through the material particle size distribution function and the crushing power function. The particle size distribution is expressed by the Rosen-Rammler function as:

[0085] ;

[0086] where, is the percentage of materials with a particle size smaller than , is the characteristic particle size, is the distribution index, which describes the steepness of the particle size distribution. The crushing power consumption is based on the Bond power calculation formula:

[0087] ;

[0088] where is the crushing energy consumption per unit time, is the Bond work index, are the 80% passing particle sizes of the material before and after crushing respectively. By combining the particle size distribution function and the Bond power formula, a crushing kinetics equation is established:

[0089] ;

[0090] where is the crushing rate constant, is the time, and the model describes the dynamic relationship between the particle size change and the energy consumption. After completing the crushing kinetics modeling, a probability analysis is performed on the screening process data. The screening efficiency depends on the probability of the material passing through the sieve and is related to the Rosin-Rammler particle size distribution function. The material passing rate is expressed as:

[0091] ;

[0092] where is the probability of the material with particle size passing through the sieve, is the characteristic size of the sieve hole, is the screening index. Combining the material packing thickness function , the screening kinetics model is expressed as:

[0093] ;

[0094] where is the distance in the screening process, is the screening efficiency coefficient. Through this model, the separation efficiency of the material particle size in the screening process is predicted. For the sorting process, the material sorting trajectory is calculated based on the motion equation of the multi-body system. Assuming that during the sorting process, the material is under the action of gravity and fluid force, the trajectory is described by the following equation:

[0095] ;

[0096] where is the material mass, is the material position, is the gravity, is the fluid resistance, is the resistance coefficient, is the fluid density, is the windward area of the material, is the relative velocity. The separation efficiency is calculated through the separation function:

[0097] ;

[0098] where is the separation, and are the average positions of the two materials respectively, and are the standard deviations of the positions. The separation kinetic model describes the relationship between the material trajectory and the separation efficiency. By coupling the output parameters of the crushing kinetic model and the input parameters of the screening efficiency model, a material balance equation is established. The material flow follows the continuity principle at each link and is expressed by the following formula:

[0099] ;

[0100] where and are the input and output flows of each node respectively. By solving the above equation by the iterative method, the material balance constraint relationship is obtained, and a complete set of material balance equations is formed. On this basis, combined with the energy consumption parameters of the crushing, screening and separation models, an energy balance equation is established. Let the input energy of the system be , the output energy be , and the intermediate energy loss be , then the energy conservation relationship is:

[0101] ;

[0102] By solving this set of equations by the matrix method, the energy transfer efficiency of each process is calculated, and a complete set of energy balance equations is formed. By combining the material balance equations and the energy balance equations, an objective function is constructed, with the crushing ratio, screening efficiency and separation accuracy as the optimization variables, and the genetic algorithm is used to optimize the objective function. The genetic algorithm finds the optimal solution through selection, crossover and mutation operations. After optimization, the least squares method is used to fit the parameter optimization solution to generate the final model parameter identification result, and the dynamic behaviors of the crushing, screening and separation links are recalculated based on the updated model parameters.

[0103] In an example, autoregressive analysis is performed based on the process mathematical model, and the current process parameter combination and historical quality data are input into the ε-constraint optimizer to obtain the Pareto optimal solutions of resource conversion rate and energy consumption, including:

[0104] Based on the process mathematical model, the objective functions of resource conversion rate and energy consumption are calculated and weighted combined to obtain a multi-objective optimization function;

[0105] Construct a p-order autoregressive model based on the current process parameter combination, determine the optimal order through the AIC criterion, and obtain the autoregressive model structure;

[0106] Input the historical quality data into the autoregressive model structure, calculate the autoregressive coefficients and the white noise variance, obtain the autoregressive prediction model, and perform rolling prediction on the process parameters based on the autoregressive prediction model to obtain the process parameter prediction sequence;

[0107] Input the process parameter prediction sequence into the ε-constraint optimizer, set the resource conversion rate constraint interval and the energy consumption constraint interval, perform grid search, and obtain the constraint condition matrix;

[0108] Perform piecewise optimization on the constraint condition matrix to obtain the piecewise optimal solution set, perform non-dominated sorting on the piecewise optimal solution set, select the solutions with a larger crowding degree as elite individuals, and obtain the non-dominated solution set;

[0109] Perform local search optimization on the non-dominated solution set, and use the simulated annealing strategy to search for the optimal solution in the neighborhood of the solution to obtain the Pareto optimal solutions of the resource conversion rate and the energy consumption.

[0110] In this example, define the resource conversion rate objective function and the energy consumption objective function. These two objective functions respectively describe the economic efficiency and the energy consumption level of the process, and have the characteristic of mutual restriction. The resource conversion rate objective function is expressed by the following formula:

[0111] ;

[0112] Among them, is the resource conversion rate, and are the output and input material qualities respectively, is the optimization variable, including process parameters such as the crushing ratio, screening efficiency, and separation accuracy. The energy consumption objective function is expressed as:

[0113] ;

[0114] Among them, is the total energy consumption, is the power of the th process link, is the corresponding running time, is the total number of links. By performing weighted combination on these two objective functions, construct a multi-objective optimization function:

[0115] ;

[0116] Among them, and is the weight coefficient, which is used to adjust the balance between resource efficiency and energy consumption. An autoregressive model of order p is constructed based on the current process parameter combination to analyze the time series characteristics of the process parameters. The expression form of the autoregressive model is:

[0117] ;

[0118] where, is the value of the time series at time , are the autoregressive coefficients, is white noise, is the model order. The optimal order is determined by the Akaike information criterion (AIC):

[0119] ;

[0120] where, is the likelihood function value of the model. The order with the minimum AIC value is selected as the optimal order, thereby determining the autoregressive model structure. After obtaining the autoregressive model structure, the historical quality data is input into the model, and the autoregressive coefficients and white noise variance are calculated by the least squares method to generate an autoregressive prediction model. Based on this model, the future process parameters are predicted recursively to generate a process parameter prediction sequence:

[0121] ;

[0122] where, is the predicted value at time . The predicted process parameter sequence is input into the -constrained optimizer. The constraint intervals of the resource conversion rate and energy consumption are set respectively, and grid search is performed to generate a constraint condition matrix. Grid search discretizes the variable space, covers all possible constraint combinations, and constructs a constraint condition in the form of a multi-dimensional matrix. The constraint condition matrix is optimized and solved in segments to obtain a series of segment-optimal solution sets. The segment-optimal solution sets are screened by non-dominated sorting, and those solutions with no obvious disadvantages in terms of the resource conversion rate and energy consumption objectives are preferentially selected. For regions with a large crowding degree, more solutions are selected as elite individuals to generate a non-dominated solution set:

[0123] ;

[0124] where F is the non-dominated solution set, and are solution vectors respectively, and the symbol represents the domination relationship. The simulated annealing strategy is used for local search optimization of the non-dominated solution set. Simulated annealing jumps and searches in the neighborhood through the following acceptance probability :

[0125] ;

[0126] Among them, is the difference in the objective function values between the current solution and the new solution, is the current temperature. By gradually decreasing the temperature, the algorithm gradually converges to the Pareto optimal solution of resource conversion rate and energy consumption.

[0127] In one example, the process parameter prediction sequence is input into the ε-constraint optimizer, the resource conversion rate constraint interval and the energy consumption constraint interval are set, and grid search is performed to obtain the constraint condition matrix, including:

[0128] The process parameter prediction sequence is input into the ε-constraint optimizer for normality test to obtain the parameter distribution characteristics;

[0129] Based on the parameter distribution characteristics, the process parameter prediction sequence is normalized to obtain the standardized prediction sequence, and based on the standardized prediction sequence, the constraint boundary is calculated to obtain the resource conversion rate constraint interval. At the same time, the energy consumption analysis is performed on the standardized prediction sequence to obtain the energy consumption constraint interval;

[0130] Based on the resource conversion rate constraint interval and the energy consumption constraint interval, grid points are generated to obtain the discrete constraint point set, and the discrete constraint point set is input into the ε vector generator for equidistant sampling to obtain the ε-constraint vector;

[0131] The ε-constraint vector and the discrete constraint point set are combined and calculated to obtain the constraint condition matrix.

[0132] In this example, the process parameter prediction sequence is subjected to a normality test to obtain the distribution characteristics of the parameters. Let the process parameter prediction sequence be , and the normality test is performed through the Shapiro-Wilk test. The calculation formula of its statistic is:

[0133] ;

[0134] Among them, is the sorted sample value, is the sample mean, is a constant generated based on the normal distribution. The test result determines whether the parameter follows a normal distribution by comparing with the significance level. If it does not conform to the normal distribution, it is transformed into a normal distribution through methods such as Box-Cox transformation. The formula is:

[0135] ;

[0136] Among them, is a transformation parameter, determined by maximum likelihood estimation. After obtaining the parameter distribution characteristics, the process parameter prediction sequence is normalized to eliminate the scale influence of the data and improve the optimization efficiency. The normalization is achieved through the min-max normalization formula:

[0137] ;

[0138] where, is the value after normalization, and are the minimum and maximum values of the sequence respectively. After normalization, the standardized prediction sequence is obtained, and its range is uniformly [0, 1]. Based on the standardized prediction sequence, the constraint boundaries of the resource conversion rate and energy consumption are calculated. The constraint interval of the resource conversion rate is determined by the mean and standard deviation of the prediction sequence, and the constraint interval is expressed as:

[0139] ;

[0140] where, is the interval expansion coefficient, usually taking 2 or 3 to cover more than 95% of the sample values. Similarly, the energy consumption constraint interval is calculated, and the energy consumption interval is:

[0141] ;

[0142] where, and are the mean and standard deviation of the energy consumption prediction value respectively. After obtaining the constraint intervals of the resource conversion rate and energy consumption, the intervals are discretized by the grid point generation method to obtain the discrete constraint point set. Assuming that the interval is divided into uniform grids, the discrete points are expressed as:

[0143] ;

[0144] where, . The discrete point set is all possible combinations. The discrete constraint point set is input into the ε vector generator for equidistant sampling to generate the ε constraint vector. Let the ε vector be , where is the equidistant sampling point, and the calculation formula is:

[0145] ;

[0146] where, , and are the minimum and maximum values of the ε vector. The ε constraint vector and the discrete constraint point set are combined and operated to generate a constraint condition matrix. The constraint condition matrix is expressed as:

[0147] ;

[0148] Each matrix element represents a combination of specific resource conversion rate, energy consumption, and ε value.

[0149] In one example, a control strategy is generated based on the Pareto optimal solution, and the control quantity of the actuator is calculated by a model predictive controller. The process parameters are adjusted by combining feedforward compensation and feedback compensation to generate an optimal process parameter combination, including:

[0150] The Pareto optimal solution is input into the model predictive controller, a quadratic objective function is used to construct a control performance index, and a step response analysis is performed to obtain a control increment sequence;

[0151] A feedforward compensation model is constructed based on the control increment sequence, and feedforward compensation calculations are performed on the rotational speed, vibration frequency, and separation ratio parameters of the crusher, screening machine, and separation device to obtain a feedforward compensation parameter matrix;

[0152] The feedforward compensation parameter matrix is input into the feedback compensator to calculate the feedback compensation amount of each actuator, and a feedback compensation parameter matrix is obtained;

[0153] Compensation control calculations are performed based on the feedforward compensation parameter matrix and the feedback compensation parameter matrix to obtain a compensation control parameter matrix;

[0154] The compensation control parameter matrix is optimized. By setting the rotational speed range of the crusher, the vibration frequency range of the screening machine, and the separation ratio range of the separation device, constraint conditions are constructed to obtain a parameter optimization interval;

[0155] Based on the compensation control parameter matrix and the parameter optimization interval, an optimal solution is obtained to get the optimal process parameter combination.

[0156] In this example, the Pareto optimal solution is passed as an input to the model predictive controller to achieve multi-objective regulation. The model predictive controller constructs a control performance index through a quadratic objective function to minimize the sum of the squares of the target deviation and the control increment. The objective function is expressed as:

[0157] ;

[0158] where is the objective function value, is the system output at time , is the target reference value, is the control increment, is the weight factor for controlling the increment, is the prediction step. By optimizing this objective function, the control increment sequence is calculated, which describes the adjustment path for the system to gradually approach the target state from the current state. Based on the control increment sequence, a feedforward compensation model is constructed to perform feedforward compensation calculations for the key parameters (such as rotational speed, vibration frequency, and separation ratio) of crushers, screens, and sorting devices. Let the input disturbance be , the current process parameters be , and the target process parameters be . The feedforward compensation relationship is expressed as:

[0159] ;

[0160] where is the parameter value after feedforward compensation, is the feedforward compensation gain matrix. By combining the control increment sequence, the feedforward compensation values for each device are calculated to generate the feedforward compensation parameter matrix , which describes the pre-adjustment strategies for different devices. The feedforward compensation parameter matrix is input into the feedback compensator to correct the system deviation in real time. The feedback compensator calculates the feedback compensation amount by collecting the deviation between the actual output and the desired output . The feedback control relationship is expressed as:

[0161] ;

[0162] where is the feedback compensation value, are the proportional, integral, and differential gains respectively. The feedback compensator generates the feedback compensation parameter matrix to correct the system deviation caused by non-linear disturbances and modeling errors. After obtaining the feedforward compensation parameter matrix and the feedback compensation parameter matrix, compensation control calculations are performed, and the two are combined to generate the final compensation control parameter matrix , and its formula is:

[0163] ;

[0164] This matrix includes both the active adjustment of feedforward compensation for disturbances and the real-time correction of feedback compensation for deviations, and can ensure the stability and accuracy of the system under complex working conditions. Parameter optimization is performed on the compensation control parameter matrix, and reasonable ranges for the rotational speed of the crusher, the vibration frequency of the screen, and the separation ratio of the sorting device are set to construct constraint conditions. Assume the rotational speed range of the crusher is , the vibration frequency range of the screen is , and the separation ratio range is , the constraint conditions are expressed as:

[0165] ;

[0166] Combining these constraint conditions, the parameter optimization interval is determined. Based on the compensation control parameter matrix and the parameter optimization interval, an optimization algorithm is used to find the optimal process parameter combination. The objective function of the optimization problem is expressed as:

[0167] ;

[0168] where is the resource conversion rate, is the energy consumption, and are the weight factors. Through a constraint optimization algorithm (such as a genetic algorithm or a particle swarm optimization algorithm), the solution space is searched within the parameter optimization interval, and finally the optimal process parameter combination that satisfies the constraint conditions is obtained.

[0169] Referring to Figure 2 , this embodiment provides a real-time monitoring and optimization system for construction waste resource utilization, including:

[0170] The acquisition module 1 is used to perform construction waste resource treatment based on the current process parameter combination, and collect vibration frequency, material image, and weight data of crushers, screeners, and sorting devices through a multi-source sensing network to obtain an original parameter data set;

[0171] The feature extraction module 2 is used to input the original parameter data set into a multi-dimensional feature extraction model for feature extraction and feature standardization processing to obtain a standardized feature data set;

[0172] The establishment module 3 is used to establish a crushing dynamics model, a screening efficiency model, and a sorting process model according to the standardized feature data set, couple them through a material balance equation and an energy balance equation, and use the least squares method and a genetic algorithm for model parameter identification to obtain a mathematical model of the process;

[0173] The analysis module 4 is used to perform autoregressive analysis based on the mathematical model of the process, and input the current process parameter combination and historical quality data into an ε-constraint optimizer to obtain the Pareto optimal solutions of the resource conversion rate and the energy consumption;

[0174] The generation module 5 is used to generate a control strategy based on the Pareto optimal solutions, calculate the control amount of the actuator through a model predictive controller, and perform process parameter adjustment by combining feedforward compensation and feedback compensation to generate the optimal process parameter combination.

[0175] In this embodiment, for the specific implementation of each unit in the above system embodiment, please refer to the description in the above method embodiment, and details are not repeated here.

[0176] Refer to Figure 3 , in an embodiment of the present invention, a computer device is further provided. The computer device may be a server, and its internal structure may be as Figure 3 shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface, and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0177] Those skilled in the art can understand that Figure 3 the structure shown in

[0178] is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0179] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium provided by the present invention and used in the embodiments can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0180] It should be noted that in this article, the term "including", "comprising", or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, apparatus, article, or method including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, apparatus, article, or method including that element.

[0181] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. A real-time monitoring and optimization method for resource utilization of construction waste, characterized in that: The following steps are involved: Based on the current process parameter combination, construction waste is processed for resource use, and the vibration frequency, material image and weight data of the crusher, screener and sorting device are collected through a multi-source sensor network to obtain the original parameter data set; Inputting the original parameter data set into a multidimensional feature extraction model to perform feature extraction and feature standardization processing to obtain a standardized feature data set; According to the standardized characteristic data set, a crushing dynamics model, a screening efficiency model and a sorting process model are established, and the material balance equation and the energy balance equation are coupled, and the least square method and genetic algorithm are used to identify the model parameters to obtain a mathematical model of the process; specifically, the following steps are performed: a time series analysis is performed on the crushing process data in the standardized characteristic data set, a crushing dynamics equation is established based on the material particle size distribution function and the crushing power function, and the Bond power calculation formula is used to calculate the crushing energy consumption per unit time to obtain a crushing dynamics model; a probability analysis is performed on the screening process data in the standardized characteristic data set, the material pass rate is calculated based on the Rosin-Rammler particle size distribution function, and a screening dynamics equation is established in combination with the material stacking thickness function to obtain a screening efficiency model; a trajectory analysis is performed on the sorting process data in the standardized characteristic data set, the material sorting trajectory is calculated based on the multi-body system motion equation, and the sorting dynamics equation is constructed through the separation function to obtain a sorting process model; according to the crushing dynamics The output parameters of the model and the input parameters of the screening efficiency model are used to establish a material balance equation, the material flow of each process is calculated, and the material balance constraint relationship is solved by an iterative method to obtain a material balance equation group; an energy balance equation is established according to the energy consumption parameters of the crushing dynamics model, the screening efficiency model and the sorting process model, the energy transfer efficiency of each process is calculated and the energy balance constraint relationship is solved by a matrix method to obtain an energy balance equation group; the material balance equation group and the energy balance equation group are used to construct an objective function, and the optimization variables including crushing ratio, screening efficiency and sorting accuracy are set, and the parameters are optimized by a genetic algorithm to obtain a model parameter optimization solution; the model parameter optimization solution is parameter-fitted by the least squares method to obtain a model parameter identification result, and the crushing dynamics model, the screening efficiency model and the sorting process model are updated based on the model parameter identification result, and the coupling relationship between the process parameters is calculated by the material balance equation group and the energy balance equation group to obtain a process mathematical model; Performing autoregressive analysis based on the mathematical model of the process, and inputting the current process parameter combination and historical quality data into an ε-constrained optimizer to obtain a Pareto optimal solution for resource conversion rate and energy consumption; A control strategy is generated based on the Pareto optimal solution, and the control amount of the actuator is calculated through a model predictive controller. The process parameters are adjusted in combination with feedforward compensation and feedback compensation to generate an optimal process parameter combination.

2. The real-time monitoring and optimization method for resource utilization of construction waste according to claim 1 is characterized in that: The construction waste resource processing is carried out based on the current process parameter combination, and the vibration frequency, material image and weight data of the crusher, the screening machine and the sorting device are collected through the multi-source sensor network to obtain the original parameter data set, including: Based on the current process parameter combination, construction waste is processed for resource use. Vibration sensors are installed at the main bearing seats of crushers, screeners, and sorting devices to collect vibration signals and obtain equipment vibration frequency data. An image sensor is installed at the material conveying channel to collect material images and obtain material shape data, and a weight sensor is installed on the conveyor belt to collect material weight and obtain material flow data; Based on the equipment vibration frequency data, the material shape data and the material flow data, a sensor calibration curve database is established, and calibration parameter calculation is performed to obtain a calibration parameter matrix; Based on the calibration parameter matrix, compensation calculation and data integration are performed on the equipment vibration frequency data, the material shape data and the material flow data to obtain an original parameter data set.

3. The real-time monitoring and optimization method for resource utilization of construction waste according to claim 2 is characterized in that: The step of inputting the original parameter data set into a multidimensional feature extraction model for feature extraction and feature standardization to obtain a standardized feature data set includes: Inputting the device vibration frequency data in the original parameter data set into the first graph convolution layer in the multidimensional feature extraction model, the first graph convolution layer comprising 8 5×5 convolution kernels, to obtain a vibration feature mapping matrix; Inputting the material morphology data in the original parameter data set into the second graph convolution layer in the multidimensional feature extraction model, the second graph convolution layer comprising 16 3×3 convolution kernels, to obtain an image feature mapping matrix; Inputting the material flow data in the original parameter data set into the third graph convolution layer in the multidimensional feature extraction model, the third graph convolution layer comprising 32 7×7 convolution kernels, to obtain a weight feature mapping matrix; Input the vibration feature mapping matrix, the image feature mapping matrix and the weight feature mapping matrix into the feature fusion layer in the multi-dimensional feature extraction model, calculate the feature weights through a multi-head attention mechanism, and obtain a multimodal feature fusion matrix; Input the multimodal feature fusion matrix into the three fully connected layers in the multidimensional feature extraction model, wherein the number of neurons in the first fully connected layer is 512, the number of neurons in the second fully connected layer is 256, and the number of neurons in the third fully connected layer is 128, and each layer adopts the LeakyReLU activation function to obtain the target feature vector; The target feature vector is standardized and decomposed by multi-scale wavelet, and a standardized feature data set is obtained by reconstructing the target feature vector.

4. The real-time monitoring and optimization method for resource utilization of construction waste according to claim 1 is characterized in that: The method of performing autoregressive analysis based on the mathematical model of the process and inputting the current process parameter combination and historical quality data into the ε-constrained optimizer to obtain the Pareto optimal solution of resource conversion rate and energy consumption includes: Based on the mathematical model of the process, the resource conversion rate objective function and the energy consumption objective function are calculated, and weighted combination is performed to obtain a multi-objective optimization function; Based on the current process parameter combination, a p-order autoregressive model is constructed, and the optimal order is determined by the AIC criterion to obtain the autoregressive model structure; Inputting historical quality data into the autoregressive model structure, calculating the autoregressive coefficient and the white noise variance, obtaining the autoregressive prediction model, and performing rolling prediction on the process parameters based on the autoregressive prediction model to obtain a process parameter prediction sequence; Input the process parameter prediction sequence into the ε-constraint optimizer, set the resource conversion rate constraint interval and the energy consumption constraint interval, perform grid search, and obtain the constraint condition matrix; The constraint matrix is ​​solved by piecewise optimization to obtain a piecewise optimal solution set, and the piecewise optimal solution set is sorted by non-dominated sorting, and solutions with a larger congestion degree are selected as elite individuals to obtain a non-dominated solution set; The non-dominated solution set is locally searched and optimized, and a simulated annealing strategy is used to find the optimal solution in the neighborhood of the solution to obtain the Pareto optimal solution of resource conversion rate and energy consumption.

5. The real-time monitoring and optimization method for resource utilization of construction waste according to claim 4 is characterized in that: The process parameter prediction sequence is input into the ε-constraint optimizer, the resource conversion rate constraint interval and the energy consumption constraint interval are set, and a grid search is performed to obtain a constraint condition matrix, including: Inputting the process parameter prediction sequence into the ε-constrained optimizer for normality test to obtain parameter distribution characteristics; Based on the parameter distribution characteristics, the process parameter prediction sequence is normalized to obtain a standardized prediction sequence, and constraint boundary calculation is performed based on the standardized prediction sequence to obtain a resource conversion rate constraint interval, and at the same time, energy consumption analysis is performed on the standardized prediction sequence to obtain an energy consumption constraint interval; Generate grid points based on the resource conversion rate constraint interval and the energy consumption constraint interval to obtain a discrete constraint point set, and input the discrete constraint point set into an ε vector generator for equal-spaced sampling to obtain an ε constraint vector; The ε constraint vector and the discrete constraint point set are combined to obtain a constraint condition matrix.

6. The real-time monitoring and optimization method for resource utilization of construction waste according to claim 5 is characterized in that: The control strategy is generated based on the Pareto optimal solution, and the control amount of the actuator is calculated by the model predictive controller, and the process parameters are adjusted in combination with feedforward compensation and feedback compensation to generate the optimal process parameter combination, including: The Pareto optimal solution is input into a model predictive controller, a quadratic objective function is used to construct a control performance index, and a step response analysis is performed to obtain a control increment sequence; Based on the control increment sequence, a feedforward compensation model is constructed, and feedforward compensation calculations are performed on the rotation speed, vibration frequency and sorting ratio parameters of the crusher, the screening machine and the sorting device to obtain a feedforward compensation parameter matrix; Inputting the feedforward compensation parameter matrix into the feedback compensator, calculating the feedback compensation amount of each actuator, and obtaining the feedback compensation parameter matrix; Perform compensation control calculation based on the feedforward compensation parameter matrix and the feedback compensation parameter matrix to obtain a compensation control parameter matrix; Optimizing the compensation control parameter matrix, constructing constraint conditions by setting the crusher speed range, the screening machine vibration frequency range and the sorting ratio range of the sorting device, and obtaining the parameter optimization interval; An optimization solution is performed based on the compensation control parameter matrix and the parameter optimization interval to obtain an optimal process parameter combination.

7. A real-time monitoring and optimization system for resource utilization of construction waste, characterized in that: The steps for implementing the real-time monitoring and optimization method for resource utilization of construction waste as described in any one of claims 1 to 6, the real-time monitoring and optimization system for resource utilization of construction waste comprises: The acquisition module is used to process the construction waste resources based on the current process parameter combination, and collect the vibration frequency, material image and weight data of the crusher, screener and sorting device through a multi-source sensor network to obtain the original parameter data set; A feature extraction module is used to input the original parameter data set into a multidimensional feature extraction model to perform feature extraction and feature standardization processing to obtain a standardized feature data set; Establishing a module for establishing a crushing dynamics model, a screening efficiency model and a sorting process model according to the standardized characteristic data set, coupling the material balance equation and the energy balance equation, using the least square method and genetic algorithm to identify the model parameters, and obtaining a process mathematical model; An analysis module, for performing autoregressive analysis based on the mathematical model of the process, and inputting the current process parameter combination and historical quality data into an ε-constrained optimizer to obtain a Pareto optimal solution of resource conversion rate and energy consumption; A generation module is used to generate a control strategy based on the Pareto optimal solution, calculate the control amount of the actuator through a model predictive controller, adjust the process parameters in combination with feedforward compensation and feedback compensation, and generate an optimal process parameter combination.

8. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the real-time monitoring and optimization method for resource utilization of construction waste described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the real-time monitoring and optimization method for resource utilization of construction waste described in any one of claims 1 to 6 are implemented.

Images