Solid waste classification control system based on multi-sensor recognition

Abstract

The present invention discloses a solid waste classification control system based on multi-sensor identification, which relates to the technical field of waste classification, including: transferring mixed waste to a pre-treatment area for vibration screening; receiving the mixed waste after vibration screening, establishing weight fuzzy recognition results and volume fuzzy recognition results, sending the weight fuzzy recognition results and volume fuzzy recognition results to a decision-making layer, setting the acceleration of an accelerating conveyor belt, and transferring mixed waste; when the mixed waste reaches a preset area, performing individual identification of the transferred waste and establishing an individual data set; performing mixed separation evaluation based on the individual data set, controlling a pneumatic sorting system and a classification mechanical arm to perform linkage sorting based on the mixed separation evaluation results, and performing a diversion transmission mark, and completing solid waste classification control according to the diversion transmission mark. The present invention solves the technical problem of low classification accuracy in the prior art when performing solid waste classification, and achieves the technical effect of accurate identification and classification of waste.

Description

Technical Field

[0001] The present invention relates to the technical field of waste classification, and in particular to a solid waste classification control system based on multi-sensor identification. Background Art

[0002] With the rapid development of the global economy and the acceleration of urbanization, the amount of solid waste generated has shown explosive growth. Solid waste is of various types and complex in composition. How to efficiently and accurately classify and treat these wastes has become an important challenge faced by governments and enterprises. The existing automated solid waste sorting system has improved the sorting efficiency to a certain extent, but there are still many shortcomings. Traditional automated sorting systems mainly rely on single or a few sensor technologies, such as visual sensors based on image recognition or simple weight sensors. These systems often find it difficult to achieve comprehensive and accurate identification and classification when faced with complex and diverse waste components, resulting in low classification accuracy. Summary of the invention

[0003] The present application provides a solid waste classification control system based on multi-sensor identification, which is used to solve the technical problem of low classification accuracy in the prior art when classifying solid waste.

[0004] In view of the above problems, the present application provides a solid waste classification control system based on multi-sensor identification.

[0005] The present application provides a solid waste classification control system based on multi-sensor recognition, the system comprising:

[0006] The pretreatment unit is used to convey the mixed waste to the pretreatment area through the primary conveyor belt, and vibrate the mixed waste in the pretreatment area; the transmission and separation unit is used to receive the mixed waste after vibration screening based on the acceleration conveyor belt, and perform fuzzy identification of individual weights in the mixed waste through the first weight sensor built under the acceleration conveyor belt, establish a weight fuzzy identification result, collect individual images of the mixed waste through the image sensor, establish a volume fuzzy identification result, send the weight fuzzy identification result and the volume fuzzy identification result to the decision layer, set the acceleration of the acceleration conveyor belt based on the decision layer, and transmit the mixed waste after vibration screening through the acceleration conveyor belt; the identification unit is used to activate the linkage sensor group to perform individual identification of the transmitted waste when the transmitted mixed waste reaches the preset area, and establish an individual data set, and the linkage sensor includes a visual sensor, a near-infrared spectrum sensor, an X-ray sensor, and a second weight sensor; the classification unit is used to perform mixed separation evaluation based on the individual data set, and control the pneumatic sorting system and the classification robot arm to perform linkage sorting based on the mixed separation evaluation result; the identification unit is used to perform diversion and transmission identification on the linkage sorting result, and complete the solid waste classification control according to the diversion and transmission identification.

[0007] The technical solution provided in this application has at least the following technical effects or advantages:

[0008] The present application transmits mixed waste to a pretreatment area through a primary conveyor belt, and performs vibration screening of the mixed waste in the pretreatment area; receives the mixed waste after vibration screening based on an accelerating conveyor belt, and performs fuzzy identification of individual weights in the mixed waste through a first weight sensor built under the accelerating conveyor belt, establishes a weight fuzzy identification result, collects individual images of the mixed waste through an image sensor, establishes a volume fuzzy identification result, sends the weight fuzzy identification result and the volume fuzzy identification result to a decision layer, sets the acceleration of the accelerating conveyor belt based on the decision layer, and transmits the mixed waste after vibration screening through the accelerating conveyor belt; when the transmitted mixed waste reaches a preset area, activates a linkage sensor group to perform individual identification of the transmitted waste, and establishes an individual data set, and the linkage sensor includes a visual sensor, a near-infrared spectrum sensor, an X-ray sensor, and a second weight sensor; performs mixed separation evaluation based on the individual data set, and controls a pneumatic sorting system and a classification robot arm to perform linkage sorting based on the mixed separation evaluation result; performs a diversion transmission mark on the linkage sorting result, and completes solid waste classification control according to the diversion transmission mark. The present invention solves the technical problem of low classification accuracy in the prior art when classifying solid waste, and achieves the technical effect of accurate identification and classification of waste by integrating multiple sensing technologies such as vibration screening, weight fuzzy recognition, image acquisition, near-infrared spectroscopy, X-ray and precise weighing. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A schematic diagram of the structure of a solid waste classification control system based on multi-sensor recognition provided in an embodiment of the present application;

[0011] Figure 2 A flow chart of a solid waste classification control method based on multi-sensor recognition provided in an embodiment of the present application.

[0012] Explanation of the reference numerals: pre-processing unit 11 , transmission and separation unit 12 , recognition unit 13 , classification unit 14 , identification unit 15 . DETAILED DESCRIPTION

[0013] The present application provides a solid waste classification control system based on multi-sensor identification, aiming to solve the technical problem of low classification accuracy in the prior art when classifying solid waste. By integrating multi-sensor technologies such as vibration screening, weight fuzzy recognition, image acquisition, near-infrared spectroscopy, X-rays and precise weighing, the technical effect of accurate identification and classification of waste is achieved.

[0014] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0015] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0016] like Figure 1 As shown, the present application provides a solid waste classification control system based on multi-sensor recognition, which is used to perform the following Figure 2 The solid waste classification control method based on multi-sensor recognition shown in the figure comprises:

[0017] The pre-treatment unit 11 is used to convey the mixed waste to a pre-treatment area through a primary conveyor belt, and vibratory screening of the mixed waste is performed in the pre-treatment area.

[0018] In the embodiment of the present application, the pretreatment unit transports the mixed waste from the initial collection point to the pretreatment area through the primary conveyor belt. In the pretreatment area, the mixed waste is preliminarily treated by vibrating screening, and the large and small pieces of materials in the mixed waste are separated by the high-speed vibration of the vibrating screen, thereby achieving the effect of preliminary screening.

[0019] The conveying and separation unit 12 is used to receive the mixed waste after vibration screening based on the accelerating conveyor belt, and perform fuzzy recognition of individual weights in the mixed waste through the first weight sensor built under the accelerating conveyor belt, establish a weight fuzzy recognition result, collect individual images of the mixed waste through the image sensor, establish a volume fuzzy recognition result, send the weight fuzzy recognition result and the volume fuzzy recognition result to the decision layer, set the acceleration of the accelerating conveyor belt based on the decision layer, and transmit the mixed waste after vibration screening through the accelerating conveyor belt.

[0020] In an embodiment of the present application, when the waste enters the conveying and separation unit, the first weight sensor built under the conveyor belt starts working. The weight sensor, such as a strain gauge or piezoelectric sensor, measures the weight of the waste passing through the conveyor belt in real time, and processes these data through a fuzzy logic algorithm, such as MATLAB's Fuzzy Logic Toolbox, to generate a weight fuzzy recognition result. Next, the image sensor collects individual images of the mixed waste. The image sensor is an industrial camera that uses image processing algorithms, such as edge detection and shape recognition in the OpenCV library, to analyze the shape and size of the waste, thereby estimating its volume and generating a volume fuzzy recognition result.

[0021] Next, the fuzzy weight recognition results and volume recognition results are sent to the decision layer, which is the core part of the system and uses the received recognition results for analysis. The rule-based decision system or machine learning model processes this data and decides to adjust the speed of the acceleration conveyor belt. These control logics are executed through a modern PLC system, which adjusts the motor speed through a variable frequency drive (VFD) to accurately control the acceleration of the conveyor belt for the mixed waste transmission after vibration screening.

[0022] Furthermore, the transmission separation unit 12 in the system provided in the embodiment of the application is also used for:

[0023] The total number of individuals of the mixed waste after vibration screening is obtained through the decision layer; the slip prediction network of the decision layer is activated, the weight fuzzy recognition result and the volume fuzzy recognition result are input into the slip prediction network, the slip distance prediction under different accelerations is performed, and the slip distance prediction result is established, and the slip distance prediction result is marked with an acceleration mapping mark; the individual initial positions of the mixed waste are obtained, and a centralized evaluation of the separation effect is performed according to the individual initial positions and the slip distance prediction results; an acceleration decision is generated according to the centralized evaluation result of the separation effect, and the acceleration of the acceleration conveyor belt is set according to the acceleration decision result.

[0024] In an embodiment of the present application, the decision layer uses the Internet of Things technology to connect various sensors and conveyor belts to monitor and record the amount of waste in real time and obtain the total number of individuals of mixed waste after vibration screening. Next, the decision layer activates its slip prediction network, which is a machine learning-based model for predicting the slip distance of waste on the conveyor belt under different accelerations. The slip prediction network is pre-built. When it is built, the weight fuzzy recognition results and volume fuzzy recognition results of a large number of mixed wastes after vibration screening are first collected from the historical database. These data are provided by the weight sensor and image sensor under the conveyor belt and processed by the fuzzy logic algorithm to generate recognition results. The collected data is then pre-processed, including data cleaning, normalization, and feature selection to ensure that the data quality is high and suitable for machine learning model training. After that, a suitable machine learning model is selected, such as a multilayer perceptron, a long short-term memory network, or a convolutional neural network, and implemented by a deep learning framework such as TensorFlow. The model is trained using a training set so that it can predict the slip distance under different accelerations. During the training process, the model parameters are adjusted by the back propagation algorithm to reduce errors and improve prediction accuracy. Subsequently, the trained model is validated using the test set to evaluate its prediction performance. If the prediction accuracy of the model does not meet the requirements, the model structure or parameters are adjusted and retrained.

[0025] When the slip prediction network is ready, it receives the weight fuzzy recognition result and the volume fuzzy recognition result as input, performs slip distance prediction under different acceleration conditions, and generates slip distance prediction results with corresponding acceleration mapping identifiers.

[0026] Next, the individual initial positions of the mixed waste are obtained through visual sensors and image processing technology, and combined with the predicted results of the slip distance to conduct a centralized evaluation of the separation effect. The centralized evaluation of the separation effect is to evaluate the separation effect of the waste during the transmission process based on the initial position of the waste and the predicted slip distance, and generate a centralized evaluation result of the separation effect.

[0027] Finally, the acceleration decision is generated based on the centralized evaluation results of the separation effect. Specifically, the rule-based decision system uses pre-set rules to directly calculate the required conveyor belt acceleration based on the centralized evaluation results of the separation effect. For example, if the waste concentration exceeds the preset threshold, the conveyor belt acceleration is reduced; if the waste dispersion exceeds the preset threshold, the conveyor belt acceleration is increased. Based on the generated acceleration decision, the decision is used to actually control the acceleration of the conveyor belt to achieve the transmission of mixed waste after vibration screening.

[0028] Furthermore, the transmission separation unit 12 in the system provided in the embodiment of the application is also used for:

[0029] Establish a calibration distance value; establish the final position of the individual based on the individual initial position and the slip distance prediction result; compare the distance of the nearest neighbor individual with the calibration distance value through the final position, if the calibration distance value is not met, assign a linear concentration value according to the distance; if the distance of the nearest neighbor individual meets the calibration distance value, set the concentration value to 0; complete the centralized evaluation of the separation effect according to the linear concentration value and the concentration value of all individuals.

[0030] In the embodiment of the present application, when the separation effect is evaluated, a calibration distance value is first established, which is a preset distance used to evaluate the separation effect of individual wastes on the conveyor belt. The calibration distance value is an ideal distance set by technical experts based on actual operating experience to ensure that individual wastes are not too concentrated or dispersed during the conveying process.

[0031] Next, the final position of each waste object is determined based on its initial position and the predicted slip distance. The predicted slip distance is calculated by the slip prediction network and reflects the slip distance of the waste under different acceleration conditions. By combining these predicted results with the initial position, the position of each waste object at the end of the transport process is located.

[0032] Then the final position of each waste individual is analyzed. Specifically, the distance between each waste individual and its nearest neighbor is compared and compared with the preset calibration distance value. If the distance between a waste individual and its nearest neighbor does not meet the calibration distance value, that is, the distance is less than the calibration distance value, a linear concentration value is assigned according to the distance. The size of the linear concentration value is inversely proportional to the distance between the waste individuals. The smaller the distance, the larger the concentration value, indicating that the waste individuals are too concentrated and the separation effect is not good. Specifically, when calculating the linear concentration value, the linear concentration value is obtained by dividing the calibration distance value by the difference between the calibration distance value and the actual distance. If the distance between a waste individual and its nearest neighbor meets or exceeds the calibration distance value, the concentration value is set to 0. A concentration value of 0 indicates that the distance between these waste individuals has reached the optimal dispersion effect and no further adjustment is required. This means that when the concentration value is 0, the waste individuals are best dispersed, and no matter how large the distance is, it will not affect the dispersion effect.

[0033] By calculating the linear concentration value and concentration value of all waste individuals, the centralized evaluation of separation effect is completed.

[0034] Furthermore, the transmission separation unit 12 in the system provided in the embodiment of the application is also used for:

[0035] The batch transfer interval of the mixed waste is obtained, and a slip distance penalty coefficient is established based on the batch transfer interval; a trigger analysis of the slip distance prediction result is performed using the slip distance penalty coefficient, and compensation for the centralized evaluation of the separation effect is performed based on the trigger analysis result.

[0036] In the embodiment of the present application, the batch transmission interval of the mixed waste is first obtained, that is, the time interval or space interval between two batches of waste. The batch transmission interval is obtained by a sensor on the conveyor belt, for example, using a photoelectric sensor or an infrared sensor to detect the gap between batches.

[0037] Based on the batch transfer interval, a slip distance penalty coefficient is established. The slip distance penalty coefficient is used to adjust the prediction results during the slip distance prediction process to prevent individual waste from sliding too far. Slipping too far will not only affect the separation effect of the current batch, but may also interfere with the transmission and processing of the next batch of waste. When calculating the slip distance penalty coefficient, the penalty coefficient is obtained by dividing the batch transfer interval by 1. The slip distance penalty coefficient is then used to trigger analysis of the slip distance prediction results. Trigger analysis refers to adjusting the slip distance prediction results according to the penalty coefficient to prevent individual waste from sliding too far. For example, if the slip distance prediction result is 10 cm and the penalty coefficient is 0.2, the adjusted slip distance prediction result = original prediction distance × (1-penalty coefficient) = 10 × (1-0.2) = 8 cm. Based on the trigger analysis results, compensation for centralized evaluation of separation effect is performed.

[0038] The identification unit 13 is used to activate the linkage sensor group to perform individual identification of the transmitted waste and establish an individual data set when the transmitted mixed waste reaches the preset area. The linkage sensor includes a visual sensor, a near-infrared spectrum sensor, an X-ray sensor, and a second weight sensor.

[0039] In the embodiment of the present application, when the mixed waste is conveyed to the preset area by the accelerating conveyor belt, the identification unit starts to work. The preset area is a specific location where various sensors are provided for detecting and identifying the characteristics of the waste. In the preset area, the linkage sensor group is activated to perform individual identification of the waste. The linkage sensor includes a visual sensor, a near infrared spectrum sensor, an X-ray sensor, and a second weight sensor.

[0040] Vision sensors such as industrial cameras are used to capture images of the waste. Near infrared spectroscopy sensors are used to analyze the chemical composition of the waste. Near infrared spectroscopy technology identifies different types of waste by detecting the spectral characteristics of the material. X-ray sensors are used to detect the internal structure of the waste. Through X-ray imaging, the density and internal structure of the waste are identified, and certain types of waste, such as composite materials containing metals, are identified. The second weight sensor is used to accurately measure the weight of the waste.

[0041] Through the data collection of the above sensors, an individual data set of waste is established.

[0042] Furthermore, the identification unit 13 in the system provided in the embodiment of the application is also used for:

[0043] The image of the individual is collected by a visual sensor to establish an image data set, and the individual size data is established based on the image data set; the spectral characteristics of the individual are collected by a near-infrared spectral sensor to establish a spectral data set, and the individual component data is established with the spectral data set; the internal data of the individual is collected by an X-ray sensor to establish the individual internal data; the weight data of the individual is collected by a second weight sensor to establish the individual weight data; and the individual data set is established according to the individual size data, the individual component data, the individual internal data, and the individual weight data.

[0044] In the embodiment of the present application, firstly, individual image acquisition is performed by a visual sensor to obtain a high-resolution image of the waste, and image processing technology, such as edge detection and shape recognition in the OpenCV library, is used to analyze the shape and size of the waste and establish individual size data. Next, individual spectral feature acquisition is performed by a near-infrared spectral sensor. The near-infrared spectral sensor detects the spectral features of the waste and stores the data in a spectral data set. Based on the spectral data set, the chemical composition of the waste is analyzed to establish individual component data. Infrared spectroscopy technology can identify waste of different materials such as plastic, metal, and paper. Then, individual internal data acquisition is performed by an X-ray sensor. The X-ray sensor uses X-ray imaging technology to detect the internal structure of the waste, and these data are stored in an individual internal data set. By analyzing the X-ray image, the density and internal structure of the waste are identified. Finally, individual weight data acquisition is performed by a second weight sensor. Through the above process, an individual data set is established based on the acquired individual size data, individual component data, individual internal data, and individual weight data.

[0045] Furthermore, the identification unit 13 in the system provided in the embodiment of the application is also used for:

[0046] Establish identification retention features for individuals of the same type; when performing individual identification, identify the collected data of the linkage sensor group based on the identification retention features to establish an individual data set.

[0047] In the embodiment of the present application, identification retention features of similar individuals are established, and data collection and identification are performed based on these retention features when performing individual identification. First, based on the identified waste individuals, key features of a specific type of waste are extracted and saved to form an identification retention feature library. These features include visual features, spectral features, internal structure features, and weight features.

[0048] When it is necessary to identify a new waste individual, data collection and identification are performed based on the established identification retention features. First, the visual sensor, near-infrared spectrum sensor, X-ray sensor and second weight sensor are activated to start data collection for the new waste individual. After each sensor collects the image data, spectral data, internal data and weight data of the waste, these data are preprocessed. For example, edge detection is performed on the image, normalization is performed on the spectral data, filtering is performed on the X-ray image, and correction is performed on the weight data. The collected data is matched with the stored identification retention features. Based on the shape and color features of the image data, the spectral features of the spectral data, the density features of the internal data and the weight data, identification is performed through a feature matching algorithm. Specifically, the newly collected image data is compared with the shape and color features in the identification retention features to determine whether they match; the spectral features of the spectral data are compared to determine whether they match; the internal structure features of the X-ray data are compared to determine whether they match; and finally, the weight data is compared to determine whether they match. If all features match, the new waste individual is identified as being consistent with a certain type of waste in the identification retention feature library.

[0049] After identification is completed, the data collected by each sensor is integrated to establish a complete individual data set.

[0050] The classification unit 14 is used to perform mixed separation evaluation based on the individual data sets, and control the pneumatic sorting system and the classification robot arm to perform linked sorting based on the mixed separation evaluation results.

[0051] In an embodiment of the present application, individual data sets are first evaluated for mixed separation. Mixed separation evaluation refers to analyzing the data of individual wastes to determine the type of each waste and the appropriate separation method. Based on image data, the shape and size of the waste are determined using image processing techniques such as edge detection and shape matching. Based on spectral data, the chemical composition of the waste, such as plastic, metal or paper, is identified by analyzing spectral feature peaks. Based on internal data, the internal structure and density of the waste are identified by X-ray imaging, which is particularly useful for composite materials and waste containing metals. Based on weight data, the weight of the waste is measured by a high-precision weighing sensor to provide more classification basis.

[0052] The results of the mixed separation evaluation are used to decide on the waste separation method. For light and fragile materials, such as plastics and paper, pneumatic sorting systems are used. Pneumatic sorting systems use airflow to blow different types of waste into different collection containers. For heavier or complex waste, such as metals and composites, a sorting robot is used. A sorting robot is a robotic arm that precisely grasps and places individual waste items through preprogrammed motion paths and a flexible end effector.

[0053] Based on the mixed separation evaluation results, the pneumatic sorting system and the classification robot arm are controlled to perform linkage sorting. Linkage sorting refers to coordinating the operations of different sorting systems to ensure that each waste individual is correctly classified. For example, when a piece of plastic waste is identified, the pneumatic sorting system will be activated immediately to blow the waste into the corresponding collection container. At the same time, if a piece of metal waste is identified, the classification robot arm will be activated to move the metal waste to the corresponding position through precise grabbing and placement actions.

[0054] Furthermore, the classification unit 14 in the system provided in the application embodiment is also used for:

[0055] The start times of pneumatic sorting and robotic arm sorting are accumulated according to the individual data sets, and a first evaluation result is established based on the accumulated start times; the influence of pneumatic sorting on the position movement of robotic arm sorting individuals is obtained to establish a second evaluation result; the influence of pneumatic sorting individuals on the grasping of robotic arm sorting individuals is obtained to establish a third evaluation result; and the mixed separation evaluation is completed based on the first evaluation result, the second evaluation result, and the third evaluation result.

[0056] In the embodiment of the present application, each waste is first classified and judged according to the individual data set, and the number of times the pneumatic sorting and the robotic arm sorting are started is recorded. The pneumatic sorting system uses airflow to blow lightweight and fragile materials, such as plastics and paper, into different collection containers, while the robotic arm sorting system accurately grabs and places heavier or complex waste, such as metals and composite materials, through preprogrammed motion paths and flexible end effectors. The number of times each pneumatic sorting and robotic arm sorting is started is recorded by sensors and controllers, and these data are stored in a database and used to establish a first evaluation result to analyze the frequency and efficiency of the sorting system.

[0057] When performing pneumatic sorting, obtain the impact of pneumatic sorting on the position movement of the robot arm sorting individuals. The airflow during pneumatic sorting may blow some waste out of the working range of the robot arm, causing the robot arm to be unable to accurately grasp it. Use position sensors, such as laser sensors or infrared sensors, to monitor the position changes of waste during pneumatic sorting, and use motion simulation software such as ANSYS Fluent to simulate the impact of airflow on the position of waste, and adjust airflow parameters to optimize the sorting effect. Based on these data, establish a second evaluation result to evaluate the impact of pneumatic sorting on the sorting efficiency of the robot arm.

[0058] At the same time, the impact of pneumatic sorting on the grasping of individual robot arm sorting is evaluated. In some cases, light waste that is not effectively separated by the pneumatic sorting system may affect the grasping operation of the robot arm. When the robot arm grasps heavier waste, unseparated light waste may be accidentally grasped or block the grasping path. Abnormal conditions encountered during the grasping process are detected by grasping sensors such as force sensors or tactile sensors installed at the end of the robot arm, and the grasping area of ​​the robot arm is monitored through visual systems such as industrial cameras and image processing software to identify unseparated light waste. Based on these data, a third evaluation result is established to evaluate the grasping impact.

[0059] The system completes the mixed separation evaluation by combining the first, second and third evaluation results. This evaluation process takes into account the mutual influence of pneumatic sorting and robotic arm sorting, and determines the optimal sorting order and strategy. If pneumatic sorting has little effect on the position of robotic arm sorting, pneumatic sorting is used first; if pneumatic sorting has a greater impact on the gripping of robotic arm sorting, the strength or order of pneumatic sorting is adjusted to optimize the overall sorting effect.

[0060] The identification unit 15 is used to identify the linkage sorting results and perform the solid waste classification control according to the identification.

[0061] In the embodiment of the present application, after the classification unit completes the mixed separation evaluation, the classification results of each waste individual will be transmitted to the identification unit. These classification results include the type of each waste and the recommended treatment method, such as using pneumatic sorting or robotic arm sorting. The identification unit 5 generates a diversion transmission identification based on these linkage sorting results, and records the classification results and transmission path of each waste individual through the electronic identification system. These electronic identifications are stored in the database management system and are associated with the position of the waste individual on the conveyor belt.

[0062] According to the generated diversion and transfer identification, each waste individual is assigned to the corresponding conveying path. The conveying path indicates the direction of movement of the waste on the conveyor belt and the final collection container. Use automatic conveyor belts and diverters to guide the waste to the correct path according to the electronic identification. Ensure that the waste is accurately guided through automatic conveyor belt systems and diverters, such as electric diverter valves or mechanical diverters.

[0063] During the waste transfer process, sensors such as photoelectric sensors and cameras monitor the movement of each waste in real time to ensure that it moves according to the diversion and transfer mark. When the waste reaches the final collection container, the classification control is completed according to the diversion and transfer mark.

[0064] Furthermore, the system provided in the application embodiment is also used for:

[0065] The authentication unit is used to identify the linkage sorting result through the identification module, establish an additional identification result, perform consistency authentication through the additional identification result and the diversion transmission identifier, and establish a consistency authentication result; the early warning unit is used to issue an abnormal diversion warning if the consistency authentication result is an inconsistent result.

[0066] In an embodiment of the present application, the authentication unit identifies the linkage sorting results through the identification module, and uses the visual sensor, near-infrared spectrum sensor, X-ray sensor and weight sensor to re-identify the waste to establish additional identification results. These additional identification results include the visual characteristics, spectral characteristics, internal structure characteristics and weight characteristics of the waste. The authentication unit compares the additional identification results with the diversion and transmission identification, and performs consistency authentication to ensure that the waste is correctly classified and transmitted. A consistency authentication result is established based on the comparison result. If the authentication result is "passed", it means that the waste is correctly classified and transmitted; if the authentication result is "failed", the early warning unit issues an abnormal diversion warning, and notifies the operator to check and correct the abnormal situation through a sound alarm or a light alarm.

[0067] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects:

[0068] The present application transmits mixed waste to a pretreatment area through a primary conveyor belt, and performs vibration screening of the mixed waste in the pretreatment area; receives the mixed waste after vibration screening based on an accelerating conveyor belt, and performs fuzzy identification of individual weights in the mixed waste through a first weight sensor built under the accelerating conveyor belt, establishes a weight fuzzy identification result, collects individual images of the mixed waste through an image sensor, establishes a volume fuzzy identification result, sends the weight fuzzy identification result and the volume fuzzy identification result to a decision layer, sets the acceleration of the accelerating conveyor belt based on the decision layer, and transmits the mixed waste after vibration screening through the accelerating conveyor belt; when the transmitted mixed waste reaches a preset area, activates a linkage sensor group to perform individual identification of the transmitted waste, and establishes an individual data set, and the linkage sensor includes a visual sensor, a near-infrared spectrum sensor, an X-ray sensor, and a second weight sensor; performs mixed separation evaluation based on the individual data set, and controls a pneumatic sorting system and a classification robot arm to perform linkage sorting based on the mixed separation evaluation result; performs a diversion transmission mark on the linkage sorting result, and completes solid waste classification control according to the diversion transmission mark. The present invention solves the technical problem of low classification accuracy in the prior art when classifying solid waste, and achieves the technical effect of accurate identification and classification of waste by integrating multiple sensing technologies such as vibration screening, weight fuzzy recognition, image acquisition, near-infrared spectroscopy, X-ray and precise weighing.

[0069] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0070] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0071] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A solid waste classification control system based on multi-sensor recognition, characterized in that: The system comprises: a pre-processing unit for conveying the mixed waste via a primary conveyor belt to a pre-processing area where vibration screening of the mixed waste is performed; A conveying and separation unit is used to receive the mixed waste after vibration screening based on an accelerating conveyor belt, and perform fuzzy identification of individual weights in the mixed waste through a first weight sensor built under the accelerating conveyor belt, establish a weight fuzzy identification result, collect individual images of the mixed waste through an image sensor, establish a volume fuzzy identification result, send the weight fuzzy identification result and the volume fuzzy identification result to a decision layer, set the acceleration of the accelerating conveyor belt based on the decision layer, and transmit the mixed waste after vibration screening through the accelerating conveyor belt; An identification unit, used for activating a linkage sensor group to identify the individual of the transmitted waste and establish an individual data set when the transmitted mixed waste reaches a preset area, wherein the linkage sensor includes a visual sensor, a near infrared spectrum sensor, an X-ray sensor, and a second weight sensor; A classification unit is used to perform mixed separation evaluation based on individual data sets, and to control the pneumatic sorting system and the classification robot arm to perform linkage sorting based on the mixed separation evaluation results; An identification unit is used to identify the linkage sorting results and to complete the solid waste classification control according to the identification of the diversion and transmission; The step of sending the weight fuzzy recognition result and the volume fuzzy recognition result to the decision layer and setting the acceleration of the conveyor belt based on the decision layer includes: Obtaining the total number of individuals of mixed waste after vibration screening through the decision layer; Activate the slip prediction network of the decision layer, input the weight fuzzy recognition result and the volume fuzzy recognition result into the slip prediction network, perform slip distance prediction under different accelerations, and establish a slip distance prediction result, wherein the slip distance prediction result carries an acceleration mapping mark; Obtain the individual initial positions of the mixed wastes, and conduct a centralized evaluation of the separation effect based on the individual initial positions and the predicted sliding distances; Generate an acceleration decision according to the centralized evaluation result of the separation effect, and set the acceleration of the acceleration conveyor belt according to the acceleration decision result; The centralized evaluation of separation effect based on individual initial positions and slip distance prediction results includes: Establish calibration distance value; The final position of the individual is established based on the individual initial position and the slip distance prediction results; Comparing the distance of the nearest neighbor individual with the calibrated distance value through the final position, if the calibrated distance value is not satisfied, assigning a linear concentration value according to the distance; If the distance of the nearest neighbor individual meets the calibration distance value, the concentration value is set to 0; Complete the centralized evaluation of separation effect based on the linear centralized values ​​and centralized values ​​of all individuals; The separation effect centralized evaluation is completed according to the linear centralized values ​​and centralized values ​​of all individuals, including: Obtaining the batch transfer interval of mixed waste, and establishing a slip distance penalty coefficient based on the batch transfer interval; The slip distance penalty coefficient is used to perform a trigger analysis of the slip distance prediction result, and compensation for the centralized evaluation of the separation effect is performed according to the trigger analysis result.

2. The solid waste classification control system based on multi-sensor recognition as claimed in claim 1, characterized in that: The activating the linkage sensor group to perform individual identification of the transported waste and establish an individual data set also includes: Collect images of individuals through a visual sensor to establish an image data set, and establish individual size data based on the image data set; Collecting individual spectral features through a near-infrared spectral sensor, establishing a spectral data set, and establishing individual component data with the spectral data set; Collect the internal data of the individual through X-ray sensors to establish the internal data of the individual; Collecting individual weight data through a second weight sensor to establish individual weight data; An individual data set is established based on individual size data, individual composition data, individual internal data, and individual weight data.

3. The solid waste classification control system based on multi-sensor recognition as claimed in claim 1, characterized in that: The mixed separation evaluation based on the individual data set also includes: Accumulate the start times of pneumatic sorting and robotic arm sorting according to the individual data sets, and establish a first evaluation result according to the accumulated start times; Obtain the influence of pneumatic sorting on the position movement of the robot arm sorting individual, and establish a second evaluation result; Obtain the influence of pneumatic sorting individuals on the grasping of robotic arm sorting individuals, and establish the third evaluation result; The mixed separation evaluation is completed based on the first evaluation result, the second evaluation result, and the third evaluation result.

4. The solid waste classification control system based on multi-sensor recognition according to claim 1, characterized in that: The system further comprises: An authentication unit, used to identify the linkage sorting result through the identification module, establish an additional identification result, perform consistency authentication through the additional identification result and the diversion transmission identifier, and establish a consistency authentication result; The early warning unit issues an abnormal diversion early warning if the consistency authentication result is an inconsistent result.

5. The solid waste classification control system based on multi-sensor recognition as claimed in claim 1, characterized in that: The activating the linkage sensor group to perform individual identification of the transported waste and establish an individual data set also includes: Establish identification and retention characteristics of similar individuals; When performing individual identification, the collected data of the linkage sensor group is identified based on the identification retention features to establish an individual data set.

Images