Crushing particle size processing management method and system for particle material in uniform state

Abstract

The invention discloses a crushing granularity processing management method and system under a uniform state of a particle material, and belongs to the field of intelligent processing, and the method comprises the steps: building a drying portrait of the particle material through the recognition of the humidity, hardness and other characteristics of the particle material; respectively constructing processing evaluation feature libraries of mechanical crushing and airflow crushing, and forming an optimal crushing feature path based on indexes such as energy consumption and time consumption; searching in the characteristic paths by taking the drying portrait of the material as a starting point and the target granularity as an end point, and determining a minimum crushing path; and performing optimization in combination with the equipment constraint and the processing target, and generating a crushing granularity processing management strategy. Accurate control and dynamic adjustment in the crushing machining process are achieved, the particle size control problem of particle material crushing under the uniform drying condition is effectively solved, and the technical effects of improving the product quality, reducing the machining energy consumption, shortening the machining time and improving the equipment utilization rate are achieved.

Description

Technical Field

[0001] This invention relates to the field of intelligent processing, specifically to a method and system for managing the crushing particle size of granular materials in a uniform state. Background Technology

[0002] In the building materials, chemical, and food processing industries, particulate material crushing is a key pretreatment process, and the industry's requirements for particle size uniformity, processing efficiency, and energy consumption control after crushing are constantly increasing. In particular, the state of high-moisture or multi-form particulate materials directly affects the crushing difficulty and particle size stability.

[0003] Existing technologies such as CN120169490A (a crusher and crushing process for large pieces of calcined sodium sulfide), CN119819417A (a crusher and crushing process for large pieces of calcined sodium sulfide), and CN120801736A (a fully automatic crushing and detection method and system for grain granules) are only applicable to specific items, lacking versatility and outdated system control. They rely on mechanical crushing without considering the synergistic selection of mechanical and airflow crushing; they ignore the influence of material conditions on crushing, failing to establish a drying-crushing linkage; and they lack multi-dimensional quantitative evaluation and optimal path planning, limiting automation to localized processes. Although mechanical crushing and airflow crushing each have their advantages and disadvantages, a systematic data-driven management method is still lacking to intelligently select or combine different crushing methods based on the real-time state of the material to achieve advanced control and plan the optimal processing path. This results in a need for further improvement in processing efficiency and resource utilization. Summary of the Invention

[0004] This invention provides a method and system for managing the crushing particle size of granular materials in a uniform state, aiming to solve the technical problems of low processing efficiency and low resource utilization caused by the reliance on experience and lack of systematic planning in the crushing of granular materials in the prior art.

[0005] In view of the above problems, the present invention provides a method and system for managing the crushing particle size of granular materials in a uniform state.

[0006] The first aspect of this invention discloses a method for managing the crushing particle size of particulate materials under uniform conditions. This method includes: identifying the state of the particulate material and establishing a drying profile of the particulate material; acquiring a processing evaluation feature library of mechanical crushing space and airflow crushing space, sorting the crushed particle size based on the processing evaluation, and constructing a crushing feature path; using the drying profile of the particulate material as the starting point and the target crushed particle size as the ending point, performing a minimum path search based on the crushing feature path to determine the minimum crushing path; and performing a response relationship optimization search based on the minimum crushing path, according to equipment constraints and processing targets, to obtain a crushing particle size management strategy for crushing processing management and control.

[0007] Another aspect of this invention discloses a crushing particle size processing management system for particulate materials in a uniform state. This system includes: a profile creation module for identifying the state of the particulate material and creating a dry profile of the particulate material; a feature path construction module for acquiring a processing evaluation feature library of mechanical crushing space and airflow crushing space, sorting the crushed particle size based on the processing evaluation, and constructing a crushing feature path; a minimum path determination module for searching for the minimum crushing path based on the dry profile of the particulate material, with the target crushed particle size as the endpoint, and determining the minimum crushing path; and a management strategy acquisition module for optimizing the response relationship based on the minimum crushing path, according to equipment constraints and processing objectives, to obtain a crushing particle size processing management strategy for crushing processing management and control.

[0008] One or more technical solutions provided in this invention have at least the following technical effects or advantages: By employing a technical solution that identifies the state of particulate materials and establishes a drying profile, acquires a processing evaluation feature library of mechanical crushing space and airflow crushing space, constructs a crushing feature path, and performs a minimum path search with the drying profile as the starting point and the target crushing particle size as the ending point, and combines equipment constraints and processing target optimization to obtain a crushing particle size processing management strategy, this solution solves the technical problems of low processing efficiency and low resource utilization caused by the reliance on experience and lack of systematic planning in the crushing of particulate materials in existing technologies. It achieves the technical effect of forming a complete closed loop from material perception to processing control, thereby improving crushing processing efficiency and resource utilization.

[0009] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0010] Figure 1 A flowchart illustrating the particle size management method for crushing particulate materials in a uniform state is provided for embodiments of the present invention. Figure 2 A schematic diagram of the structure of a crushing particle size processing management system for granular materials in a uniform state is provided for embodiments of the present invention.

[0011] Figure labeling: Image creation module 11, feature path construction module 12, minimum path determination module 13, management strategy acquisition module 14. Detailed Implementation

[0012] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structure, features and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0013] The overall concept of the technical solution provided by this invention is as follows: This invention provides a method and system for managing the crushing particle size of granular materials in a uniform state. The method involves identifying the state of the granular material and establishing a drying profile, constructing a processing evaluation feature library and crushing feature paths for mechanical and pneumatic crushing spaces, searching for the minimum path starting from the drying profile and ending at the target particle size, and combining equipment constraints and processing target optimization to form a crushing particle size management strategy.

[0014] After introducing the basic principles of the present invention, various non-limiting embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0015] Example 1, as Figure 1 As shown in the figure, this invention provides a method for managing the particle size of crushed particulate materials in a uniform state. The method includes: Step S100: Perform state identification on the particulate material and establish a state profile of the particulate material.

[0016] In this embodiment of the application, the state profile refers to a structured and reusable data set formed after processing based on the drying state parameters of particulate materials, which contains core characteristic information of the materials.

[0017] Specifically, a combination of sensor and image recognition technologies is used to collect basic information about particulate materials, including humidity, hardness, particle morphology, density, and surface characteristics. The collected raw data is then cleaned and standardized. The standardized data is then integrated according to a preset structure to form a unique drying profile for each particulate material.

[0018] This step avoids the problem of ambiguous judgment of the state of traditional particulate materials through precise data collection and structured processing. It improves data accuracy with the help of sensors and image recognition. The drying profile provides accurate input for subsequent crushing path matching, reducing issues such as unqualified particle size and energy waste from the source.

[0019] Step S200: Obtain the processing evaluation feature library of mechanical crushing space and airflow crushing space, sort them according to the crushing particle size based on the processing evaluation, and construct the crushing feature path.

[0020] Specifically, the mechanical crushing space refers to the parameter range involved in processing systems that use mechanical forces, such as extrusion and shearing, for crushing, including equipment type and processing conditions. The airflow crushing space refers to the parameter range of processing systems that utilize high-speed airflow impact for crushing, encompassing equipment parameters such as airflow velocity, pressure, and operating conditions. The crushing characteristic path refers to the path formed by sequentially listing optimal processing conditions in ascending order of particle size, reflecting an efficient processing sequence from initial particle size to target particle size.

[0021] Specifically, for mechanical crushing and airflow crushing, samples with different processing conditions and material characteristics are collected to obtain processing time-series data. A particle size time-series chain is established according to the particle size from smallest to largest. Based on the time-series chain, processing time, energy consumption, and particle size change ratio are calculated. These evaluation indicators are associated with corresponding particle size nodes to construct a processing evaluation feature library for mechanical and airflow crushing spaces. Based on the processing evaluation results of each particle size in the feature library, the optimal processing conditions are selected through a weighted algorithm. The optimal conditions of each node are then connected in sequence according to the particle size to form the crushing feature paths for mechanical crushing and airflow crushing spaces.

[0022] This step involves systematically constructing an evaluation feature library and granular ordered paths for dual crushing spaces, providing a quantitative basis for subsequent optimal processing path selection, and reducing inefficiency and energy waste caused by blindly selecting processing conditions.

[0023] Step S300: Starting from the state profile of the particulate material and ending at the target particle size, perform a minimum path search based on the crushing characteristic path to determine the minimum crushing path.

[0024] In this embodiment, the minimum crushing path refers to the optimal processing path obtained through minimum path search, from the initial particle size corresponding to the current state of the material to the target crushed particle size. It is divided into two categories: mechanical crushing minimum path and airflow crushing minimum path. The starting point refers to the particle size time sequence node corresponding to the current state of the material when the particulate material drying profile is matched to the crushing feature path. The ending point refers to the target particle size time sequence node corresponding to the target crushed particle size when matched to the crushing feature path.

[0025] Specifically, the core characteristics of the particulate material drying profile are extracted to find the corresponding starting particle size node. The preset target crushing particle size is matched with the particle size time sequence nodes in the two paths to determine the target endpoint. A path search algorithm is used, with the comprehensive evaluation quantification result of each particle size node in each path as the path weight. All possible processing sequences are traversed from the starting point to the endpoint to select the path with the lowest weight. The corresponding optimal paths, namely the minimum mechanical crushing path and the minimum airflow crushing path, are obtained in the two crushing characteristic paths, respectively, and stored in the path database for subsequent optimization.

[0026] This step avoids the problem of relying on experience for path selection in traditional crushing by using precise node matching and algorithmic path search. At the same time, it obtains two types of minimum paths, namely mechanical and airflow, providing two options for subsequent combined optimization of equipment constraints and processing objectives, thus improving the scientific nature and flexibility of crushing path selection.

[0027] Step S400: Based on the minimum crushing path, perform response relationship optimization search according to equipment constraints and processing objectives to obtain crushing particle size processing management strategy, and feed it back to the Internet of Things control system for crushing processing management and control.

[0028] In the embodiments of this application, equipment constraints refer to the operating limitations of the crushing equipment itself, such as the maximum speed and minimum crushing gap of a mechanical crusher, and the maximum air pressure and maximum feed rate of an airflow crusher, which determine the feasible range of processing parameters.

[0029] Specifically, real-time constraint data of the equipment is collected, and the processing target is retrieved. With the equipment constraints as boundary conditions and the processing target as the optimization direction, the response relationship optimization search is performed based on the parameters of the two minimum crushing paths of mechanical and airflow. The parameter combination that meets both the equipment constraints and the processing target is selected, and the optimal parameter combination is organized into a structured crushing particle size processing management strategy.

[0030] This step, through dual control of equipment constraints and processing targets, generates management strategies that can directly guide production, avoiding the deviations of traditional experience-based control and achieving precise and efficient management of crushing and processing.

[0031] Furthermore, the process of identifying the dry state of particulate materials and establishing a dry profile of the particulate materials includes: collecting basic information of the particulate materials through sensors and image recognition technology, including humidity, hardness, particle shape, density, and surface characteristics; cleaning and standardizing the collected basic information; establishing a state profile of the particulate materials based on the processed data, and storing it in a database as a material characteristic file.

[0032] In this embodiment, a humidity sensor is first used to collect material humidity, and a pressure sensor is used to obtain hardness. Hardness-related parameters are obtained indirectly by applying controllable pressure to the particle sample using a pressure sensor. Pressure is stopped when the particle surface cracks during the pressure application process, indicating the critical pressure the particle can withstand. The peak pressure at this point represents the crushing strength, used to characterize hardness. Material images are captured using an industrial camera, and particle morphology, such as particle outline, size distribution, and surface characteristics like agglomeration and smoothness, is extracted using image recognition technology. Density is calculated using mass and volume measurement tools to obtain basic information about the particle material. The collected information undergoes data cleaning, such as removing abnormal humidity values ​​falsely reported by the sensor. Then, standardization processing is used to convert humidity percentages and hardness values ​​into data of a unified dimension. Specifically, box plots are used to identify and remove outliers, such as negative or out-of-range values ​​falsely reported by the humidity sensor. Finally, dimensional unification processing is performed using a min-max normalization method. The method linearly maps physical values ​​such as humidity and hardness, which have different meanings, to the interval [0, 1]. The calculation formula is: Standardized value = (Original value - Minimum value) / (Maximum value - Minimum value), where the minimum and maximum values ​​are determined based on historical data or preset parameters. Finally, through weighted fusion, that is, by assigning humidity weight and hardness weight according to material characteristics, a feature vector of a unified dimension is generated and stored in the database as a standardized input for state profile, ensuring the comparability and fusion of multi-source data, and providing a consistent data foundation for subsequent crushing path optimization. Among them, the box plot method is a statistical outlier detection method based on quartiles. By calculating the first quartile (Q1), third quartile (Q3), and interquartile range (IQR) of the data, the outlier boundaries are set as Q1-1.5IQR and Q3+1.5IQR. In the data standardization process, this method is used to automatically identify outliers in sensor data such as humidity and hardness and remove them to ensure the accuracy and reliability of the material condition profile. The integrated and processed data forms a drying profile, which is stored in a relational or non-relational database to generate a material characteristic file that can be accessed at any time.

[0033] Specifically, during image acquisition, an industrial camera was used with a ring LED fill light with a color temperature of 5000K-6500K and an IP67 dustproof lens. Short shutter speeds and synchronous trigger shooting (1 / 1000s-1 / 2000s) were employed to avoid motion blur, with ≥10 frames acquired per batch. The acquired images were then balanced using CLAHE adaptive histogram equalization to mitigate the impact of ambient light and dust atomization on image quality. NLM nonlocal mean filtering was used to remove residual noise, and the Otsu adaptive threshold segmentation algorithm was employed for initial separation of particles from the background. Finally, the Canny edge detection algorithm, combined with morphological opening and closing operations, was used to repair edge breaks and accurately extract particle contours. Morphological opening and closing operations are image processing operations based on structural elements, such as circles and rectangles. The opening operation first eliminates small noises such as dust and thins out interference through erosion, and then restores the main shape of the particle through dilation. The closing operation first dilates to fill small holes in the contour and connect broken edges, and then erodes to restore the true size of the particle. The two operations are used together for particle image preprocessing to ensure the integrity and accuracy of contour extraction. Then, contour moments are calculated to obtain the equivalent diameter, area and other morphological parameters of each particle. Specifically, the pixel coordinates (x, y) of the preprocessed particle contour are obtained first, and then the zeroth moment, first moment, second moment and so on are calculated according to the mathematical definition of moments. The quantitative morphological parameters of the particles are derived through moment values, providing accurate data support for subsequent size distribution statistics and cluster analysis. Furthermore, the parameters are classified using the K-Means clustering algorithm, and the proportions of fine, medium, and coarse particles are statistically analyzed to form particle size distribution data. At the same time, a lightweight MobileNetV2 deep learning model is introduced to optimize and correct the contours of irregular or deformed particles. Finally, multiple clear frames are stitched together using inter-frame fusion technology, and the majority voting method (≥7 frames of consistency) is used to determine the valid verification results, ensuring the accuracy and stability of particle morphology extraction.

[0034] Among them, the lightweight MobileNetV2 deep learning model is based on the optimized design of the MobileNetV2 core architecture. The input is an RGB preprocessed granular image, the size of which is adjusted to 224×224 pixels, and the total number of parameters is controlled within 3 million. The core structure includes: 1. An input layer and preprocessing module, which normalizes the image, maps pixel values ​​to [-1,1] and aligns them with channels; 2. A feature extraction backbone, consisting of 6 concatenated inverse residual blocks. Each block uses a structure of 1×1 convolution for dimensionality increase → 3×3 depthwise separable convolution for feature extraction → 1×1 convolution for dimensionality reduction. The expansion factor is set to 6, and gradient flow is preserved through shortcut connections, focusing on extracting key features such as particle contours and edge textures; 3. A linear bottleneck layer, which removes the traditional ReLU activation function to avoid loss of feature information and strengthens the linear expression of particle morphology features; 4. A global average pooling layer and a classification output layer, which compresses the feature map into a 128-dimensional vector and outputs three types of results through a fully connected layer: normal particle contours, deformed particle contours, and contours requiring correction, achieving accurate identification and optimized correction of particle morphology.

[0035] The model employs a construction and training approach combining transfer learning and fine-tuning with a custom dataset. It collects particle images from various fields, including building materials and food, encompassing different humidity levels, agglomeration, and dust scenarios. These images are labeled with contour morphology and deformity types, and the training, validation, and test sets are divided in a 7:2:1 ratio. By randomly rotating, horizontally flipping, adding Gaussian noise, and adjusting brightness and contrast, it simulates industrial lighting changes and material motion blur, enhancing the model's robustness. Lightweight weights pre-trained on the ImageNet dataset are loaded, and the first three inverse residual blocks are frozen, training only the last three feature extraction blocks and the output layer. The Adam optimizer is used, with an initial learning rate of 1e-4, a cross-entropy loss function, and a batch size of 32. For the first 20 rounds, all layers are unfrozen for fine-tuning, and the step size is adjusted using a cosine annealing strategy. An early stopping mechanism is implemented—stopping the model if the validation set loss does not decrease after three rounds—prevents overfitting. The final model achieves a particle morphology recognition accuracy of ≥92% on the test set, meeting the requirements of real-time industrial processing.

[0036] This step, through precise data collection and standardized processing, makes material characteristic data more reliable, providing a precise basis for subsequent crushing path matching and reducing situations such as unqualified particle size and wasted processing energy caused by inaccurate material information.

[0037] Furthermore, the processing evaluation feature library for mechanical crushing space and airflow crushing space is obtained, including: collecting samples for mechanical crushing and airflow crushing methods respectively, including different processing conditions and different particle material characteristics; evaluating the processing time sequence particle size for mechanical crushing and airflow crushing methods respectively based on the collected sample sets, and establishing a particle size time sequence chain in order of particle size from smallest to largest; and evaluating processing time, processing energy consumption, and processing particle size change ratio based on the particle size time sequence chain, and establishing the processing evaluation feature library.

[0038] In the embodiments of this application, particulate material characteristics refer to the physical properties of the particles themselves, such as moisture content, hardness, density, and initial particle size. The particle size time sequence chain refers to a sequence formed by linking particle size data from different processing stages in descending (or ascending) order of particle size, reflecting the gradual change in particle size. The particle size change ratio during processing and crushing refers to the ratio of the difference in particle size before and after a certain crushing stage to the initial particle size, reflecting the crushing efficiency.

[0039] Specifically, samples were collected for mechanical crushing and airflow crushing respectively. Mechanical crushing required coverage of different processing conditions such as rotation speed and crushing gap, as well as different material properties such as hardness and moisture content. Airflow crushing required coverage of different processing conditions such as airflow velocity and feed rate, as well as corresponding material properties. During the collection process, a data acquisition instrument was used to record processing parameters, and a laser particle size analyzer was used to monitor particle size in real time. Based on the collected sample sets, the processing time-series particle size was evaluated for both crushing methods. Particle size data at different time points were extracted and connected in descending order of particle size to establish a particle size time-series chain. Based on the particle size time-series chain, the processing time, processing energy consumption, and processing particle size change ratio were evaluated to establish a processing evaluation feature library.

[0040] This step, through systematic data collection and analysis, established a quantitative evaluation system for the two crushing methods, providing a precise evaluation basis for subsequent crushing path optimization and reducing the blindness in the selection of processing conditions.

[0041] Furthermore, based on the particle size time sequence chain, processing time, processing energy consumption, and processing particle size change ratio are evaluated, and the processing evaluation feature library is established, including: aligning particle size time sequence chains with different processing conditions and different particulate material characteristics based on particle size; evaluating processing time, processing energy consumption, and processing particle size change ratio according to processing conditions and particulate material characteristics based on the aligned particle size time sequence chain, and establishing an evaluation feature list; inserting the evaluation feature list into the corresponding particle size time sequence node to construct the processing evaluation feature library.

[0042] In the embodiments of this application, granularity time sequence chain alignment refers to the operation of adjusting the node positions of granularity time sequence chains under different processing conditions and different material characteristics according to the same granularity size, so as to ensure that the evaluation data corresponding to different scenarios of the same granularity can be compared.

[0043] Specifically, using data processing software such as Excel and Python Pandas, the particle size time series chains corresponding to different processing conditions and material characteristics are aligned based on particle size. For example, the 30mm particle size nodes in all time series chains are adjusted to the same position to ensure data dimension uniformity. Based on the aligned time series chains, they are grouped according to processing conditions and material characteristics. A timer is used to count the processing time from one particle size node to the next particle size node in each group, and an energy meter is used to record the processing energy consumption at the corresponding stage. The particle size change ratio of the processed crushing is calculated using the formula (previous particle size - subsequent particle size) / previous particle size. These three types of indicators are organized into an evaluation feature list by group. Using a database management tool such as MySQL, the evaluation feature list of each group is inserted into the dedicated node of the particle size time series chain, so that each node simultaneously contains particle size, processing conditions, material characteristics, and the three types of evaluation indicators, ultimately constructing a complete processing evaluation feature library.

[0044] This step addresses the issue of data incomparability across different scenarios through temporal chain alignment. Combined with structured evaluation and node binding, it makes the processing evaluation feature library more systematic and practical, providing accurate and comparable quantitative data support for subsequent selection of optimal processing conditions and construction of crushing feature paths.

[0045] Furthermore, based on the particle size of the processed material, a pulverization characteristic path is constructed by sorting the pulverized particles. This includes: configuring screening weights for processing time, processing energy consumption, and the ratio of changes in pulverized particle size; using these screening weights to fuse the evaluation features and determine the comprehensive evaluation quantification result; for different particulate material characteristics, the processing conditions are screened based on the maximum comprehensive evaluation quantification result to obtain the screening processing condition paths corresponding to different particle size time sequence nodes; and sequentially connecting the screening processing condition paths corresponding to the particle size time sequence nodes to obtain the pulverization characteristic path.

[0046] In this embodiment, the screening weight refers to the weighting coefficient assigned to the three evaluation indicators—processing time, processing energy consumption, and the ratio of particle size change—based on actual processing objectives, such as prioritizing efficiency or energy saving. This weighting is used to quantify the impact of each indicator on the processing effect. The screening processing condition path refers to selecting the processing condition with the highest comprehensive evaluation quantification result at each particle size time-series node, forming a single-node optimal processing condition set for a specific particulate material characteristic.

[0047] Specifically, the allocation of the aforementioned weighting coefficients is adjusted according to production needs. That is, the weighting coefficient allocation is determined based on production requirements, assigning specific weighting coefficients to the three evaluation indicators: processing time, processing energy consumption, and the ratio of particle size change in processing. For example, if low energy consumption is prioritized, a higher weight is assigned to processing energy consumption, such as 0.4; if efficiency is prioritized, processing time is assigned 0.4. The screening weights for processing time, processing energy consumption, and the ratio of particle size change in processing are configured through data processing software to ensure that the sum of the weights is 1. A weighted summation formula is then applied, such as: Overall Score = Time × Time Weight + Energy × Energy Weight. The evaluation index of different processing conditions at each particle size time node is integrated with the corresponding weight by adding the particle size change ratio × the change ratio weight to calculate the comprehensive evaluation quantitative result of each processing condition. For each particulate material characteristic, the processing condition with the largest comprehensive evaluation quantitative result is selected at each particle size time node to form the screening processing condition path corresponding to the material characteristic. Through path integration tools, such as Excel spreadsheets or database queries, the optimal processing conditions of each particle size node under the same material characteristic are connected in sequence according to the particle size gradient to finally obtain the crushing characteristic path of the material.

[0048] This step achieves precise matching of processing targets through weight configuration, and selects the optimal processing conditions by combining comprehensive evaluation. The final crushing characteristic path provides a targeted optimal condition sequence for subsequent path search, helping to improve the stability and efficiency of crushing processing.

[0049] Furthermore, the crushing characteristic path includes: a crushing characteristic path in a mechanical crushing space and a crushing characteristic path in an airflow crushing space.

[0050] In this embodiment, the crushing characteristic path of the mechanical crushing space is a processing sequence formed by concatenating the optimal processing conditions of each particle size time sequence node of mechanical crushing, based on the processing evaluation feature library of the mechanical crushing space. The crushing characteristic path of the airflow crushing space is also a processing sequence formed by concatenating the optimal processing conditions of each particle size time sequence node of airflow crushing, based on the processing evaluation feature library of the airflow crushing space.

[0051] This step, by clearly classifying the crushing characteristic paths of the two crushing spaces, breaks through the limitations of the traditional single crushing method path, provides a clear data carrier for subsequent searching of the optimal crushing scheme from both mechanical and airflow paths, improves the flexibility of crushing and processing path selection, and lays the foundation for subsequent combined optimization of equipment constraints and processing objectives.

[0052] Furthermore, taking the dried profile of the particulate material as the starting point and the target particle size as the ending point, a minimum path search is performed based on the crushing characteristic path to determine the minimum crushing path. This includes: matching the dried profile of the particulate material with the characteristics of the particulate material in the crushing characteristic path of the mechanical crushing space and the crushing characteristic path of the airflow crushing space to locate the starting point; matching the ending particle size in the crushing characteristic path with the target particle size to locate the target ending point; and performing a minimum path search based on the starting point and the target ending point in the crushing characteristic path of the mechanical crushing space and the crushing characteristic path of the airflow crushing space to obtain the minimum crushing path, which includes the minimum mechanical crushing path and the minimum airflow crushing path.

[0053] In this embodiment, the core characteristics of the particulate material drying profile are extracted and matched with the material characteristic-particle size node data stored in the crushing feature paths of the mechanical crushing space and the airflow crushing space. That is, the profile characteristics are compared one by one with the characteristics of the two types of paths, and the entry with the highest characteristic similarity is found. The corresponding path particle size node is the starting point after matching. For example, in the mechanical path, a humidity of 13%–15% and a hardness of 45N–55N correspond to a particle size node of 8mm, locating the starting point of both paths. Simultaneously, the target particle size is compared with the particle size values ​​of each node, and the node that is consistent or closest is found; this node is the located target endpoint. An optimal path algorithm, such as the A* algorithm, is used to quantify the comprehensive evaluation results between each particle size node in each path, such as time consumption and energy consumption weighted as the path cost. All possible processing sequences are traversed from the starting point to the target endpoint, and the path with the lowest cost is selected from the crushing feature paths of mechanical and airflow crushing, ultimately obtaining the minimum mechanical crushing path and the minimum airflow crushing path.

[0054] This step, by simultaneously obtaining the optimal paths for both mechanical and airflow types, provides two options for subsequent combined optimization that integrates equipment constraints and processing objectives, thereby enhancing the scientific rigor and flexibility of path selection.

[0055] Furthermore, based on the minimum crushing path, a response relationship optimization search is performed according to equipment constraints and processing targets to obtain a crushing particle size processing management strategy. This includes: acquiring processing constraint information for mechanical crushing equipment and airflow crushing equipment, as well as acquiring processing target time, processing target cost, and target equipment load; using the equipment constraints as constraints, constructing an optimization function based on the processing target time, processing target cost, and target equipment load; performing dual-crushing combination optimization based on the minimum mechanical crushing path and the minimum airflow crushing path to obtain a combined optimization strategy; and adjusting the switching relationship of the combined optimization strategy based on the switching frequency limit of the dual-crushing combination to obtain the crushing particle size processing management strategy.

[0056] In this embodiment, real-time constraint data is collected using equipment sensors, while the processing target is retrieved. Using equipment constraints as boundary conditions and meeting the processing target as the direction, an optimization function is constructed using MATLAB, such as minimizing the single-batch crushing cost while ensuring that the time consumed does not exceed 2 hours and the equipment load does not exceed 80%. A genetic algorithm is used to optimize the response relationship based on the rotational speed sequence of the minimum path for mechanical crushing and the air pressure sequence of the minimum path for airflow crushing, selecting parameter combinations that both meet equipment constraints and achieve the processing target. Finally, the optimal parameter combinations are organized into a structured crushing particle size processing management strategy, clarifying the crushing method, specific parameters, and abnormal adjustment rules for each stage, and stored in the production control system for real-time management and control.

[0057] This step generates a management strategy through dual control of equipment constraints and processing targets, which can directly guide production, avoid the deviations of traditional experience-based control, and achieve precise and efficient management of crushing and processing.

[0058] Furthermore, the switching relationship of the combined optimization strategy is adjusted based on the switching frequency limit of the dual crushing combination to obtain the crushing particle size processing management strategy, including: determining the switching time window constraint according to the number of switching and the switching interval time; performing moving particle size crushing evaluation in the combined optimization strategy based on the switching time window constraint, obtaining the maximum evaluation difference within the time window, and determining the switching node; and dividing the combined optimization strategy into crushing mode switching segments according to the switching node to obtain the crushing particle size processing management strategy.

[0059] In this embodiment of the application, the maximum evaluation difference within the time window refers to the maximum value of the evaluation indicators (such as energy consumption difference) corresponding to different switching nodes within the same moving window, which is used to determine the necessity of switching and the optimal position.

[0060] Specifically, switching frequency limits are determined based on production experience or equipment manuals, such as a maximum of 3 switching times per day with an interval of ≥1 hour. These limits are then converted into switching time window constraints using time window analysis tools. Based on these constraints, a sliding window, covering 2 particle size nodes, is used to evaluate the moving particle size crushing process within the granularity sequence of the combined optimization strategy. The differences in energy consumption, time consumption, and other indicators among different switching nodes within the window are calculated by weighted summation, and the node with the smallest maximum evaluation difference is selected as the optimal switching node. The combined optimization strategy is then segmented according to the switching nodes to divide the crushing method, dividing the particle size sequence into continuous intervals such as mechanical crushing and airflow crushing. This forms a crushing particle size processing management strategy that includes clearly defined switching nodes and methods, and is stored in the production control system.

[0061] This step, by limiting the switching frequency and precisely locating the switching nodes, avoids equipment wear and particle size fluctuations caused by disordered switching in the combined optimization strategy, making the processing management strategy more compatible with the actual production rhythm and improving the stability of the crushing process and the service life of the equipment.

[0062] In summary, the particle size management method for crushing particulate materials in a uniform state provided by the embodiments of the present invention has the following technical effects: 1. By establishing a profile through state recognition, constructing characteristic paths in the dual crushing space, searching for the minimum path, and optimizing equipment constraints, the technical effect of precise control and dynamic adjustment of the crushing process under advanced control was achieved. This enabled intelligent optimization management of particle size of particulate materials, effectively improving product quality, reducing processing energy consumption and time, and increasing equipment utilization.

[0063] 2. Samples were collected for both mechanical and pneumatic crushing, a particle size time series chain was established, and indicators such as time consumption and energy consumption were evaluated to construct a processing evaluation feature library. This solves the problem of the lack of quantitative evaluation data in traditional crushing, forming a structured evaluation system for the two crushing methods, and providing key data support for subsequent selection of optimal processing conditions and construction of crushing paths.

[0064] 3. By combining equipment constraints and processing objectives, the optimal combination of two minimum paths is searched to form a management strategy. This solves the problem that parameters exceed equipment capabilities and objectives cannot be achieved when the minimum path for crushing is implemented, ensuring that the strategy conforms to the actual operating capacity of the equipment and meets production requirements, thus achieving precise management of crushing and processing.

[0065] Example 2, based on the same inventive concept as the method for managing the crushing particle size of granular materials in a uniform state as described in the previous examples, such as... Figure 2 As shown, this embodiment of the invention provides a crushing particle size processing management system for particulate materials in a uniform state. The system includes: a profile building module 11, used to identify the state of the particulate materials and build a state profile of the particulate materials; a feature path construction module 12, used to acquire a processing evaluation feature library of mechanical crushing space and airflow crushing space, sort the crushing particle size based on the processing evaluation, and construct a crushing feature path; a minimum path determination module 13, used to search for the minimum path based on the state profile of the particulate materials as the starting point and the target crushing particle size as the ending point, and determine the minimum crushing path; and a management strategy acquisition module 14, used to perform response relationship optimization search based on the minimum crushing path, according to equipment constraints and processing targets, to obtain a crushing particle size processing management strategy, and feed it back to the Internet of Things control system for crushing processing management and control.

[0066] Furthermore, the profile creation module 11 is also used to perform the following steps: collecting basic information of particulate materials through sensors and image recognition technology, including humidity, hardness, particle shape, density, and surface characteristics; cleaning and standardizing the collected basic information; creating a state profile of the particulate materials based on the processed data, and storing it in the database as a material characteristic file.

[0067] Furthermore, the feature path construction module 12 is also used to perform the following steps: collecting samples for mechanical crushing and airflow crushing methods respectively, including different processing conditions and different particle material characteristics; evaluating the processing time sequence particle size of mechanical crushing and airflow crushing methods according to the collected sample set, and establishing a particle size time sequence chain in order of particle size from small to large; and evaluating the processing time, processing energy consumption, and processing particle size change ratio based on the particle size time sequence chain, and establishing the processing evaluation feature library.

[0068] Furthermore, the feature path construction module 12 is also used to perform the following steps: aligning the particle size time sequence chains of different processing conditions and different particulate material characteristics based on particle size; evaluating the processing time, processing energy consumption, and processing particle size change ratio according to the aligned particle size time sequence chains, based on the processing conditions and particulate material characteristics, and establishing an evaluation feature list; inserting the evaluation feature list into the corresponding particle size time sequence nodes to construct the processing evaluation feature library.

[0069] Furthermore, the feature path construction module 12 is also used to perform the following steps: configuring screening weights for processing time, processing energy consumption, and the ratio of changes in particle size during processing; using the screening weights to fuse the evaluation features and determine the comprehensive evaluation quantification result; for different particulate material characteristics, screening processing conditions based on the maximum comprehensive evaluation quantification result to obtain screening processing condition paths corresponding to different particle size time sequence nodes; and sequentially connecting the screening processing condition paths corresponding to the particle size time sequence nodes to obtain the crushing feature path.

[0070] Furthermore, the feature path construction module 12 is also used to perform the following steps: the crushing feature path includes: the crushing feature path of the mechanical crushing space and the crushing feature path of the airflow crushing space.

[0071] Furthermore, the minimum path determination module 13 is also used to perform the following steps: matching the dry profile of the particulate material with the characteristics of the particulate material in the crushing characteristic path of the mechanical crushing space and the crushing characteristic path of the airflow crushing space, respectively, to locate the starting point; matching the endpoint particle size in the crushing characteristic path with the target crushing particle size, and locating the target endpoint; based on the starting point and the target endpoint, performing minimum path search on the crushing characteristic path of the mechanical crushing space and the crushing characteristic path of the airflow crushing space, respectively, to obtain the minimum crushing path, which includes the minimum mechanical crushing path and the minimum airflow crushing path.

[0072] Furthermore, the management strategy acquisition module 14 is also used to perform the following steps: acquiring the processing constraint information of the mechanical crushing equipment and the airflow crushing equipment, as well as acquiring the processing target time, processing target cost, and target equipment load; using the equipment constraints as constraints, constructing an optimization function based on the processing target time, processing target cost, and target equipment load, performing dual-crushing combination optimization based on the minimum path of mechanical crushing and the minimum path of airflow crushing, and obtaining a combination optimization strategy; adjusting the switching relationship of the combination optimization strategy based on the switching frequency limit of the dual-crushing combination, and obtaining the crushing particle size processing management strategy.

[0073] Furthermore, the management strategy acquisition module 14 is also used to perform the following steps: determine the switching time window constraint based on the number of switching and the switching interval time; perform moving particle size crushing evaluation in the combined optimization strategy based on the switching time window constraint, obtain the maximum evaluation difference within the time window, and determine the switching node; and perform crushing mode switching segmentation on the combined optimization strategy according to the switching node to obtain the crushing particle size processing management strategy.

[0074] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for managing the particle size of crushed granular materials in a uniform state, characterized in that, include: Step S100: Perform state identification on the particulate material and establish a state profile of the particulate material; Step S200: Obtain the processing evaluation feature library of mechanical crushing space and airflow crushing space, sort them according to the crushing particle size based on the processing evaluation, and construct the crushing feature path; Step S300: Starting from the state profile of the particulate material and ending at the target particle size, perform a minimum path search based on the crushing characteristic path to determine the minimum crushing path; Step S400: Based on the minimum crushing path, perform response relationship optimization search according to equipment constraints and processing objectives to obtain crushing particle size processing management strategy, and feed it back to the Internet of Things control system for crushing processing management and control.

2. The method for managing the crushing particle size of granular materials in a uniform state according to claim 1, characterized in that, Step S100: The step of identifying the state of particulate materials and establishing a state profile of the particulate materials includes: Basic information about particulate materials is collected through sensors and image recognition technology, including humidity, hardness, particle shape, density, and surface properties. The collected basic information is cleaned and standardized. A state profile of the particulate material is created based on the processed data and stored in the database as a material characteristic file.

3. The method for managing the crushing particle size of granular materials in a uniform state according to claim 1, characterized in that, Step S200: Obtain the processing evaluation feature library of the mechanical crushing space and the airflow crushing space, including: Samples were collected for mechanical crushing and airflow crushing methods, including different processing conditions and different particulate material characteristics. Based on the collected sample set, the processing time particle size of mechanical crushing and airflow crushing methods were evaluated, and a particle size time sequence chain was established in order of particle size from small to large. Based on the particle size time sequence chain, the processing time, processing energy consumption, and processing particle size change ratio are evaluated, and the processing evaluation feature library is established.

4. The method for managing the crushing particle size of granular materials in a uniform state according to claim 3, characterized in that, Based on the particle size time sequence chain, processing time, processing energy consumption, and processing particle size change ratio are evaluated, and the processing evaluation feature library is established, including: Align the particle size time sequence chain based on different processing conditions and different particulate material characteristics; Based on the aligned particle size time sequence chain, the processing time, processing energy consumption, and processing particle size change ratio are evaluated according to processing conditions and particle material characteristics, and an evaluation feature list is established. The evaluation feature list is inserted into the corresponding granularity time sequence node to construct the processing evaluation feature library.

5. The method for managing the crushing particle size of granular materials in a uniform state according to claim 4, characterized in that, The grinding particle size is sorted based on processing evaluation, and a grinding characteristic path is constructed, including: Configure screening weights for processing time, processing energy consumption, and the change ratio of processing and crushing particle size; use the screening weights to fuse the evaluation features and determine the comprehensive evaluation quantitative result. For different particulate material characteristics, the processing conditions are screened based on the maximum comprehensive evaluation and quantitative results, and the screening processing condition paths corresponding to different particle size time nodes are obtained. The crushing characteristic path is obtained by sequentially connecting the screening and processing condition paths corresponding to the particle size time sequence nodes.

6. The method for managing the crushing particle size of granular materials in a uniform state according to claim 5, characterized in that, The crushing characteristic path includes: the crushing characteristic path in the mechanical crushing space and the crushing characteristic path in the airflow crushing space.

7. The method for managing the crushing particle size of granular materials in a uniform state according to claim 6, characterized in that, Step S300: Starting from the dried profile of the particulate material and ending at the target particle size, perform a minimum path search based on the pulverization characteristic path to determine the minimum pulverization path, including: Based on the drying profile of the particulate material, the characteristics of the particulate material in the crushing characteristic path of the mechanical crushing space and the crushing characteristic path of the airflow crushing space are matched to locate the starting point. The target crushing particle size is matched to the endpoint particle size in the crushing characteristic path to locate the target endpoint. Based on the starting point and the target endpoint, minimum path searches are performed on the crushing characteristic paths in the mechanical crushing space and the airflow crushing space to obtain the minimum crushing path, which includes the minimum mechanical crushing path and the minimum airflow crushing path.

8. The method for managing the crushing particle size of granular materials in a uniform state according to claim 7, characterized in that, Step S400: Based on the minimum crushing path, perform a response relationship optimization search according to equipment constraints and processing objectives to obtain a crushing particle size processing management strategy, including: Obtain the processing constraint information of mechanical crushing equipment and airflow crushing equipment respectively, as well as the processing target time, processing target cost, and target equipment load; Using the equipment constraints as constraints, an optimization function is constructed based on the target processing time, target processing cost, and target equipment load. Based on the minimum path of mechanical crushing and the minimum path of airflow crushing, a dual crushing combination optimization is performed to obtain a combined optimization strategy. The switching relationship of the combination optimization strategy is adjusted based on the switching frequency limit of the dual crushing combination to obtain the crushing particle size processing management strategy.

9. The method for managing the crushing particle size of granular materials in a uniform state according to claim 8, characterized in that, The switching relationship of the combination optimization strategy is adjusted based on the switching frequency limit of the dual crushing combination to obtain the crushing particle size processing management strategy, including: Determine the switching time window constraint based on the number of switching operations and the switching interval. Based on the switching time window constraint, the moving particle size crushing evaluation is performed in the combined optimization strategy to obtain the maximum evaluation difference within the time window and determine the switching node. The combined optimization strategy is divided into crushing mode switching based on the switching node to obtain the crushing particle size processing management strategy.

10. A crushing particle size processing management system for granular materials in a uniform state, characterized in that, The system is used to perform the particle size management method for crushing particulate materials in a uniform state as described in any one of claims 1 to 9, the system comprising: The profile creation module is used to identify the state of particulate materials and create a state profile of the particulate materials. The feature path construction module is used to obtain the processing evaluation feature library of mechanical crushing space and airflow crushing space, sort them according to the crushing particle size based on the processing evaluation, and construct crushing feature paths; The minimum path determination module is used to search for the minimum path for crushing based on the crushing characteristic path, starting from the state profile of the particulate material and ending at the target crushing particle size. The management strategy acquisition module is used to perform response relationship optimization search based on the minimum crushing path, equipment constraints, and processing objectives to obtain the crushing particle size processing management strategy, and feed it back to the Internet of Things control system for crushing processing management and control.

Images