Solid waste treatment equipment collaborative control system based on industrial internet of things

Abstract

The application discloses a solid waste treatment equipment collaborative control system based on an industrial Internet of Things. The system quantifies material complexity in real time by extracting visual entropy features, and solves the optimal flow rate by combining the frequency response limit of actuators and the dynamic fatigue state, thereby accurately controlling the speed of the conveyor belt using a time-space synchronization mechanism. In this way, it can effectively smooth the load fluctuations under complex working conditions, prevent the response collapse of the actuator under continuous high load, and suppress the inertial displacement of the material and the failure of visual positioning caused by sudden speed changes. At the same time of ensuring the reliability of the equipment throughout its life cycle, the collaborative efficiency and operation accuracy of solid waste sorting are significantly improved.

Description

Technical Field

[0001] This application relates to the field of intelligent control, and more specifically, to a collaborative control system for solid waste treatment equipment based on the Industrial Internet of Things. Background Technology

[0002] With increasingly stringent global environmental protection requirements and the large-scale development of the solid waste treatment industry, achieving efficient linkage of heterogeneous equipment such as crushing, conveying, and sorting has become a core demand for industry development. The core significance of constructing a collaborative control scheme for solid waste treatment equipment lies in solving bottleneck problems such as low production efficiency, drastic equipment load fluctuations, and difficulty in guaranteeing sorting accuracy caused by the complex and variable composition of materials through dynamic optimization of the entire process.

[0003] However, existing solid waste treatment control schemes still face the challenge of closed-loop coordination of heterogeneous equipment under dynamic load balancing in practical industrial applications. Due to the highly heterogeneous composition of solid waste, such as moisture, hardness, and volume, most existing technologies rely on fixed preset parameters, lacking the ability to adaptively adjust to complex operating conditions. Once the material composition undergoes a sudden change, such as a large influx of large-sized or heavy waste into the production line, it can easily trigger a surge in current in the upstream crusher and instantaneous blockage in the downstream sorting equipment, causing frequent production line shutdowns. This static mapping mechanism often ignores the physical fatigue of actuators, dynamic air pressure recovery, and inertial constraints during material movement, leading to system response collapse under continuous high-load conditions, or excessively abrupt conveyor belt speed changes causing material displacement on the belt, resulting in the failure of preceding visual positioning and severely impacting the overall stability and sorting accuracy of the system.

[0004] Therefore, an optimized collaborative control system for solid waste treatment equipment based on the Industrial Internet of Things is needed. Summary of the Invention

[0005] To address the aforementioned technical issues, this application provides a collaborative control system for solid waste treatment equipment based on the Industrial Internet of Things.

[0006] According to one aspect of this application, a collaborative control system for solid waste treatment equipment based on the Industrial Internet of Things is provided, comprising: The image acquisition module is used to acquire real-time image streams and depth-height maps captured by the upstream predictive camera; The solid waste flow visual entropy feature extraction and quantization module is used to extract and quantize the solid waste flow visual entropy features from the real-time image stream and depth height map to obtain the material visual entropy index of the current frame. The valve status acquisition module is used to acquire the highest operating frequency and valve spatial resolution of the valve array. The optimal flow rate calculation module is used to calculate the optimal flow rate based on the actuator frequency response of the material visual entropy index, the highest operating frequency of the valve array, and the valve spatial resolution in the current frame to obtain the optimal target flow rate at the current moment. The control status acquisition module is used to acquire the physical distance from the sensor to the actuator and the current conveyor belt speed; The spatiotemporal synchronization and cooperative speed change module is used to perform spatiotemporal synchronization and cooperative speed change on the physical distance from the sensor to the actuator, the current conveyor belt speed, and the optimal target flow speed at the current moment to obtain a time-stamped speed command sequence to be sent to the frequency converter.

[0007] Compared with existing technologies, this application provides a collaborative control system for solid waste treatment equipment based on the Industrial Internet of Things (IIoT). This system extracts visual entropy features to quantify material complexity in real time and calculates the optimal flow rate by combining actuator frequency response limits and dynamic fatigue states. This allows for precise control of the conveyor belt speed using a spatiotemporal synchronization mechanism. This approach effectively smooths load fluctuations under complex operating conditions, prevents response collapse of actuators under continuous high loads, and suppresses material inertial displacement and visual positioning failure caused by sudden speed changes. While ensuring the reliability of the equipment throughout its entire lifecycle, it significantly improves the collaborative efficiency and operational accuracy of solid waste sorting. Attached Figure Description

[0008] The above and other objects, features, and advantages of this application will become more apparent from the more detailed description of the embodiments of this application in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the embodiments of this application to explain this application and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.

[0009] Figure 1 This is a block diagram of a collaborative control system for solid waste treatment equipment based on the Industrial Internet of Things, according to an embodiment of this application. Figure 2 This is a schematic diagram of data flow in a collaborative control system for solid waste treatment equipment based on the Industrial Internet of Things, according to an embodiment of this application. Detailed Implementation

[0010] Hereinafter, exemplary embodiments according to this application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments of this application. It should be understood that this application is not limited to the exemplary embodiments described herein.

[0011] As indicated in this application and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not specifically singular and may include plural forms. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of explicitly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0012] While this application makes various references to certain modules of the systems according to embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The modules described are merely illustrative, and different aspects of the systems and methods may use different modules.

[0013] Flowcharts are used in this application to illustrate the operations performed by the system according to embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0014] The technical solution of this application proposes a collaborative control system for solid waste treatment equipment based on the Industrial Internet of Things. Figure 1 This is a block diagram of a collaborative control system for solid waste treatment equipment based on the Industrial Internet of Things, according to an embodiment of this application. Figure 2 This is a schematic diagram of data flow in a collaborative control system for solid waste treatment equipment based on the Industrial Internet of Things (IIoT) according to an embodiment of this application. Figure 1 and Figure 2 As shown, the collaborative control system 300 for solid waste treatment equipment based on the Industrial Internet of Things according to an embodiment of this application includes: an image acquisition module 310, used to acquire real-time image streams and depth-height maps collected by an upstream predictive camera; a solid waste stream visual entropy feature extraction and quantization module 320, used to extract and quantize solid waste stream visual entropy features from the real-time image streams and depth-height maps to obtain the material visual entropy index of the current frame; a valve status acquisition module 330, used to acquire the highest operating frequency and valve spatial resolution of the valve array; an optimal flow rate calculation module 340, used to calculate the optimal flow rate based on the actuator frequency response of the material visual entropy index, the highest operating frequency of the valve array, and the valve spatial resolution of the current frame to obtain the optimal target flow rate at the current moment; a control status acquisition module 350, used to acquire the physical distance from the sensor to the actuator and the current conveyor belt speed; and a spatiotemporal synchronization and collaborative speed change module 360, used to perform spatiotemporal synchronization and collaborative speed change on the physical distance from the sensor to the actuator, the current conveyor belt speed, and the optimal target flow rate at the current moment to obtain a time-stamped speed command sequence sent to the frequency converter.

[0015] Specifically, the image acquisition module 310 is used to acquire real-time image streams and depth-height maps collected by the upstream predictive camera. It should be understood that the solid waste treatment environment is highly uncertain, and the physical properties of materials, such as volume, texture, and stacking state, are dynamically changing. The real-time image stream contains surface visual information of the solid waste material on the conveyor belt, such as color, texture, and contour, and is the cornerstone for subsequent waste target segmentation and visual feature extraction. The importance of the depth-height map lies in its provision of height information and volume distribution of the material in three-dimensional space, enabling the quantification of the material's stacking thickness and spatial topological complexity. By acquiring these two modalities of perception data, the system can anticipate the energy status of the upstream material flow, providing feedforward compensation for the downstream valve array's operating frequency adjustment and the coordinated speed change of the conveyor belt, thereby avoiding system response collapse or blockage caused by sudden changes in material flow.

[0016] In practice, the system first continuously acquires data using a predictive camera (typically an industrial camera with integrated depth sensing capabilities, such as a structured light or binocular camera) deployed at a specific location upstream of the conveyor belt. This camera is configured to synchronously capture RGB color images and depth information of the scene at a fixed frame rate (e.g., 30 frames per second). Next, the system establishes a stable network connection with the predictive camera via an industrial IoT gateway, ensuring low-latency and high-reliability data transmission. Then, the image acquisition module within the system continuously listens to a designated data port, receiving data packets encapsulated by the camera SDK or standard video streaming protocols (such as RTSP) in real time. This module then parses and decodes the received raw data stream, separating it into two independent but strictly timestamped data sequences: one is a two-dimensional pixel matrix reflecting the appearance of the object's surface, i.e., the real-time image stream; the other is a matrix representing the distance of each pixel to the Z-axis distance in the camera coordinate system, i.e., the depth-height map. Finally, the system caches the parsed real-time image stream and depth-height map in a designated buffer in memory, and adds a unified timestamp and frame sequence number label for subsequent solid waste stream visual entropy feature extraction and quantization modules to call in sequence. The entire acquisition process requires ensuring data integrity and temporal consistency, ensuring that the image of each frame can be accurately matched with its corresponding depth-height map.

[0017] Specifically, the solid waste flow visual entropy feature extraction and quantification module 320 is used to extract and quantify the solid waste flow visual entropy features from the real-time image stream and depth height map to obtain the material visual entropy index of the current frame. It should be understood that the material composition at the solid waste treatment site, such as humidity, hardness, volume, and stacking state, exhibits extremely high heterogeneity. Traditional control schemes often operate solely based on set parameters; once the material composition undergoes a sudden change, it can easily lead to production line blockage, a surge in instantaneous load on the sorting machine, or frequent equipment shutdowns. In the technical solution of this application, by extracting visual entropy from the real-time image stream and depth height map, visual perception data can be transformed into physical property constants reflecting the essential characteristics of the material (such as fluidization performance and breakability), thereby solving the problems of missing physical meaning in perception data and lagging control feedback in existing schemes, and providing accurate feedforward compensation constraints for subsequent optimal flow rate calculation.

[0018] In practice, the system first performs waste target segmentation and multimodal feature mapping coefficient extraction on the real-time image stream and depth height map to obtain a material contour set and a material category mapping table. During this process, the system first inputs the timestamp-aligned real-time image stream (RGB images) and depth height map into a pre-trained deep learning segmentation model (e.g., based on U-Net or Mask R-CNN architecture). This model utilizes the color and texture information of the RGB images and the geometric information of the depth map to classify each pixel in the image, distinguishing between the background (conveyor belt) and the foreground (each solid waste object), thus obtaining the precise pixel-level contour of each individual waste object. The set of these contours constitutes the material contour set. Simultaneously, the model identifies the material category (e.g., plastic, metal, paper) of each segmented foreground object and organizes the identification results (object ID, material category, confidence score) into a structured table, i.e., the material category mapping table.

[0019] Next, the overlap rate of the material contour set is quantized based on spatial topological complexity to obtain the stacking overlap rate. In this process, the system calculates the total area of ​​the minimum outer envelope rectangle occupied by all material contours; then, by calculating the intersection between each contour, the overlapping areas of any two or more contours are determined, and the areas of these overlapping areas are accumulated, i.e., the ratio of the total overlapping area to the total area covered by the contours, to identify the mutual occlusion of materials in three-dimensional space. Specifically, the overlap rate of the material contour set is quantized based on spatial topological complexity using the following formula: ; in, The sum of the areas where the outlines of two or more solid waste targets overlap. The total effective coverage area within the smallest outer envelope rectangle occupied by all solid waste targets.

[0020] Furthermore, the information entropy of material distribution heterogeneity is calculated on the material category mapping table to obtain the category heterogeneity index. That is, the information entropy of material distribution heterogeneity is calculated on the material category mapping table to obtain the category heterogeneity index, which reflects the degree of disorder of components in the current material flow. Specifically, suppose there are N different material categories, and the ratio of the number of objects of the k-th material category to the total number of objects is... Then the category heterogeneity index The entropy is calculated using the following formula: a higher entropy value indicates a more chaotic material distribution and stronger heterogeneity. Specifically, the formula is: ; in, Let k be the ratio of the number of objects of the kth material to the total number of objects in the current area. This represents the total number of different material categories present within the current visual recognition area.

[0021] Subsequently, a weighted convergence of the category heterogeneity index and the stacking overlap rate is performed to obtain the material visual entropy index of the current frame. That is, the stacking overlap rate and the category heterogeneity index, which are calculated above, representing spatial complexity and material complexity respectively, are fused. Specifically, this is achieved through preset weighting coefficients (e.g., ...). and ,and The two indicators are weighted and summed to obtain a comprehensive material visual entropy index, which reflects the overall complexity of the current material flow. The higher the index, the more severe the material stacking or the more complex the material types.

[0022] Specifically, the valve status acquisition module 330 is used to acquire the maximum operating frequency and spatial resolution of the valve array. It should be understood that actuators in a solid waste treatment environment are not ideal electronic switches; their physical actions are strictly constrained by response limits. The maximum operating frequency of the valve array defines the upper limit of the actuator's operation in the time dimension. Acquiring this parameter can effectively prevent instantaneous pressure drops in the air supply system or overload of the solenoid valve coil due to excessively rapid command issuance, thereby avoiding system response collapse under continuous high-load conditions. The valve spatial resolution quantifies the impact accuracy of the equipment in the spatial dimension, determining the minimum particle size of material that can be processed per unit distance, and is the fundamental physical benchmark for calculating the balance between target flow rate and sorting accuracy. If the control command exceeds the physical capabilities of the valve array, such as requiring it to operate at a speed higher than the maximum operating frequency, it will cause the valve to fail to respond in time or increase mechanical wear; if the material distribution density is higher than the spatial resolution of the valve, it will lead to a decrease in sorting accuracy. By obtaining the highest operating frequency and spatial resolution of the valve array, it is possible to ensure that the control commands meet both the sorting requirements and the rigid constraints of the physical equipment, which is the foundation for achieving precise collaborative control.

[0023] Among them, the highest operating frequency of the air valve array refers to the maximum number of cycles that the air valve can complete per second while ensuring the blowing efficiency; the air valve spatial resolution refers to the physical arrangement spacing or effective coverage diameter of the air valve nozzles in the sorting area, which reflects the precision of the system in spatially cutting the material flow.

[0024] In practice, the configuration protocol of the underlying actuator can be read in real time through the industrial IoT gateway to obtain its rated maximum operating frequency and spatial resolution in the spatial layout parameters.

[0025] Specifically, the optimal flow rate calculation module 340 is used to calculate the optimal target flow rate based on the actuator frequency response of the material visual entropy index, the maximum operating frequency of the valve array, and the valve spatial resolution in the current frame to obtain the optimal target flow rate at the current moment. It should be understood that the sorting accuracy in solid waste treatment is highly dependent on the matching degree between the actuator (valve array) and the material flow rate. Due to the physical maximum operating frequency limit of the valves, when the upstream material complexity (visual entropy) surges, if the conveyor belt continues to operate at a high rated speed, the valve array will be unable to achieve precise strikes due to insufficient response frequency, leading to decreased sorting efficiency or equipment overload. Through optimal flow rate calculation based on actuator frequency response, the system can dynamically adjust the conveyor belt speed according to the real-time status of the material, ensuring that each sorting target can be processed within the effective response window of the actuator, thereby preventing system response collapse.

[0026] In practice, firstly, the maximum operating frequency of the valve array is calibrated using the actuator's time-domain limit to obtain the minimum operating cycle period. During this process, the system reads the pre-acquired maximum operating frequency of the valve array. The minimum operating cycle period is the reciprocal of the maximum operating frequency, representing the shortest time required for the valve array to complete one full "open-close" action. This calibration logic establishes a mapping relationship between frequency and time step. For example, if the maximum operating frequency of the valve is 100 Hz, the minimum operating cycle period is 0.01 seconds (10 milliseconds). This calibrates the physical lower limit of the actuator's time response.

[0027] Next, based on the minimum action cycle period and the entropy safety threshold, the entropy deviation and time dilation coefficient of the material visual entropy index in the current frame are evaluated to obtain the processing time density required per unit space. It should be understood that traditional visual entropy-based solid waste sorting collaborative control mechanisms calculate the time dilation coefficient in isolation, only based on the difference between the entropy value of the current frame and a preset threshold, assuming a simple linear or quasi-linear relationship between time dilation and entropy overshoot. This approach ignores two key physical constraints in the Industrial Internet of Things (IIoT) scenario: the dynamic correlation between actuator physical fatigue and air pressure recovery, and the influence of material flow inertial coupling. Specifically, in industrial solid waste sorting, the actuator (valve array) is not an ideal electronic switch. Continuous high-entropy (high-complexity) materials can cause instantaneous drops in air source pressure or temperature rises in the solenoid valve coil; this cumulative effect of complexity over time is not considered. Simultaneously, when the material stack height exceeds a certain threshold, the blowing efficiency of the valve decreases exponentially, and linear mapping cannot reflect this nonlinear distortion. Furthermore, the conveyor belt's speed variation inherently involves mechanical inertia. If the speed command is drastically changed solely based on the current entropy value, it will cause equipment vibration and material displacement due to inertia, rendering the preceding visual positioning ineffective. To address these shortcomings, this application proposes a state-aware dynamic evolution model, improving upon the traditional mechanism by introducing actuator state memory factors and inertial constraint weights.

[0028] In this process, firstly, the actuator load accumulation state is evaluated by assessing the material visual entropy index, entropy safety threshold, and load index of the previous moment in the current frame to obtain the actuator dynamic fatigue factor. This step aims to evaluate the load accumulation state of the actuator under continuous high-intensity sorting tasks. That is, it quantifies the cumulative impact of continuous complexity in the time dimension on physical devices to compensate for the shortcomings of the original mechanism that only focuses on instantaneous entropy values. Specifically, using first-order hysteresis filtering logic, combined with the load index of the previous moment and the entropy overshoot of the current frame, a dynamic fatigue factor characterizing the current fatigue level of the equipment is calculated. This factor can identify continuous high-load scenarios. When the system detects multiple frames of continuous complex material flow, the fatigue factor will continue to rise, which means that the air pressure reserve of the sorting air valve array may be insufficient or the solenoid valve may be at risk of overheating. Therefore, a more conservative and predictive time offset is needed to ensure the effectiveness of the sorting action. In this way, a memory index that reflects the historical working intensity of the actuator can be obtained, providing a key state input for subsequent nonlinear responses. More specifically, this process is expressed by the formula:

[0029] in, The dynamic fatigue factor of the actuator represents a dimensionless scalar at time t, characterizing the cumulative load state of the actuator. It is a coefficient between 0 and 1, representing the historical memory weight, which is used to adjust the degree of influence of historical load on the current fatigue state; The load index at the previous moment, i.e. The value; The material visual entropy index is a complex quantification of the current frame; This is the entropy safety threshold, representing a preset complexity critical point; This is a function that maximizes the value of a function to ensure that negative fatigue does not occur when the entropy value does not exceed the threshold.

[0030] Next, a nonlinear complexity-response coupling solution is performed on the actuator dynamic fatigue factor, the material visual entropy index of the current frame, the entropy safety threshold, and the sensitivity adjustment factor to obtain a nonlinear expansion gain. Since traditional linear mapping cannot accurately simulate the phenomenon of efficiency rapidly decreasing with load in the physical world, the technical solution of this application uses an exponential function to couple the actuator dynamic fatigue factor obtained in the previous step with the current entropy value, thereby calculating a nonlinear expansion gain. Specifically, when the actuator is already in a high fatigue state and the current material complexity is still high, the introduced exponential term... This will cause the expansion gain to increase nonlinearly, or even exponentially. This generates an amplification factor that reflects the physical constraints under extremely complex operating conditions, forcing the system to adopt a more drastic deceleration strategy when risks accumulate, ensuring sufficient pressure recovery window or preventing coil burnout, thus guaranteeing system robustness. Specifically, this process is expressed by the following formula: ; in, The nonlinear expansion gain is a dimensionless value used to amplify the fundamental time period. This is a sensitivity adjustment factor used to adjust the system's response strength to changes in conventional entropy. The physical environment coupling coefficient is a constant calibrated based on parameters such as the rated flow rate of a specific gas supply system. It is used to adjust the exponential influence of fatigue effects on the gain.

[0031] Furthermore, the processing time density is synthesized by constraining and optimizing the nonlinear expansion gain, minimum action cycle period, current conveyor belt speed, and inertial constraint weights to obtain the processing time density required per unit space. Specifically, the nonlinear expansion gain calculated in the previous step is applied to the basic minimum action cycle of the valve, and a momentum constraint penalty term related to velocity change is introduced. This process is expressed by the following formula: ; in, This represents the processing time density required per unit space, expressed in seconds, and is the final time parameter used for velocity calculation. The minimum action cycle period represents the lower limit of the physical time of the actuator; The inertial constraint weights are used to adjust the smoothness of velocity changes. The current conveyor belt speed, The optimal speed for the previous cycle, i.e., the speed calculated in the previous control cycle. Value. By introducing This feature allows the system to detect drastic changes in speed commands. Specifically, if the calculated new speed differs significantly from the current speed, this feature increases the final time density, thus smoothly suppressing sudden speed changes. In this way, it ensures that changes in time density do not create physically unfeasible speed jumps, preventing material slippage due to relative inertia on the conveyor belt. This allows for the synthesis of a final unit space processing time density that reflects material complexity, accommodates actuator states, and guarantees smooth motion, while ensuring sorting accuracy. This provides a stable, reliable, and physically complete input for subsequent speed calculations.

[0032] In summary, this mechanism transforms the process from static, isolated complexity mapping to dynamic, state-aware processing time assessment. This allows the system to dynamically adjust processing time based on the instantaneous complexity of the material, while also considering the physical fatigue state of the actuators over time and the dynamic recovery capability of the air supply system. Through an exponential response mechanism, it effectively prevents system response collapse and equipment damage under continuous high-load conditions. Simultaneously, by introducing inertial constraints, the smoothness of the conveyor belt speed change process is ensured, resolving material slippage and visual positioning failure issues caused by sudden speed changes, thus ensuring the physical consistency of high-speed, high-precision sorting. Ultimately, this mechanism enables the system to improve the robustness, stability, and reliability of the entire industrial IoT collaborative control system throughout the equipment's lifecycle while ensuring the sorting accuracy of complex materials, achieving globally optimal control under multiple physical constraints.

[0033] Subsequently, based on a comparison between the material visual entropy index and the entropy safety threshold of the current frame, a nonlinear mapping solution is performed on the valve spatial resolution, the system's rated maximum speed, and the processing time density required per unit space to obtain the optimal target flow rate at the current moment. This step aims to solve the problem of production line blockage and decreased sorting accuracy caused by sudden changes in material composition in the solid waste sorting process. Specifically, through this step, the system can dynamically adjust the flow rate at the downstream execution end based on the material complexity (visual entropy) predicted upstream, ensuring that each sorting target can be accurately processed within the actuator's response window. The entropy safety threshold is a system-preset complexity threshold that guarantees sorting accuracy at the rated flow rate. The processing time density required per unit space is a dynamic correction cycle considering actuator fatigue and material characteristics. The final optimal target flow rate refers to the highest theoretical operating speed that the conveyor belt can achieve under the premise of meeting the complexity of the current sorting task and not exceeding the physical limits of the actuator.

[0034] In this process, the rated maximum speed of the system and the processing time density required per unit space are solved nonlinearly using the following formula: ; in, For the space resolution of the air valve, The processing time density required per unit space, The system's rated maximum speed is used. This means that when the comparison result shows the material's visual entropy index is less than or equal to the entropy safety threshold, the material complexity in the current frame is below or equal to the safety threshold, and the material is considered to be in a standard state. In this case, the system does not need to reserve extra time for sorting; therefore, the optimal strategy is to allow the conveyor belt to run at the system's maximum allowable speed, i.e., directly outputting the system's rated maximum speed as the optimal target flow rate, aiming to maximize the system's processing throughput. When the comparison result shows the material's visual entropy index is greater than the entropy safety threshold, it means the material complexity exceeds the safety threshold and is in a complex state. In this case, the system must reduce its speed to ensure the valve array has sufficient time to accurately sort the complex material. In this situation, the optimal target flow rate is determined by the ratio of the valve spatial resolution to the processing time density required per unit space.

[0035] Specifically, the control state acquisition module 350 is used to acquire the physical distance between the sensor and the actuator and the current conveyor belt speed. Since solid waste sorting involves cross-spatial collaboration between "upstream sensing and downstream execution," there is a physical transport delay from when the material is captured by the pre-judgment camera to when it enters the air valve spraying area. Therefore, in the technical solution of this application, by acquiring the physical distance between the two and combining it with the real-time conveyor belt speed, the system can establish a dynamic mapping model of the material flow on the belt, thereby solving the positioning deviation problem caused by the failure of spatiotemporal decoupling between sensing and execution. The physical distance between the sensor and the actuator refers to the linear displacement between the projection point of the optical center of the upstream pre-judgment camera onto the conveyor belt plane and the spray centerline of the downstream air valve array; the current conveyor belt speed is the instantaneous linear speed of the belt, collected in real time by a frequency converter feedback or an encoder installed on the roller.

[0036] In practice, the static configuration parameters and dynamic sensor data of the underlying devices are read in real time through the industrial IoT gateway. During this process, the physical distance can be determined by retrieving the physical location information stored in the system's EEPROM, and a specific data read command frame can be sent to the frequency converter to request the frequency value currently output by the frequency converter or directly obtain the motor speed feedback value calculated internally by the frequency converter.

[0037] Specifically, the spatiotemporal synchronization and cooperative speed control module 360 ​​is used to perform spatiotemporal synchronization and cooperative speed control on the physical distance from the sensor to the actuator, the current conveyor belt speed, and the optimal target flow rate at the current moment to obtain a time-stamped speed command sequence to be sent to the frequency converter. It should be understood that in the collaborative control system of solid waste treatment equipment based on the Industrial Internet of Things, the visual sensor (upstream predictive camera) and the actuator (valve array) are physically separated, and the material moves between them via a conveyor belt. If the optimal target flow rate command calculated based on the current visual information is directly sent to the frequency converter, the material may not have arrived or may have passed the predetermined position when the valve is activated due to the conveyor belt transport time between the material's identification point and the sorting point, resulting in sorting failure or reduced efficiency. Therefore, in the technical solution of this application, through spatiotemporal synchronization and cooperative speed control, the system can coordinate the calculated target flow rate with the material's transport delay on the conveyor belt and the current operating state to generate a speed command that takes effect at the correct time, thereby ensuring that visual recognition, position prediction, and valve action are precisely aligned in time and space, achieving reliable sorting control.

[0038] In practice, the first step is to calculate the time required for the material to move from the sensor (camera) position to the actuator (valve array) position, i.e., the transmission delay. This delay is determined by both the physical distance between the sensor and the actuator and the instantaneous speed of the material. Specifically, using the current conveyor belt speed as the instantaneous speed of the material, the calculation process for the transmission delay can be expressed by the formula: ; in, The physical distance from the sensor to the actuator. The current conveyor belt speed, This is for transmission delay. It's worth noting that this calculation is based on the assumption of uniform motion, providing a baseline delay for setting timestamps for subsequent instructions.

[0039] Next, after obtaining the transmission delay and the optimal target flow rate calculated at the current moment, the system needs to plan the timing of speed adjustments. The timing of adjustments must consider two key factors: first, ensuring that the affected material arrives at the valve sorting area precisely after the speed change takes effect; and second, recognizing that the speed change process itself takes time and should avoid interfering with the position of materials still in transit. Therefore, the system generates a future speed switching point. This process can be expressed by the formula: ; in, The system response and mechanical inertia compensation time is a constant obtained based on the inverter's response characteristics and the conveyor belt's mechanical inertia calibration. It is used to compensate for the time required from sending a command to speed stabilization, ensuring a smooth speed transition before the material arrives. As the speed switching point, This is the current time.

[0040] Then, the arranged speed instructions are encapsulated into a timestamped instruction sequence. This sequence contains at least two instructions: the first instruction is timestamped. Effective, setting the conveyor belt speed to the optimal target flow rate. The second instruction, serving as a safety or default instruction, can be set for the more distant future or as a hold instruction. The sequence is organized in the form of digital communication messages, each containing a precise timestamp and a corresponding target speed value. This sequence is then sent to the conveyor belt frequency converter, which performs speed switching at specified times based on the timestamps, thereby achieving strict spatiotemporal synchronization with upstream visual sensing and downstream valve action.

[0041] As described above, the industrial IoT-based solid waste treatment equipment collaborative control system 300 according to the embodiments of this application can be implemented in various wireless terminals, such as servers with industrial IoT-based solid waste treatment equipment collaborative control algorithms. In one possible implementation, the industrial IoT-based solid waste treatment equipment collaborative control system 300 according to the embodiments of this application can be integrated into the wireless terminal as a software module and / or hardware module. For example, the industrial IoT-based solid waste treatment equipment collaborative control system 300 can be a software module in the operating system of the wireless terminal, or it can be an application developed for the wireless terminal; of course, the industrial IoT-based solid waste treatment equipment collaborative control system 300 can also be one of many hardware modules of the wireless terminal.

[0042] Alternatively, in another example, the industrial IoT-based solid waste treatment equipment collaborative control system 300 and the wireless terminal can also be separate devices, and the industrial IoT-based solid waste treatment equipment collaborative control system 300 can be connected to the wireless terminal via wired and / or wireless networks, and transmit interactive information in accordance with an agreed data format.

[0043] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A collaborative control system for solid waste treatment equipment based on the Industrial Internet of Things, characterized in that, include: The image acquisition module is used to acquire real-time image streams and depth-height maps captured by the upstream predictive camera; The solid waste flow visual entropy feature extraction and quantization module is used to extract and quantize the solid waste flow visual entropy features from the real-time image stream and depth height map to obtain the material visual entropy index of the current frame. The valve status acquisition module is used to acquire the highest operating frequency and valve spatial resolution of the valve array. The optimal flow rate calculation module is used to calculate the optimal flow rate based on the actuator frequency response of the material visual entropy index, the highest operating frequency of the valve array, and the valve spatial resolution in the current frame to obtain the optimal target flow rate at the current moment. The control status acquisition module is used to acquire the physical distance from the sensor to the actuator and the current conveyor belt speed; The spatiotemporal synchronization and cooperative speed change module is used to perform spatiotemporal synchronization and cooperative speed change on the physical distance from the sensor to the actuator, the current conveyor belt speed, and the optimal target flow speed at the current moment to obtain a time-stamped speed command sequence to be sent to the frequency converter.

2. The collaborative control system for solid waste treatment equipment based on the Industrial Internet of Things as described in claim 1, characterized in that, The solid waste stream visual entropy feature extraction and quantification module includes: The target segmentation and feature mapping unit is used to segment waste targets and extract multimodal feature mapping coefficients from real-time image streams and depth height maps to obtain material contour sets and material category mapping tables; The overlap rate quantization unit is used to perform overlap rate quantization on the material profile set based on spatial topological complexity to obtain the stacking overlap rate. The information entropy calculation unit is used to calculate the information entropy of material distribution heterogeneity in the material category mapping table to obtain the category heterogeneity index. The weighted convergence unit is used to perform weighted convergence on the category heterogeneity index and the stacking overlap rate to obtain the material visual entropy index of the current frame.

3. The collaborative control system for solid waste treatment equipment based on the Industrial Internet of Things as described in claim 2, characterized in that, The overlap rate quantization unit is used to: quantify the overlap rate of the material contour set based on spatial topological complexity using the following formula: ； in, The sum of the areas where the outlines of two or more solid waste targets overlap. The total effective coverage area within the smallest outer envelope rectangle occupied by all solid waste targets.

4. The collaborative control system for solid waste treatment equipment based on the Industrial Internet of Things as described in claim 2, characterized in that, The information entropy calculation unit is used to calculate the information entropy of material distribution heterogeneity in the material category mapping table using the following formula: ； in, Let k be the ratio of the number of objects of the kth material to the total number of objects in the current area. This represents the total number of different material categories present within the current visual recognition area.

5. The collaborative control system for solid waste treatment equipment based on the Industrial Internet of Things as described in claim 1, characterized in that, The optimal flow rate calculation module includes: The time-domain limit calibration unit is used to perform time-domain limit calibration of the actuator action at the highest operating frequency of the valve array in order to obtain the minimum operating cycle period. The entropy deviation and time dilation coefficient evaluation unit is used to evaluate the entropy deviation and time dilation coefficient of the visual entropy index of the material in the current frame based on the minimum action cycle period and the entropy safety threshold in order to obtain the processing time density required per unit space. The nonlinear mapping solution unit is used to perform nonlinear mapping solutions on the gas valve spatial resolution, the system's rated maximum speed, and the processing time density required per unit space based on the comparison between the material visual entropy index and the entropy safety threshold in the current frame, in order to obtain the optimal target flow rate at the current moment.

6. The collaborative control system for solid waste treatment equipment based on the Industrial Internet of Things as described in claim 5, characterized in that, Entropy deviation and time dilation coefficient evaluation unit, used for: The actuator's dynamic fatigue factor is obtained by evaluating the cumulative load status of the actuator based on the material's visual entropy index, entropy safety threshold, and load index from the previous moment in the current frame. The nonlinear complexity-response coupling solution is performed on the actuator dynamic fatigue factor, the material visual entropy index of the current frame, the entropy safety threshold, and the sensitivity adjustment factor to obtain the nonlinear expansion gain; The processing time density required per unit space is synthesized by constraining and optimizing the nonlinear expansion gain, minimum action cycle period, current conveyor belt speed, and inertial constraint weight.

7. The collaborative control system for solid waste treatment equipment based on the Industrial Internet of Things as described in claim 1, characterized in that, The nonlinear mapping solution unit is used to perform a nonlinear mapping solution for the system's rated maximum speed and the required processing time density per unit space using the following formula: ； in, For the space resolution of the air valve, The processing time density required per unit space, This is the system's rated maximum speed.

Images