Electrostatic separation system, method and device based on Internet of Things and artificial intelligence

Abstract

The invention provides an electrostatic separation system, method and device based on the Internet of Things and artificial intelligence, and relates to the technical field of waste recovery and resource regeneration, the system comprises a feeding system, a receiving system, an electrified processing module, an electric field control module, an artificial intelligence optimization system and an Internet of Things system; the Internet of Things system is used for collecting feature data of the materials and operation data of the electrostatic separation system and carrying out data processing and analysis; the artificial intelligence optimization system is used for generating an optimal electric field adjustment instruction and an electrification control strategy according to the received feature data and the operation data; the electrification processing module is used for adjusting an electrification mode and an electrification parameter of the material according to an electrification control strategy; the electric field control module is used for optimizing electric field distribution according to the electric field adjusting instruction; and the material receiving system is used for receiving the materials subjected to electrostatic separation. The system response capability, the sorting precision and the energy efficiency level can be greatly improved, and intelligent optimization and self-adaptive adjustment of the electrostatic sorting process are achieved.

Description

Technical Field

[0001] This application relates to the technical field of waste recycling and resource regeneration, and particularly to an electrostatic separation system, method and device based on the Internet of Things and artificial intelligence. Background Art

[0002] With the development of the waste recycling and resource regeneration industry, electrostatic separation technology has been widely used in the treatment of complex mixed materials. However, the traditional electrostatic separation equipment in the prior art usually has the following problems to be solved urgently:

[0003] 1. Fixed power output and low energy efficiency: During the electrostatic separation process, the change of the electric field intensity needs to be precisely regulated. The traditional equipment lacks an intelligent adjustment system, resulting in unnecessary energy waste.

[0004] 2. Lack of real-time adjustment function: The electric field configuration and charging mode of the traditional equipment are usually fixed and cannot be optimized in real time according to the change of materials.

[0005] 3. Frequent equipment maintenance: The operation and maintenance of the traditional equipment usually rely on manual labor, lacking an automated and intelligent diagnosis and maintenance mechanism.

[0006] Currently, with the deep integration of AI and Internet of Things technologies in intelligent manufacturing and environmental protection equipment, the intelligent separation technology that uses AI to identify and optimize the control of electrostatic parameters and uses IoT to sense the operating state of equipment has gradually become an important research direction for improving the resource recovery rate and system efficiency. Summary of the Invention

[0007] In view of the problems in the prior art, the embodiments of this application provide an electrostatic separation system, method and device based on the Internet of Things and artificial intelligence, which can solve the problems in the prior art that the electric field output of the traditional system is fixed, the energy efficiency is low, and it cannot be adjusted in real time according to the change of materials; the charging treatment method is fixed, resulting in poor separation adaptability; and there is a lack of abnormal detection and energy efficiency optimization mechanisms, and the operation stability and energy consumption control are insufficient.

[0008] In the first aspect, this application provides an electrostatic separation system based on the Internet of Things and artificial intelligence, including: a feeding system, a material receiving system, a charging treatment module, an electric field control module, an artificial intelligence optimization system and an Internet of Things system;

[0009] The feeding system is connected to the charging treatment module and is used to evenly feed materials into the electrostatic separation system;

[0010] The Internet of Things system is connected to the artificial intelligence optimization system through network communication, and is used to collect the characteristic data of the materials and the operation data of the electrostatic separation system and perform data processing and analysis on the characteristic data and the operation data;

[0011] The artificial intelligence optimization system is used to generate optimal electric field adjustment instructions and charging control strategies according to the received characteristic data and operation data;

[0012] The charging processing module is connected to the electric field control module and connected to the artificial intelligence optimization system through a communication bus, and is used to adjust the charging mode and charging parameters of the material according to the received charging control strategy;

[0013] The electric field control module is connected to the artificial intelligence optimization system through a communication bus, and is used to optimize the electric field distribution according to the received electric field adjustment instructions;

[0014] The material receiving system is connected to the electric field control module and is used to receive the material after electrostatic separation.

[0015] Further, the Internet of Things system includes a sensor module, an Internet of Things communication module, and a cloud platform;

[0016] The sensor module is used to collect the characteristic data of the material and the operation data of the electrostatic separation system;

[0017] The Internet of Things communication module is used to receive the characteristic data and the operation data and send them to the artificial intelligence optimization system and the cloud platform;

[0018] The cloud platform is used to perform data processing and analysis on the received characteristic data and operation data.

[0019] Further, the artificial intelligence optimization system is specifically used for:

[0020] Receive the characteristic data from the Internet of Things system;

[0021] Determine the charging mode according to the characteristic data and a pre-trained decision tree model;

[0022] Generate charging parameters according to the characteristic data and a pre-trained reinforcement learning model corresponding to the charging mode;

[0023] Generate a charging control strategy using the charging mode and the charging parameters and send it to the charging processing module.

[0024] Further, the artificial intelligence optimization system is specifically used for:

[0025] Receive the characteristic data and the operation data from the Internet of Things system;

[0026] Perform simulation modeling on the electric field of the electrostatic separation system according to the operation data to obtain a numerical model of the electric field distribution;

[0027] Generate the electric field distribution state according to the numerical model of the electric field distribution;

[0028] Generate the electric field distribution control parameters according to the characteristic data, the electric field distribution state, and the reinforcement learning agent that pre-constructs and calls the numerical model of the electric field distribution;

[0029] Generate an electric field adjustment instruction by using the electric field distribution control parameters and send it to the electric field control module.

[0030] Further, it further includes: a safety intelligent management system;

[0031] The safety intelligent management system is connected to the Internet of Things system and the artificial intelligence optimization system, and is used to identify the abnormal state and fault trend of the electrostatic separation system according to the received operation data, and send the identification result to the artificial intelligence optimization system to adjust the operation strategy.

[0032] Further, the safety intelligent management system is specifically used for:

[0033] Receive the operation data from the Internet of Things system;

[0034] Identify the abnormal state of the electrostatic separation system according to the operation data and the pre-trained anomaly detection model; the anomaly detection model is obtained by unsupervised training of the historical operation data of the electrostatic separation system in the normal operation state;

[0035] Identify the fault trend of the electrostatic separation system according to the operation data and the pre-trained fault prediction model; the fault prediction model is obtained by supervised training of the historical operation data and fault records of the electrostatic separation system in different operation cycles;

[0036] Send the identification result to the artificial intelligence optimization system to adjust the operation strategy.

[0037] Further, it further includes: an energy efficiency optimization system;

[0038] The energy efficiency optimization system is connected to the Internet of Things system and the electric field control module, and is used to generate an optimization control strategy according to the received operation data to perform adaptive optimization adjustment on the electric field power.

[0039] Further, the energy efficiency optimization system is specifically used for:

[0040] Receive the operation data from the Internet of Things system;

[0041] Generate an optimization control strategy through dynamic analysis according to the operation data;

[0042] Send the optimized control strategy to the electric field control module to adaptively optimize and adjust the electric field power.

[0043] In a second aspect, the present application provides an electrostatic separation method based on the Internet of Things and artificial intelligence, which is applied to the electrostatic separation system based on the Internet of Things and artificial intelligence described in any of the above embodiments, and includes:

[0044] Receive the feature data from the Internet of Things system;

[0045] Determine the charging mode according to the feature data and the pre-trained decision tree model;

[0046] Generate charging parameters according to the feature data and the pre-trained reinforcement learning model corresponding to the charging mode;

[0047] Generate a charging control strategy using the charging mode and the charging parameters and send it to the charging processing module.

[0048] Further, it further includes:

[0049] Receive the feature data and the operation data from the Internet of Things system;

[0050] Perform simulation modeling on the electric field of the electrostatic separation system according to the operation data to obtain an electric field distribution numerical model;

[0051] Generate an electric field distribution state according to the electric field distribution numerical model;

[0052] Generate electric field distribution control parameters according to the feature data, the electric field distribution state, and a pre-constructed reinforcement learning agent that calls the electric field distribution numerical model;

[0053] Generate an electric field adjustment instruction using the electric field distribution control parameters and send it to the electric field control module.

[0054] In a third aspect, the present application provides an electrostatic separation device based on the Internet of Things and artificial intelligence, which is applied to the electrostatic separation system based on the Internet of Things and artificial intelligence described in any of the above embodiments, and includes:

[0055] A first data receiving unit for receiving the feature data from the Internet of Things system;

[0056] A charging mode determination unit for determining the charging mode according to the feature data and the pre-trained decision tree model;

[0057] A charging parameter generation unit for generating charging parameters according to the feature data and the pre-trained reinforcement learning model corresponding to the charging mode;

[0058] The charged control strategy output unit is configured to generate a charged control strategy by using the charged mode and the charged parameters and send it to the charged processing module.

[0059] Further, it further includes:

[0060] The second data receiving unit is configured to receive the feature data and the operation data from the IoT system;

[0061] The model establishment unit is configured to perform simulation modeling on the electric field of the electrostatic separation system according to the operation data to obtain a numerical model of the electric field distribution;

[0062] The electric field distribution state generation unit is configured to generate an electric field distribution state according to the numerical model of the electric field distribution;

[0063] The control parameter generation unit is configured to generate electric field distribution control parameters according to the feature data, the electric field distribution state, and a reinforcement learning agent that pre - constructs and calls the numerical model of the electric field distribution;

[0064] The electric field adjustment instruction output unit is configured to generate an electric field adjustment instruction by using the electric field distribution control parameters and send it to the electric field control module.

[0065] In a fourth aspect, the present application provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it implements the electrostatic separation method based on the Internet of Things and artificial intelligence described in any of the above embodiments.

[0066] In a fifth aspect, the present application provides a computer - readable storage medium. The computer - readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the electrostatic separation method based on the Internet of Things and artificial intelligence described in any of the above embodiments.

[0067] In a sixth aspect, the present application provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, it implements the electrostatic separation method based on the Internet of Things and artificial intelligence described in any of the above embodiments.

[0068] The present application provides an electrostatic separation system, method and device based on the Internet of Things and artificial intelligence. The system includes: a feeding system, a material receiving system, a charging processing module, an electric field control module, an artificial intelligence optimization system and an Internet of Things system; the feeding system is connected to the charging processing module and is used for uniformly feeding materials into the electrostatic separation system; the Internet of Things system is connected to the artificial intelligence optimization system through network communication, and is used for collecting the characteristic data of the materials and the operation data of the electrostatic separation system and performing data processing and analysis on the characteristic data and the operation data; the artificial intelligence optimization system is used for generating optimal electric field adjustment instructions and charging control strategies according to the received characteristic data and operation data; the charging processing module is connected to the electric field control module and is connected to the artificial intelligence optimization system through a communication bus, and is used for adjusting the charging mode and charging parameters of the materials according to the received charging control strategy; the electric field control module is connected to the artificial intelligence optimization system through a communication bus, and is used for optimizing the electric field distribution according to the received electric field adjustment instructions; the material receiving system is connected to the electric field control module and is used for receiving the materials after electrostatic separation. The present application realizes the intelligent optimization and adaptive adjustment of electrostatic separation, and improves the separation accuracy and system stability.

[0069] Among them, the feeding system realizes the uniform transportation of materials, ensuring the stability and separation accuracy of the subsequent charging and separation processes; the Internet of Things system realizes the real-time perception and data collection of material characteristics and equipment operation states, providing data support for intelligent optimization; the artificial intelligence optimization system realizes the adaptive optimization based on material characteristics and equipment states, automatically generating control strategies for electric fields and charging parameters, and improving the intelligent level and energy efficiency of the equipment; the charging processing module realizes the intelligent selection and adjustment of charging modes and charging parameters for different materials, improving the charging uniformity of materials and the subsequent separation efficiency; the electric field control module realizes the dynamic optimization of electrode voltage, electric field strength and distribution, ensuring that the electric field adapts to the material state and reducing energy consumption; the material receiving system realizes the efficient classification and collection of the separated materials, ensuring the continuity and practicality of the entire electrostatic separation process.

[0070] By introducing the real-time perception of the Internet of Things and the dynamic optimization control of artificial intelligence, the present invention realizes the intelligent optimization and adaptive adjustment of the electrostatic separation process, significantly improving the separation accuracy, energy efficiency and system stability. Different from the traditional electrostatic separation system that relies on manual adjustment or single-sensor control methods, the present invention constructs an intelligent closed-loop optimization system of "perception - decision - execution - feedback", improving the real-time response ability of the system and the ability to adapt to diverse material processing, and having important engineering application value and promotion potential. Brief Description of the Drawings

[0071] To more clearly illustrate the technical solutions in the embodiments of the present application or in the prior art, the following will briefly introduce the accompanying drawings required for use in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0072] Figure 1 It is a schematic structural diagram of an electrostatic separation system based on the Internet of Things and artificial intelligence provided by an embodiment of the present application;

[0073] Figure 2 It is a schematic structural diagram of an electrostatic separation system based on the Internet of Things and artificial intelligence provided by an embodiment of the present application;

[0074] Figure 3 It is a schematic structural diagram of an electrostatic separation system based on the Internet of Things and artificial intelligence provided by an embodiment of the present application;

[0075] Figure 4 It is a schematic structural diagram of an electrostatic separation system based on the Internet of Things and artificial intelligence provided by an embodiment of the present application;

[0076] Figure 5 It is a schematic flowchart of an electrostatic separation method based on the Internet of Things and artificial intelligence provided by an embodiment of the present application;

[0077] Figure 6 It is a schematic flowchart of an electrostatic separation method based on the Internet of Things and artificial intelligence provided by an embodiment of the present application;

[0078] Figure 7 It is a schematic structural diagram of an electrostatic separation device based on the Internet of Things and artificial intelligence provided by an embodiment of the present application;

[0079] Figure 8 It is a schematic structural diagram of an electrostatic separation device based on the Internet of Things and artificial intelligence provided by an embodiment of the present application;

[0080] Figure 9 It is a schematic block diagram of the system composition of an electronic device. Detailed implementation manners

[0081] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer and more understandable, the following will further elaborate on the embodiments of the present application in conjunction with the accompanying drawings. Herein, the illustrative embodiments of the present application and their descriptions are used to explain the present application, but do not limit the present application. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be arbitrarily combined with each other.

[0082] Figure 1The following is a schematic structural diagram of an electrostatic separation system based on the Internet of Things and artificial intelligence provided by an embodiment of the present application. As Figure 1 shown, the electrostatic separation system based on the Internet of Things and artificial intelligence provided by the present application includes: a feeding system 101, a material collection system 104, a charging processing module 102, an electric field control module 103, an artificial intelligence optimization system 106, and an Internet of Things system 105;

[0083] The feeding system 101 is connected to the charging processing module 102 and is used to uniformly feed materials into the electrostatic separation system;

[0084] The Internet of Things system 105 is connected to the artificial intelligence optimization system 106 through network communication, and is used to collect the characteristic data of the materials and the operation data of the electrostatic separation system, and perform data processing and analysis on the characteristic data and the operation data;

[0085] The artificial intelligence optimization system 106 is used to generate optimal electric field adjustment instructions and charging control strategies according to the received characteristic data and operation data;

[0086] The charging processing module 102 is connected to the electric field control module 103 and is connected to the artificial intelligence optimization system 106 through a communication bus, and is used to adjust the charging mode and charging parameters of the materials according to the received charging control strategy;

[0087] The electric field control module 103 is connected to the artificial intelligence optimization system 106 through a communication bus, and is used to optimize the electric field distribution according to the received electric field adjustment instructions;

[0088] The material collection system 104 is connected to the electric field control module 103 and is used to receive the materials after electrostatic separation.

[0089] Specifically, the feeding system 101 uniformly feeds the mixed materials to be separated into the interior of the electrostatic separation system, ensuring the stability of the subsequent charging processing and electric field separation processes, and controlling the feeding speed and uniformity.

[0090] The charging processing module 102 is arranged between the feeding system 101 and the electric field control module 103, and is used to pre-charge the materials by means of contact charging, friction charging, corona discharge, etc. according to different material characteristics, ensuring the separation efficiency and accuracy. This module can dynamically adjust the charging mode and charging parameters according to the received control strategy.

[0091] The electric field control module 103 receives the electric field adjustment instructions from the artificial intelligence optimization system 106, and adjusts the voltage, electric field strength or polarity configuration between the electrodes to optimize the electric field distribution, improve the separation effect and reduce unnecessary energy consumption.

[0092] The artificial intelligence optimization system 106 receives material characteristic data and equipment operation status information by communicating with the Internet of Things system 105, models and optimizes the data using algorithms such as machine learning and deep reinforcement learning, automatically generates charged control strategies and electric field adjustment instructions, and realizes adaptive control.

[0093] The Internet of Things system 105 collects operation data such as material particle size, resistivity, humidity, electric field load, and power consumption, and transmits the data to the artificial intelligence optimization system 106 in real time for processing.

[0094] The material receiving system 104 is arranged downstream of the electric field control module 103, and recovers materials with different charge characteristics according to the electrostatic separation results to achieve physical separation.

[0095] The above-mentioned modules are connected through a communication bus or a physical interface. After the data is collected by the Internet of Things system 105, it is sent to the artificial intelligence optimization system 106, which automatically calculates and issues control strategies to the charged processing module 102 and the electric field control module 103 according to the data, realizing closed-loop optimization control of the electrostatic separation process.

[0096] In one embodiment, the material receiving system 104 can be configured with multiple receiving bins according to the sorting channels.

[0097] In one embodiment, the feeding system 101 transports the material to the charged processing module 102 at a uniform dispersion and stable rate to ensure uniform subsequent charged processing effects and avoid uneven charging or sorting failure caused by accumulation and overload; after receiving the charged control strategy issued by the artificial intelligence optimization system 106, the charged processing module 102 performs contact charging, friction charging, or corona discharge on the material to make the material carry suitable charges for sorting, and the charge amounts and polarities of different particles are different; subsequently, the charged material is sent into the electric field control module 103, and the electric field control module 103 dynamically adjusts the electrode voltage, spacing, and polarity switching frequency according to the electric field adjustment instruction issued by the artificial intelligence optimization system 106 to generate an optimized electrostatic separation electric field, dynamically adjust the deflection trajectory of the charged material in the electric field, and effectively sort the material based on charge characteristics; finally, materials with different charged characteristics enter the material receiving system 104 for classification and collection according to the sorting trajectory.

[0098] The charged processing module 102 selects working modes according to the received charged control strategies, including a contact charging unit, a friction charging unit, and a corona discharge unit. The contact charging unit is used to charge the material through mechanical friction; the friction charging unit is used to charge the material by contacting with a high-speed rotating friction roller; the corona discharge unit is used to ionize and charge the surface of the material through high-voltage corona discharge.

[0099] The electric field control module 103 includes parallel electrode plates, which can apply a preset voltage according to the received electric field adjustment instruction to form a spatial electric field, causing particles with different charges and polarities to deviate from the transmission trajectory. Among them, the distance between the electrode plates is adjustable to adapt to the sorting requirements of materials with different particle sizes. The electrode voltage and polarity switching frequency can be dynamically adjusted in real time according to the material characteristics and charging status to optimize the material sorting trajectory and improve the sorting accuracy. Different charge amounts result in different deviation trajectories, and the deviation distance can be obtained based on the material mass, material charge amount, electric field strength, and electric field action time.

[0100] The material receiving system 104 includes a plurality of guiding structures and a plurality of receiving bins connected thereto. According to the charged polarity and deviation trajectory of the sorted materials, the corresponding guiding structures guide the corresponding types of materials into the corresponding receiving bins respectively. When the deviation trajectory of the materials changes, the positions of the receiving bins and the guiding structures need to be adjusted to accurately receive the corresponding materials. Among them, the guiding structure can be a fixed baffle, a dynamically switched baffle, or a rail system, and the number of receiving bins can be configured into at least two groups or multiple groups according to the material types. The guiding mechanism can be a structure such as a pipeline or a flat plate, and the present application is not limited thereto.

[0101] Figure 2 It is a schematic structural diagram of an electrostatic sorting system based on the Internet of Things and artificial intelligence provided by an embodiment of the present application. As Figure 2 shown, the Internet of Things system 105 includes a sensor module 201, an Internet of Things communication module 202, and a cloud platform 203;

[0102] The sensor module 201 is used to collect the characteristic data of the material and the operation data of the electrostatic sorting system;

[0103] The Internet of Things communication module 202 is used to receive the characteristic data and the operation data and send them to the artificial intelligence optimization system 106 and the cloud platform 203;

[0104] The cloud platform 203 is used to perform data processing and analysis on the received characteristic data and operation data.

[0105] Specifically, the Internet of Things system 105 further includes: a sensor module 201, an Internet of Things communication module 202, and a cloud platform 203 to achieve comprehensive perception of the system operation status, efficient data transmission, and centralized processing.

[0106] The sensor module 201 collects the characteristic data of the material and the operation data of the electrostatic separation system in real time. The material characteristic data includes but is not limited to material type, charge requirement, particle size, humidity, resistivity, dielectric constant, etc.; the operation data includes but is not limited to key equipment parameters such as electric field voltage, electrode spacing, power consumption, discharge frequency, electric field strength, electric field uniformity, electrode state, current of the charging system, operating temperature, vibration spectrum, etc. The sensor module 201 includes an electric field strength sensor, a temperature and humidity sensor, a current / voltage sensor, a Hall sensor, a vibration sensor, a pressure sensor, an electrostatic field measurement array, a dielectric constant measuring instrument, a surface charge detector, a material flow sensor, a leakage current detector, etc., and selects appropriate installation points according to the equipment layout.

[0107] The Internet of Things communication module 202 transmits the above-collected data to the artificial intelligence optimization system 106 and the cloud platform 203 wirelessly. The Internet of Things communication module 202 supports a two-way interaction mechanism, which can not only upload the collected data but also receive the regulation commands returned by the cloud platform 203.

[0108] The cloud platform 203 centrally processes and analyzes the received characteristic data and operation data, and generates a health status report of the system. The cloud platform 203 integrates big data analysis tools and edge computing interfaces, and is responsible for functions such as data storage, status visualization, operation log generation, and historical trend prediction.

[0109] Through the collaborative operation of the above Internet of Things system 105, not only the system's all-round perception of the material state and equipment operation is realized, but also the data interaction efficiency between the optimization system and the sorting site is guaranteed, laying a foundation for subsequent intelligent analysis and precise control.

[0110] In one embodiment, the communication methods of the Internet of Things communication module 202 may include Wi-Fi, 4G / 5G, LoRa, or NB-IoT, etc., to ensure the real-time and stability of data transmission.

[0111] In one embodiment, the cloud platform 203 timely warns of possible equipment failures by analyzing equipment data and provides maintenance suggestions.

[0112] In one embodiment, the cloud platform 203 can also serve as a unified management center for multiple sorting devices, supporting multi-device parallel collaboration and remote scheduling.

[0113] In one embodiment, the artificial intelligence optimization system 106 is specifically used for:

[0114] Receiving the characteristic data from the Internet of Things system 105;

[0115] Determining the charging mode according to the characteristic data and the pre-trained decision tree model;

[0116] Generate charged parameters based on the feature data and the pre-trained reinforcement learning model corresponding to the charged mode;

[0117] Generate a charged control strategy using the charged mode and the charged parameters and send it to the charged processing module 102.

[0118] Specifically, the artificial intelligence optimization system 106 receives the feature data of the materials from the Internet of Things system 105. The feature data includes physical and electrical characteristics such as the particle size, resistivity, humidity, and dielectric constant of the materials, which are collected in real time by the sensor module 201 and uploaded to the artificial intelligence optimization system 106 for use as the input basis for subsequent charged strategy decisions.

[0119] The artificial intelligence optimization system 106 has a built-in charged mode classification model, preferably a hybrid model combining decision trees (such as XGBoost, Random Forest, etc.) and deep learning, to establish the optimal matching rules for different types of materials in terms of charging methods based on historical data. The artificial intelligence optimization system 106 automatically discriminates the optimal charging method according to the currently collected feature data, improving adaptability and sorting efficiency.

[0120] For the determined charged mode, the artificial intelligence optimization system 106 calls the corresponding deep reinforcement learning model, combines the current material characteristics and the target sorting accuracy, and outputs a set of optimized charged parameters. Under different charged modes, the adjusted parameters are different. Finally, the artificial intelligence optimization system 106 encapsulates the generated charged mode and charged parameters into a complete charged control strategy and sends it to the charged processing module 102 for execution, realizing the intelligent closed-loop control of the charging method and operating parameters. The artificial intelligence optimization system 106 can also perform online fine-tuning and self-learning based on the feedback data to continuously optimize the control strategy.

[0121] In one embodiment, the steps of pre-training the decision tree model include:

[0122] The artificial intelligence optimization system 106 conducts multiple batches of experiments by combining different types of mixed waste (such as plastics, metals, rubbers, minerals, etc.) during the actual operation process, collects the corresponding material feature data, including but not limited to: particle size, humidity, resistivity, dielectric constant, material composition labels, etc. At the same time, record the sorting effect indicators under different charged modes, such as charging uniformity, sorting rate, and residue rate, etc.

[0123] Taking the above characteristic data as the input vector and combining the "optimal charging mode" determined by manual annotation or expert system as the output label, a training sample set is constructed. The charging mode labels include: contact charging, triboelectric charging, and corona discharge. Preferably, an interpretable integrated decision tree model such as XGBoost (Extreme Gradient Boosting Tree) or RandomForest (Random Forest) is used. Through the supervised learning training method, the model structure is continuously iterated in the training set so that it can output the probability distribution or classification result of the optimal charging mode after inputting the material characteristics.

[0124] The trained model is evaluated in multiple dimensions such as accuracy, recall rate, and F1 value in an independent test set. Generalization ability can be improved through means such as cross-validation and hyperparameter adjustment to ensure the stable prediction performance of the model on new materials.

[0125] In one embodiment, plastic materials are suitable for triboelectric charging, metal materials are suitable for contact charging, and high-humidity materials preferably use the corona discharge mode.

[0126] In one embodiment, in the contact charging mode, the adjusted parameters are friction time, pressure, and friction material type; in the triboelectric charging mode, the adjusted parameters are friction speed, material flow rate, and triboelectric surface resistance; in the corona discharge mode, the adjusted parameters are electrode voltage, electrode spacing, corona frequency, etc. The charging parameter optimization strategy constructed by reinforcement learning can dynamically balance charging efficiency and energy consumption, significantly improving the operating efficiency of the system. Set a reward function to give positive or negative feedback on parameter adjustment according to material charging uniformity, energy consumption, and sorting accuracy, so that the artificial intelligence optimization system 106 continuously optimizes the adjustment strategy.

[0127] In one embodiment, the steps of pre-training the reinforcement learning model include:

[0128] After determining the charging mode, the artificial intelligence optimization system 106 needs to further determine the corresponding combination of charging parameters (such as voltage, contact pressure, rotation speed, electrode spacing, etc.) to meet the following objectives: maximizing charging efficiency and uniformity; minimizing energy consumption; ensuring stable operation and avoiding electric breakdown or charge imbalance.

[0129] Using a deep neural network (DQN (Deep Q-Network) or PPO (Proximal Policy Optimization)), the model input is material characteristic data; the model output is the optimal combination of charging parameters; the model training objective is to minimize the charging deviation between the predicted value and the historical optimal sample, while meeting the power consumption threshold and safety constraints.

[0130] In one embodiment, since indicators such as material humidity and charge retention fluctuate dynamically over time, the artificial intelligence optimization system 106 introduces a time series prediction module based on a long short-term memory network (LSTM) to perform short-term predictions on the material charge state and the sorting efficiency trend. The artificial intelligence optimization system 106 evaluates the deviation between the current charging parameters and the actual charging effect based on the feedback data, and when it detects a decrease in efficiency or a charging deviation, it performs rapid parameter fine-tuning to achieve closed-loop adjustment and adaptive correction.

[0131] The artificial intelligence optimization system 106 predicts the charge change trend of the material through LSTM (long short-term memory network) to achieve pre-adjustment and prevent the sorting efficiency from decreasing due to environmental changes (such as increased humidity and charge decay).

[0132] In one embodiment, the artificial intelligence optimization system 106 is specifically configured to:

[0133] Receive the feature data and the operation data from the Internet of Things system 105;

[0134] Perform simulation modeling on the electric field of the electrostatic sorting system according to the operation data to obtain a numerical model of the electric field distribution;

[0135] Generate an electric field distribution state according to the numerical model of the electric field distribution;

[0136] Generate electric field distribution control parameters according to the feature data, the electric field distribution state, and a reinforcement learning agent that pre-constructs and calls the numerical model of the electric field distribution;

[0137] Generate an electric field adjustment instruction using the electric field distribution control parameters and send it to the electric field control module 103.

[0138] Specifically, the artificial intelligence optimization system 106 receives the material feature data and the system operation data from the Internet of Things system 105, including: electric field strength, voltage, electrode spacing, electrode temperature, electric field volatility, charge residue, material type, resistivity, humidity, and charge state feedback, etc. The above data is collected in real time by the sensor module 201 and uploaded to the artificial intelligence optimization system 106 through the Internet of Things communication module 202 for unified processing.

[0139] The artificial intelligence optimization system 106 performs electrostatic field simulation on the electrode structure and voltage configuration in the electrostatic separation area through the finite element method, and constructs a high-precision numerical model of the electric field distribution. This model can reflect spatial characteristics such as potential distribution, electric field line direction, and field strength gradient under different input parameters. The simulation model is fitted and corrected according to historical operation data and measured electric field sensor data to ensure a high degree of consistency between the numerical model and the physical system. During operation, the artificial intelligence optimization system 106 calls this model to generate the electric field distribution state under the current configuration (including uniformity index and hot spot area identification, etc.), which is used as the input reference for the reinforcement learning agent.

[0140] A reinforcement learning agent is constructed in the artificial intelligence optimization system 106. Its role is to continuously learn and optimize the electric field control strategy in the electrostatic separation system with the support of the established numerical model of the electric field distribution. The system inputs the state vector in real time during operation. The reinforcement learning agent outputs the electric field distribution control parameters according to the policy network, including electrode voltage, electrode spacing, and electrode shape, etc., which are used to generate the final electric field adjustment instruction and send it to the electric field control module 103 to make the electric field distribution tend to be optimal.

[0141] Finally, the artificial intelligence optimization system 106 packages the generated electric field distribution control parameters into an electric field adjustment instruction and sends it to the electric field control module 103 for execution, realizing the intelligent closed-loop control of the electric field distribution control parameters. The artificial intelligence optimization system 106 can also perform online fine-tuning and self-learning according to the feedback data to continuously optimize the control strategy.

[0142] By introducing the reinforcement learning agent, the artificial intelligence optimization system 106 can not only achieve dynamic self-adaptation of the electric field configuration under complex materials and operating conditions, but also continuously learn the feedback in the separation process, improve decision-making flexibility, optimize energy efficiency control, and have a high degree of environmental adaptability and expansion ability.

[0143] In one embodiment, the reinforcement learning agent includes: an environment interface, a state space, an action space, a policy network and a learning algorithm, a reward function design, and a policy output and online call mechanism.

[0144] An integrated system of a reinforcement learning agent and a numerical model of electric field distribution, where the numerical model of electric field distribution reflects the electrostatic field response under different electrode configurations through finite element simulation (FEM). This simulation environment provides state feedback and reward signals, constituting the training environment for reinforcement learning. The states received by the reinforcement learning agent during each decision-making include the current electric field distribution state and the current material properties. The adjustable parameters of the reinforcement learning agent include electrode voltage, electrode spacing, electrode polarity switching strategy, discharge frequency, or trigger period, etc. The reinforcement learning agent is constructed using the Deep Deterministic Policy Gradient (DDPG) or PPO (Proximal Policy Optimization) algorithm, and its policy network is used to map states to actions, and the control strategy is continuously optimized through iterative training during the training process. The reward function is designed as a multi-objective function to guide the reinforcement learning agent to generate optimized electric field configuration parameters. Specific indicators include: improvement in electric field uniformity (positive reward); improvement in electrostatic separation efficiency (positive reward); reduction in system energy consumption (positive reward); electric field instability, excessive energy consumption, local overload, or breakdown (negative penalty).

[0145] In one embodiment, the artificial intelligence optimization system 106 identifies uneven regions in the electric field and interference factors that may affect the separation effect (such as electrode loss, environmental humidity impact, etc.) through potential distribution and field strength gradient analysis.

[0146] In one embodiment, the artificial intelligence optimization system 106 uses an RNN (Recurrent Neural Network) or LSTM (Long Short-Term Memory) to predict the short-term change trend of the electric field and adjusts the electrode parameters in advance to reduce sudden abnormal situations during the separation process.

[0147] In one embodiment, in an industrial operation environment, the artificial intelligence optimization system 106 adjusts the voltage source, grounding configuration, and electrode arrangement in real time to achieve adaptive electric field optimization and improve the separation stability. Combining edge computing, the AI model is deployed at the device end to reduce latency and improve the optimization response speed.

[0148] Figure 3 It is a schematic structural diagram of an electrostatic separation system based on the Internet of Things and artificial intelligence provided by an embodiment of the present application. As Figure 3 shown, the electrostatic separation system based on the Internet of Things and artificial intelligence provided by the present application further includes: a safety intelligent management system 301;

[0149] The safety intelligent management system 301 is connected to the Internet of Things system 105 and the artificial intelligence optimization system 106, and is used to identify the abnormal state and fault trend of the electrostatic separation system according to the received operation data, and send the identification result to the artificial intelligence optimization system 106 to adjust the operation strategy.

[0150] Specifically, the safety intelligent management system 301 ensures the safety, stability and maintainability of the system in the high-voltage electrostatic separation environment. The safety intelligent management system 301 is respectively connected to the Internet of Things system 105 and the artificial intelligence optimization system 106, and can identify potential abnormal states or fault trends according to the key equipment status data collected during the operation process, and feedback the processing result to the artificial intelligence optimization system 106 to dynamically adjust the operation strategy, so as to achieve the safety intelligent control goal of "running, monitoring and optimizing at the same time".

[0151] In one embodiment, the electrostatic separation system is internally integrated with overload protection, leakage protection and short-circuit protection to ensure the safety of the electrical system. When an abnormality occurs in the electrostatic separation system, the power supply is automatically cut off, and the operator is notified remotely through the Internet of Things system 105.

[0152] In one embodiment, the safety intelligent management system 301 is specifically used for:

[0153] Receiving the operation data from the Internet of Things system 105;

[0154] Identifying the abnormal state of the electrostatic separation system according to the operation data and a pre-trained anomaly detection model; the anomaly detection model is obtained by unsupervised training of the historical operation data of the electrostatic separation system in the normal operation state;

[0155] Identifying the fault trend of the electrostatic separation system according to the operation data and a pre-trained fault prediction model; the fault prediction model is obtained by supervised training of the historical operation data and fault records of the electrostatic separation system in different operation cycles;

[0156] Sending the identification result to the artificial intelligence optimization system 106 to adjust the operation strategy.

[0157] Specifically, the safety intelligent management system 301 obtains operation data from multiple sensors through the Internet of Things system 105, including but not limited to: operation environment parameters such as current, voltage, electrode temperature, electric field uniformity and charge change, vibration, noise and discharge frequency.

[0158] Based on the received real-time data, the safety intelligent management system 301 uses a pre-trained anomaly detection model to conduct real-time evaluation of the device status. The anomaly detection model adopts an unsupervised learning method and is trained with a large amount of historical operation data under normal conditions to construct the "behavior boundary" of the normal state, so as to accurately identify abnormal patterns beyond the threshold during operation, such as abnormal fluctuations in electrode current, sudden temperature rise, uneven electric field or polarity drift, etc.

[0159] The safety intelligent management system 301 further introduces a supervised-trained fault prediction model, which combines historical fault labels and operation cycle data to evaluate whether there is a potential failure risk in the current state. The prediction is carried out by analyzing the following trend indicators: the decrease in charging efficiency caused by electrode aging, the electric field imbalance caused by unstable high-voltage power supply, and the electrode misalignment caused by long-term vibration, etc.

[0160] Once an anomaly is identified or a high-risk trend is predicted, the safety intelligent management system 301 sends the relevant results to the artificial intelligence optimization system 106, which adjusts the operation strategy to avoid the impact of faults.

[0161] In one embodiment, the anomaly detection model adopts Isolation Forest or anomaly sequence detection based on LSTM. The LSTM algorithm learns the normal operation mode of the device and predicts the parameter change trend in a short period. The IsolationForest algorithm trains the model to identify abnormal characteristics of the device operation state: short circuit, leakage, and electric field imbalance caused by electrode aging. When an abnormal point is detected, the safety protection mechanism is triggered and a detailed anomaly log is generated.

[0162] In one embodiment, the fault prediction model is preferably trained using algorithms such as Random Forest, LSTM, or ARIMA, and can predict the potential future fault time points or failed components. This includes: judging whether the current / voltage exceeds the standard and predicting the short-circuit risk in advance; generating an abnormal temperature curve to detect possible high-temperature burnout faults; identifying abnormal loads in the charging system and warning of possible electrode damage. The random forest classification algorithm is used to classify faults: electrode faults: abnormal voltage fluctuations, decreased charging efficiency; unstable corona discharge: abnormal electric field distribution, wrong charging mode; too high temperature: heat dissipation system failure.

[0163] In one embodiment, the strategies for dealing with anomalies or faults include but are not limited to: reducing voltage, switching to a polarity operation strategy, starting a self-check or cleaning process, switching to an energy-saving or low-load operation mode, and remotely notifying the operation and maintenance personnel.

[0164] In one embodiment, when an anomaly occurs in the device, the Safety Intelligent Management System 301 provides a detailed fault report and remotely transmits it via the Internet of Things system 105 to inform the operator in advance for maintenance. Based on the device's operating status and historical data, the maintenance cycle of the device is evaluated, and regular maintenance suggestions and automatic maintenance notifications are provided.

[0165] In one embodiment, the Safety Intelligent Management System 301 uses Wavelet Transform for data denoising and extracts precursor fault features: abnormal current fluctuations, abnormal electrode temperatures, etc.

[0166] In one embodiment, the Safety Intelligent Management System 301 calculates the safety and health score of the system based on historical data and real-time monitoring results: 80 - 100 Safe: The device is in normal condition and no intervention is required; 50 - 80 Warning: There are minor anomalies, and it is recommended to check specific components (electrode looseness, current fluctuations); 0 - 50 High Risk: The device may experience a serious fault, automatically cut off the power supply and trigger an alarm.

[0167] Figure 4 It is a schematic structural diagram of an electrostatic separation system based on the Internet of Things and artificial intelligence provided by an embodiment of the present application. As Figure 4 shown, the electrostatic separation system based on the Internet of Things and artificial intelligence provided by the present application further includes: an Energy Efficiency Optimization System 401;

[0168] The Energy Efficiency Optimization System 401 is connected to the Internet of Things system 105 and the Electric Field Control Module 103, and is used to generate an optimization control strategy based on the received operation data to adaptively optimize and adjust the electric field power.

[0169] Specifically, the Energy Efficiency Optimization System 401 dynamically optimizes the output of the electric field power on the premise of ensuring the separation accuracy, reduces energy consumption, and improves the overall operation efficiency. The Energy Efficiency Optimization System 401 is respectively connected to the Internet of Things system 105 and the Electric Field Control Module 103. By continuously monitoring and analyzing the system operation status, it independently generates and issues an optimization control strategy to achieve the adaptive adjustment of the electric field power and the switching of the intelligent energy-saving mode.

[0170] In one embodiment, the Energy Efficiency Optimization System 401 is specifically used for:

[0171] Receiving the operation data from the Internet of Things system 105;

[0172] Generating an optimization control strategy through dynamic analysis based on the operation data;

[0173] Sending the optimization control strategy to the Electric Field Control Module 103 to adaptively optimize and adjust the electric field power.

[0174] Specifically, the energy efficiency optimization system 401 receives operation data from the Internet of Things system 105, mainly including: current material flow rate, electric field load, real-time energy consumption, equipment temperature, charge change, electrode response efficiency, etc. The energy efficiency optimization system 401 adopts a method combining fuzzy logic control (FLC) and deep reinforcement learning (DRL) to perform real-time adaptive adjustment of the electric field output power: during high-load sorting, the electric field intensity is increased to ensure sufficient charging and sorting accuracy; during low-load or material shortage stages, the voltage and energy consumption are automatically reduced to prevent over-discharge. Among them, according to the material flow rate, charge amount, and electrode load conditions, fuzzy control rules are established to adjust the electric field voltage and frequency.

[0175] The energy efficiency optimization system 401 generates an optimization control strategy based on the current material load and energy consumption status, including: adjusting the output voltage of the high-voltage power supply, controlling the electric field polarity switching frequency, switching to low-power operation or intermittent power supply mode, etc. The control instruction is sent to the electric field control module 103 through the communication bus to complete real-time power adjustment.

[0176] In one embodiment, the DRL algorithm (such as DQN or PPO) learns the balance point of energy consumption-accuracy-stability in the simulation environment through a pre-trained reinforcement learning strategy, and sets the following reward function: improvement in sorting accuracy (positive reward); improvement in electric field uniformity (positive reward); reduction in energy consumption (positive reward); electric energy waste or efficiency decline (negative penalty). The optimization goal is: when the load is high, automatically increase the electric field intensity to ensure sorting accuracy; when the load is low, reduce the electric field power to avoid excessive energy loss; adaptively adjust the cycle, learn through historical data, optimize the adjustment frequency, and reduce the impact of violent fluctuations in the electric field on the sorting effect.

[0177] In one embodiment, the energy efficiency optimization system 401 is further configured with an automatic energy-saving mode switching function, which is used to automatically determine whether to enter the energy-saving operation state when the device is in a low-load operation state, and dynamically adjust the electric field power strategy according to the prediction result, so as to effectively reduce energy consumption and improve the operation efficiency of the device.

[0178] The energy efficiency optimization system 401 continuously monitors the key operation parameters during the device operation process, and triggers the energy-saving mode evaluation when the following conditions are met:

[0179] The material entry rate is lower than the preset threshold, for example, the actual material flow rate is lower than 30% of the full-load operation capacity of the system; the electrode power consumption continues to be at a low level, indicating that the system is not in a high-load operation; the short-term prediction result based on historical data shows no high-load working condition trend in the future. When the above conditions are all met, the energy efficiency optimization system 401 will evaluate whether to enter the energy-saving operation state.

[0180] To improve the accuracy and responsiveness of energy-saving mode switching, the energy efficiency optimization system 401 introduces an LSTM (Long Short-Term Memory) prediction model to predict and analyze the changing trend of material load in the next period of time. According to the prediction results, the energy efficiency optimization system 401 can execute the following strategies:

[0181] If the prediction results show that the load remains low in the short term, the system will enter the "low-power energy-saving mode", significantly reducing the electric field voltage and energy loss; if the load is likely to recover or fluctuate in a short time, the system will maintain the "dynamic standby mode", with the electric field operating at a low intensity to quickly respond and resume efficient sorting when the load picks up.

[0182] After entering the energy-saving mode, the energy efficiency optimization system 401 will execute the following electric field control strategies based on different operating scenarios:

[0183] Power optimization adjustment: Automatically reduce the output level of the high-voltage power supply to avoid unnecessary electrode discharge and energy consumption waste;

[0184] Periodic working mode: When the equipment is in a low-load state for a long time, the system can switch to the "intermittent power supply" method, intermittently activating the electric field according to the set period to further reduce the energy consumption caused by continuous operation;

[0185] Edge computing optimization scheduling: The judgment of the energy-saving mode and the strategy switching can be completed locally through the edge computing module without relying on the cloud platform 203 frequently, improving the strategy response speed and reducing communication latency.

[0186] This application provides an electrostatic separation system, method and device based on the Internet of Things and artificial intelligence. The system includes: a feeding system, a material receiving system, a charging processing module, an electric field control module, an artificial intelligence optimization system and an Internet of Things system; the feeding system is connected to the charging processing module for evenly feeding materials into the electrostatic separation system; the Internet of Things system is connected to the artificial intelligence optimization system through network communication for collecting the characteristic data of the materials and the operation data of the electrostatic separation system and performing data processing and analysis on the characteristic data and the operation data; the artificial intelligence optimization system is used to generate optimal electric field adjustment instructions and charging control strategies according to the received characteristic data and operation data; the charging processing module is connected to the electric field control module and connected to the artificial intelligence optimization system through a communication bus for adjusting the charging mode and charging parameters of the materials according to the received charging control strategy; the electric field control module is connected to the artificial intelligence optimization system through a communication bus for optimizing the electric field distribution according to the received electric field adjustment instructions; the material receiving system is connected to the electric field control module for receiving the materials after electrostatic separation. This application realizes the intelligent optimization and adaptive adjustment of electrostatic separation, improving the separation accuracy and system stability.

[0187] Among them, the feeding system realizes the uniform conveying of materials, ensuring the stability and sorting accuracy of the subsequent charging and sorting processes; the Internet of Things system realizes the real-time perception and data collection of material characteristics and equipment operating states, providing data support for intelligent optimization; the artificial intelligence optimization system realizes the adaptive optimization based on material characteristics and equipment states, automatically generating control strategies for electric fields and charging parameters, improving the intelligence level and energy efficiency of the equipment; the charging processing module realizes the intelligent selection and adjustment of charging modes and charging parameters for different materials, improving the charging uniformity of materials and the subsequent sorting efficiency; the electric field control module realizes the dynamic optimization of electrode voltage, electric field strength and distribution, ensuring that the electric field adapts to the material state and reducing energy consumption; the material receiving system realizes the efficient classification and collection of sorted materials, ensuring the continuity and practicality of the entire electrostatic sorting process.

[0188] Figure 5 is a schematic flow chart of an electrostatic sorting method based on the Internet of Things and artificial intelligence provided by an embodiment of the present application. As Figure 5 shown, the electrostatic sorting method based on the Internet of Things and artificial intelligence provided by the present application includes:

[0189] S501: Receive the feature data from the Internet of Things system;

[0190] S502: Determine the charging mode according to the feature data and a pre-trained decision tree model;

[0191] S503: Generate charging parameters according to the feature data and a pre-trained reinforcement learning model corresponding to the charging mode;

[0192] S504: Generate a charging control strategy using the charging mode and the charging parameters and send it to the charging processing module.

[0193] From Figure 5 the shown process, it can be seen that the electrostatic sorting method based on the Internet of Things and artificial intelligence provided by the present application realizes the intelligent optimization and adaptive adjustment of electrostatic sorting by receiving the feature data from the Internet of Things system; determining the charging mode according to the feature data and a pre-trained decision tree model; generating charging parameters according to the feature data and a pre-trained reinforcement learning model corresponding to the charging mode; generating a charging control strategy using the charging mode and the charging parameters and sending it to the charging processing module, improving the sorting accuracy and system stability.

[0194] Among them, the real-time acquisition of material characteristic data is achieved by receiving the characteristic data from the Internet of Things system, providing an accurate data basis for subsequent intelligent decision-making; the charging mode is determined according to the characteristic data and the pre-trained decision tree model, which can intelligently identify the adapted charging mode, improving the pertinence of the charging method and the sorting adaptability; the charging parameters are generated according to the characteristic data and the pre-trained reinforcement learning model corresponding to the charging mode, which can generate the optimal charging parameters matching the material characteristics, ensuring stable charging effect and optimal energy efficiency; by using the charging mode and the charging parameters to generate a charging control strategy and sending it to the charging processing module, a complete charging control strategy can be formed and sent down for execution, realizing the automatic control and fine adjustment of the charging process.

[0195] Through the above automatic energy-saving mode switching function, the system can not only effectively reduce the power consumption by 20% - 30% in low-load operation scenarios, but also ensure that the sorting effect is not weakened, realizing the coordinated operation of high-precision and high energy efficiency.

[0196] Next, taking the artificial intelligence optimization system as the execution subject, each step will be explained in detail.

[0197] S501: Receive the characteristic data from the Internet of Things system;

[0198] Specifically, the artificial intelligence optimization system receives the characteristic data of the material from the Internet of Things system. The characteristic data includes physical and electrical characteristics such as the particle size, resistivity, humidity, and dielectric constant of the material, which are collected in real time by the sensor module and uploaded to the artificial intelligence optimization system for use as the input basis for subsequent charging strategy decision-making.

[0199] S502: Determine the charging mode according to the characteristic data and the pre-trained decision tree model;

[0200] Specifically, the artificial intelligence optimization system has a built-in charging mode classification model, preferably a hybrid model based on the combination of decision trees (such as XGBoost, Random Forest, etc.) and deep learning, and establishes the optimal matching rules for charging methods for different types of materials based on historical data. The artificial intelligence optimization system automatically discriminates the optimal charging method according to the currently collected characteristic data, improving the adaptability and sorting efficiency.

[0201] In one embodiment, the steps of pre-training the decision tree model include:

[0202] The artificial intelligence optimization system conducts multiple batches of experiments by combining different types of mixed waste (such as plastics, metals, rubbers, minerals, etc.) during actual operation, and collects corresponding material characteristic data, including but not limited to: particle size, humidity, resistivity, dielectric constant, material composition labels, etc. At the same time, record the sorting effect indicators under different charging modes, such as charging uniformity, sorting rate, and residue rate, etc.

[0203] Take the above characteristic data as input vectors, and combine the "optimal charging mode" determined by manual annotation or expert system as the output label to construct a training sample set. Among them, the charging mode labels include: contact charging, frictional charging, and corona discharge. It is preferably to use an interpretable integrated decision tree model, such as XGBoost (Extreme Gradient Boosting Tree) or RandomForest (Random Forest), and through the supervised learning training method, continuously iterate the model structure in the training set so that it can output the probability distribution or classification result of the optimal charging mode after inputting the material characteristics.

[0204] Conduct multi-dimensional evaluations such as accuracy, recall rate, and F1 value on the trained model in an independent test set. Generalization ability can be improved through means such as cross-validation and hyperparameter adjustment to ensure the stable prediction performance of the model on new materials.

[0205] In one embodiment, plastic materials are suitable for frictional charging, metal materials are suitable for contact charging, and high-humidity materials preferably use the corona discharge mode.

[0206] S503: Generate charging parameters according to the characteristic data and the pre-trained reinforcement learning model corresponding to the charging mode;

[0207] Specifically, for the determined charging mode, the artificial intelligence optimization system calls the corresponding deep reinforcement learning model, and combines the current material characteristics and the target sorting accuracy to output a set of optimized charging parameters. Under different charging modes, the adjusted parameters are different.

[0208] In one embodiment, under the contact charging mode, the adjusted parameters are friction time, pressure, and friction material type; under the frictional charging mode, the adjusted parameters are friction speed, material flow rate, and friction surface resistance; under the corona discharge mode, the adjusted parameters are electrode voltage, electrode spacing, corona frequency, etc. The charging parameter optimization strategy constructed by reinforcement learning can dynamically balance the charging efficiency and energy consumption, and significantly improve the system operation efficiency. Set a reward function, and give positive or negative feedback on the parameter adjustment according to the material charging uniformity, energy consumption, and sorting accuracy, so that the artificial intelligence optimization system continuously optimizes the adjustment strategy.

[0209] In one embodiment, generating charging parameters according to the characteristic data and the pre-trained reinforcement learning model corresponding to the charging mode includes:

[0210] After determining the charging mode, the artificial intelligence optimization system needs to further determine the corresponding combination of charging parameters (such as voltage, contact pressure, rotation speed, electrode spacing, etc.) to meet the following objectives: maximize charging efficiency and uniformity; minimize energy consumption; ensure stable operation and avoid electric breakdown or charge imbalance.

[0211] Adopt a deep neural network (DQN (Deep Q-Network) or PPO (Proximal Policy Optimization)), the model input is the material characteristic data; the model output is the optimal combination of charging parameters; the model training objective is to minimize the charging deviation between the predicted value and the historical optimal sample, while meeting the power consumption threshold and safety constraints.

[0212] Since indicators such as material humidity and charge retention fluctuate dynamically over time, the artificial intelligence optimization system introduces a time series prediction module based on a long short-term memory network (LSTM) to perform short-term predictions on the material charge state and sorting efficiency trend. The artificial intelligence optimization system evaluates the deviation between the current charging parameters and the actual charging effect based on the feedback data. When it is found that the efficiency decreases or the charging deviates, rapid parameter fine-tuning is performed to achieve closed-loop adjustment and adaptive correction.

[0213] In one embodiment, the artificial intelligence optimization system predicts the charge change trend of the material through LSTM (long short-term memory network) to achieve pre-regulation and prevent the sorting efficiency from decreasing due to environmental changes (such as increased humidity and charging attenuation).

[0214] S504: Generate a charging control strategy using the charging mode and the charging parameters and send it to the charging processing module.

[0215] Specifically, the artificial intelligence optimization system encapsulates the generated charging mode and charging parameters into a complete charging control strategy and sends it to the charging processing module for execution to achieve intelligent closed-loop control of the charging method and operating parameters. The artificial intelligence optimization system can also perform online fine-tuning and self-learning based on the feedback data to continuously optimize the control strategy.

[0216] Figure 6 It is a schematic flowchart of an electrostatic sorting method based on the Internet of Things and artificial intelligence provided by an embodiment of the present application. As Figure 6 shown, the electrostatic sorting method based on the Internet of Things and artificial intelligence provided by the present application further includes:

[0217] S601: Receive the characteristic data and the operation data from the Internet of Things system;

[0218] Specifically, the artificial intelligence optimization system receives material characteristic data and system operation data from the Internet of Things system, including: electric field strength, voltage, electrode spacing, electrode temperature, electric field volatility, charge residue, material type, resistivity, humidity, and charged state feedback, etc. The above data is collected in real time through the sensor module and uploaded to the artificial intelligence optimization system through the Internet of Things communication module for unified processing.

[0219] S602: Simulate and model the electric field of the electrostatic separation system according to the operation data to obtain a numerical model of the electric field distribution;

[0220] Specifically, the artificial intelligence optimization system performs electrostatic field simulation on the electrode structure and voltage configuration in the electrostatic separation area through the finite element method to construct a high-precision numerical model of the electric field distribution. This model can reflect spatial characteristics such as potential distribution, electric field line direction, and field strength gradient under different input parameters. The simulation model is fitted and corrected according to historical operation data and measured electric field sensor data to ensure a high degree of consistency between the numerical model and the physical system.

[0221] S603: Generate an electric field distribution state according to the numerical model of the electric field distribution;

[0222] Specifically, during operation, the artificial intelligence optimization system calls the numerical model of the electric field distribution to generate the electric field distribution state (including uniformity index and hotspot area identification, etc.) under the current configuration as the input reference for the reinforcement learning agent.

[0223] S604: Generate electric field distribution control parameters according to the characteristic data, the electric field distribution state, and a pre-constructed reinforcement learning agent that calls the numerical model of the electric field distribution;

[0224] Specifically, a reinforcement learning agent (Reinforcement Learning Agent) is constructed in the artificial intelligence optimization system. Its role is to continuously learn and optimize the electric field control strategy in the electrostatic separation system with the support of the established numerical model of the electric field distribution. The system inputs the state vector in real time during operation, and the reinforcement learning agent outputs the electric field distribution control parameters according to the policy network, including electrode voltage, electrode spacing, and electrode shape, etc., which are used to generate the final electric field adjustment instruction and send it to the electric field control module to make the electric field distribution tend to be optimal.

[0225] By introducing the reinforcement learning agent, the artificial intelligence optimization system can not only achieve dynamic adaptability of the electric field configuration under complex materials and operating conditions, but also continuously learn the feedback in the sorting process, improve decision-making flexibility, optimize energy efficiency control, and have high environmental adaptability and expansion ability.

[0226] In one embodiment, the reinforcement learning agent includes: an environment interface, a state space, an action space, a policy network and a learning algorithm, a reward function design, and a strategy output and online call mechanism.

[0227] The reinforcement learning agent is integrated with a numerical model of the electric field distribution, and the numerical model of the electric field distribution reflects the electrostatic field response under different electrode configurations through finite element simulation (FEM). This simulation environment provides state feedback and reward signals, constituting the training environment of reinforcement learning. The state received by the reinforcement learning agent at each decision-making includes: the current electric field distribution state and the current material properties. The adjustable parameters of the reinforcement learning agent include: electrode voltage, electrode spacing, electrode polarity switching strategy, discharge frequency or trigger period, etc. The reinforcement learning agent is constructed using the Deep Deterministic Policy Gradient (DDPG) or PPO (Proximal Policy Optimization) algorithm, and its policy network is used to map states to actions, and continuously iterates during the training process to optimize the control strategy. The reward function is designed as a multi-objective function to guide the reinforcement learning agent to generate optimized electric field configuration parameters. The specific indicators include: improvement in electric field uniformity (positive reward); improvement in electrostatic separation efficiency (positive reward); reduction in system energy consumption (positive reward); electric field instability, excessive energy consumption, local overload or breakdown (negative penalty).

[0228] In one embodiment, the artificial intelligence optimization system identifies uneven regions in the electric field and interference factors that may affect the separation effect (such as electrode loss, environmental humidity impact, etc.) through potential distribution and field strength gradient analysis.

[0229] In one embodiment, the artificial intelligence optimization system uses RNN (Recurrent Neural Network) or LSTM (Long Short-Term Memory) to predict the short-term change trend of the electric field and adjusts the electrode parameters in advance to reduce sudden abnormal situations during the separation process.

[0230] In one embodiment, in an industrial operation environment, the artificial intelligence optimization system adjusts the voltage source, grounding configuration, and electrode arrangement in real time to achieve adaptive electric field optimization and improve separation stability. Combining edge computing, the AI model is deployed at the device end to reduce latency and improve the optimization response speed.

[0231] S605: Generate an electric field adjustment instruction using the electric field distribution control parameter and send it to the electric field control module.

[0232] Specifically, the artificial intelligence optimization system encapsulates the generated electric field distribution control parameters into electric field adjustment instructions and sends them to the electric field control module for execution, realizing the intelligent closed-loop control of the electric field distribution control parameters. The artificial intelligence optimization system can also perform online fine-tuning and self-learning based on the feedback data to continuously optimize the control strategy.

[0233] The electrostatic separation method based on the Internet of Things and artificial intelligence provided by this application receives the feature data from the Internet of Things system; determines the charging mode according to the feature data and the pre-trained decision tree model; generates the charging parameters according to the feature data and the pre-trained reinforcement learning model corresponding to the charging mode; and generates the charging control strategy by using the charging mode and the charging parameters and sends it to the charging processing module, realizing the intelligent optimization and adaptive adjustment of electrostatic separation, and improving the separation accuracy and system stability.

[0234] Among them, the real-time acquisition of the material characteristic data is realized by receiving the feature data from the Internet of Things system, providing an accurate data basis for subsequent intelligent decision-making; determining the charging mode according to the feature data and the pre-trained decision tree model can intelligently identify the suitable charging mode and improve the pertinence of the charging method and the separation adaptability; generating the charging parameters according to the feature data and the pre-trained reinforcement learning model corresponding to the charging mode can generate the optimal charging parameters matching the material characteristics, ensuring stable charging effect and optimal energy efficiency; generating the charging control strategy by using the charging mode and the charging parameters and sending it to the charging processing module can form a complete charging control strategy and send it down for execution, realizing the automatic control and fine adjustment of the charging process.

[0235] Based on the same inventive concept, the embodiment of this application also provides an electrostatic separation device based on the Internet of Things and artificial intelligence, which can be used to implement the method described in the above embodiment, as described in the following embodiment. Since the principle of the electrostatic separation device based on the Internet of Things and artificial intelligence to solve problems is similar to that of the electrostatic separation method based on the Internet of Things and artificial intelligence, the implementation of the electrostatic separation device based on the Internet of Things and artificial intelligence can refer to the implementation of the method for determining the software performance benchmark, and the repeated parts will not be described again. Hereinafter, the term "unit" or "module" can be a combination of software and / or hardware that can achieve a predetermined function. Although the system described in the following embodiments is preferably implemented in software, the implementation in hardware, or a combination of software and hardware, is also possible and contemplated.

[0236] Figure 7 is a schematic structural diagram of the electrostatic separation device based on the Internet of Things and artificial intelligence provided by an embodiment of this application, as Figure 7 shown, the electrostatic separation device based on the Internet of Things and artificial intelligence provided by this application includes:

[0237] The first data receiving unit 701 is configured to receive the feature data from the Internet of Things system;

[0238] The electrified mode determining unit 702 is configured to determine the electrified mode according to the feature data and a pre-trained decision tree model;

[0239] The electrified parameter generating unit 703 is configured to generate electrified parameters according to the feature data and a pre-trained reinforcement learning model corresponding to the electrified mode;

[0240] The electrified control strategy output unit 704 is configured to generate an electrified control strategy by using the electrified mode and the electrified parameters and send the electrified control strategy to the electrified processing module.

[0241] Figure 8 FIG. is a schematic structural diagram of an electrostatic separation device based on the Internet of Things and artificial intelligence provided by an embodiment of the present application. On the basis of the embodiment, further, as Figure 7 shown, the electrostatic separation device based on the Internet of Things and artificial intelligence provided by the present application further includes: Figure 8 The second data receiving unit 801 is configured to receive the feature data and the operation data from the Internet of Things system;

[0242] The model building unit 802 is configured to perform simulation modeling on the electric field of the electrostatic separation system according to the operation data to obtain an electric field distribution numerical model;

[0243] The electric field distribution state generating unit 803 is configured to generate an electric field distribution state according to the electric field distribution numerical model;

[0244] The control parameter generating unit 804 is configured to generate electric field distribution control parameters according to the feature data, the electric field distribution state, and a reinforcement learning agent that pre-constructs and calls the electric field distribution numerical model;

[0245] The electric field adjustment instruction output unit 805 is configured to generate an electric field adjustment instruction by using the electric field distribution control parameters and send the electric field adjustment instruction to the electric field control module.

[0246] The electrostatic separation method and device based on the Internet of Things and artificial intelligence provided by the present application receive the feature data from the Internet of Things system; determine the electrified mode according to the feature data and a pre-trained decision tree model; generate electrified parameters according to the feature data and a pre-trained reinforcement learning model corresponding to the electrified mode; generate an electrified control strategy by using the electrified mode and the electrified parameters and send the electrified control strategy to the electrified processing module, realizing intelligent optimization and adaptive adjustment of electrostatic separation, and improving the separation accuracy and system stability.

[0247] The electrostatic separation method and device based on the Internet of Things and artificial intelligence provided by the present application receive the feature data from the Internet of Things system; determine the electrified mode according to the feature data and a pre-trained decision tree model; generate electrified parameters according to the feature data and a pre-trained reinforcement learning model corresponding to the electrified mode; generate an electrified control strategy by using the electrified mode and the electrified parameters and send the electrified control strategy to the electrified processing module, realizing intelligent optimization and adaptive adjustment of electrostatic separation, and improving the separation accuracy and system stability.

[0248] Among them, real-time acquisition of material characteristic data is achieved by receiving the characteristic data from the Internet of Things system, providing an accurate data basis for subsequent intelligent decision-making; determining the charging mode according to the characteristic data and the pre-trained decision tree model can intelligently identify the adapted charging mode, improving the pertinence of the charging method and the sorting adaptability; generating charging parameters according to the characteristic data and the pre-trained reinforcement learning model corresponding to the charging mode can generate the optimal charging parameters matching the material characteristics, ensuring stable charging effect and optimal energy efficiency; forming a complete charging control strategy and sending it to the charging processing module by using the charging mode and the charging parameters can form a complete charging control strategy and issue it for execution, realizing automatic control and fine adjustment of the charging process.

[0249] From the hardware level, to solve the problems that traditional electrostatic separation equipment cannot intelligently adjust the electric field strength and charging parameters according to material changes and operating states, resulting in high energy consumption, low sorting accuracy, frequent equipment maintenance, and lack of automatic optimization and remote management capabilities, this application provides an embodiment of an electronic device for implementing all or part of the content in the above-mentioned electrostatic separation method based on the Internet of Things and artificial intelligence. The electronic device specifically includes the following:

[0250] A processor, a memory, a communication interface, and a bus; wherein, the processor, the memory, and the communication interface complete communication with each other through the bus; the communication interface is used to realize information transmission between the electrostatic separation device based on the Internet of Things and artificial intelligence and related devices such as the core business system, the user terminal, and the relevant database. The logic controller can be a desktop computer, a tablet computer, a mobile terminal, etc., and this embodiment is not limited thereto. In this embodiment, the logic controller can be implemented with reference to the embodiments of the electrostatic separation method based on the Internet of Things and artificial intelligence and the embodiments of the electrostatic separation device based on the Internet of Things and artificial intelligence, and the content is incorporated herein, and the repeated parts will not be elaborated.

[0251] It can be understood that the user terminal may include a smart phone, a tablet electronic device, an Internet set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device, etc. Among them, the smart wearable device may include smart glasses, a smart watch, a smart bracelet, etc.

[0252] In practical applications, part of the electrostatic sorting method based on the Internet of Things and artificial intelligence can be executed on the side of the electronic device as described above, or all operations can be completed in the client device. Specifically, it can be selected according to the processing capacity of the client device and the limitations of the user usage scenario, etc. This application does not make any limitations in this regard. If all operations are completed in the client device, the client device may further include a processor.

[0253] The above-mentioned client device may have a communication module (i.e., a communication unit), which can communicate with a remote server to achieve data transmission with the server. The server may include a server on the side of the task scheduling center, and in other implementation scenarios, it may also include a server of an intermediate platform, such as a server of a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, or may include a server cluster composed of multiple servers, or a server structure of a distributed device.

[0254] Figure 9 It is a schematic block diagram of the system composition of the electronic device 9600 according to an embodiment of the present application. As Figure 9 shown, the electronic device 9600 may include a central processing unit 9100 and a memory 9140; the memory 9140 is coupled to the central processing unit 9100. It should be noted that this Figure 9 is exemplary; other types of structures may also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0255] In one embodiment, the function of the electrostatic sorting method based on the Internet of Things and artificial intelligence can be integrated into the central processing unit 9100. Among them, the central processing unit 9100 may be configured to perform the following controls:

[0256] S501: Receive the feature data from the Internet of Things system;

[0257] S502: Determine the charging mode according to the feature data and the pre-trained decision tree model;

[0258] S503: Generate charging parameters according to the feature data and the pre-trained reinforcement learning model corresponding to the charging mode;

[0259] S504: Generate a charging control strategy using the charging mode and the charging parameters and send it to the charging processing module.

[0260] As can be seen from the above description, the electrostatic separation method and device based on the Internet of Things and artificial intelligence provided by this application achieve adaptive adjustment of the electric field, voltage, and charging mode. While ensuring the separation accuracy, it reduces the energy consumption by 20% - 30%, and improves the intelligence level, maintenance efficiency, and long-term operation stability of the equipment. By receiving the characteristic data from the Internet of Things system, it realizes the real-time acquisition of material characteristic data, providing an accurate data basis for subsequent intelligent decision-making; determining the charging mode according to the characteristic data and the pre-trained decision tree model can intelligently identify the suitable charging mode, improving the pertinence of the charging method and the separation adaptability; generating the charging parameters according to the characteristic data and the pre-trained reinforcement learning model corresponding to the charging mode can generate the optimal charging parameters matching the material characteristics, ensuring stable charging effect and optimal energy efficiency; by using the charging mode and the charging parameters to generate the charging control strategy and sending it to the charging processing module, a complete charging control strategy can be formed and sent for execution, realizing the automatic control and fine adjustment of the charging process.

[0261] In another embodiment, the electrostatic separation device based on the Internet of Things and artificial intelligence can be separately configured from the central processor 9100. For example, the data composite transmission device, the electrostatic separation device based on the Internet of Things and artificial intelligence, can be configured as a chip connected to the central processor 9100, and the functions of the electrostatic separation method based on the Internet of Things and artificial intelligence can be realized through the control of the central processor.

[0262] As Figure 9 shown, the electronic device 9600 may further include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It should be noted that the electronic device 9600 does not necessarily have to include Figure 9 all the components shown in Figure 9 ; in addition, the electronic device 9600 may further include

[0263] components not shown in Figure 9 ; reference can be made to the prior art.

[0264] Among them, the memory 9140, for example, can be one or more of a buffer, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory, or other suitable devices. It can store the above information related to failures, and can also store programs for executing relevant information. And the central processor 9100 can execute the program stored in the memory 9140 to achieve information storage or processing, etc.

[0265] The input unit 9120 provides input to the central processing unit 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to supply power to the electronic device 9600. The display 9160 is used to display display objects such as images and texts. The display can be, for example, an LCD display, but is not limited thereto.

[0266] The memory 9140 can be a solid-state memory, for example, a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It can also be a memory that stores information even when power is off, can be selectively erased and has more data. Examples of such a memory are sometimes referred to as EPROMs, etc. The memory 9140 can also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 can include an application / function storage unit 9142, which is used to store application programs and function programs or the processes for operating the electronic device 9600 through the central processing unit 9100.

[0267] The memory 9140 can also include a data storage unit 9143, which is used to store data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 can include various drivers of the electronic device for communication functions and / or for performing other functions of the electronic device (such as a messaging application, an address book application, etc.).

[0268] The communication module 9110 is a transmitter / receiver 9110 that transmits and receives signals via the antenna 9111. The communication module (transmitter / receiver) 9110 is coupled to the central processing unit 9100 to provide input signals and receive output signals, which can be the same as in the case of a conventional mobile communication terminal.

[0269] Based on different communication technologies, multiple communication modules 9110 can be provided in the same electronic device, such as a cellular network module, a Bluetooth module, and / or a wireless local area network module, etc. The communication module (transmitter / receiver) 9110 is also coupled to the speaker 9131 and the microphone 9132 via the audio processor 9130 to provide an audio output via the speaker 9131 and receive an audio input from the microphone 9132, thereby realizing normal telecommunication functions. The audio processor 9130 can include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor 9130 is also coupled to the central processing unit 9100, so that recording can be performed on the local machine through the microphone 9132, and the sound stored on the local machine can be played through the speaker 9131.

[0270] An embodiment of the present application further provides a computer-readable storage medium capable of implementing all steps of the electrostatic separation method based on the Internet of Things and artificial intelligence, where the execution subject in the above embodiment is a server or a client. A computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, all steps of the electrostatic separation method based on the Internet of Things and artificial intelligence, where the execution subject in the above embodiment is a server or a client, are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0271] S501: Receive the feature data from the Internet of Things system;

[0272] S502: Determine the charging mode according to the feature data and a pre-trained decision tree model;

[0273] S503: Generate charging parameters according to the feature data and a pre-trained reinforcement learning model corresponding to the charging mode;

[0274] S504: Generate a charging control strategy by using the charging mode and the charging parameters, and send it to the charging processing module.

[0275] As can be seen from the above description, the electrostatic separation method and device based on the Internet of Things and artificial intelligence provided by the present application realize the adaptive adjustment of the electric field, voltage, and charging mode. While ensuring the separation accuracy, the energy consumption is reduced by 20%-30%, and the intelligence level, maintenance efficiency, and long-term operation stability of the equipment are improved. By receiving the feature data from the Internet of Things system, the real-time acquisition of material characteristic data is realized, providing an accurate data basis for subsequent intelligent decision-making; determining the charging mode according to the feature data and a pre-trained decision tree model can intelligently identify the suitable charging mode, improving the pertinence of the charging method and the separation adaptability; generating charging parameters according to the feature data and a pre-trained reinforcement learning model corresponding to the charging mode can generate the optimal charging parameters matching the material characteristics, ensuring stable charging effect and optimal energy efficiency; by generating a charging control strategy by using the charging mode and the charging parameters and sending it to the charging processing module, a complete charging control strategy can be formed and sent down for execution, realizing the automatic control and fine adjustment of the charging process.

[0276] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a device, or a computer program product. Therefore, the present invention can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0277] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (devices), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and combinations of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable data processing devices to generate a machine, such that the instructions executed by the processors of the computer or other programmable data processing devices produce means for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0278] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including instruction means that implement the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0279] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0280] Specific embodiments are applied in the present invention to elaborate on the principles and implementation manners of the present invention. The descriptions of the above embodiments are only used to help understand the method of the present invention and its core idea; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. An electrostatic sorting system based on the Internet of Things and artificial intelligence, characterized in that: include: Feeding system, receiving system, live processing module, electric field control module, artificial intelligence optimization system and Internet of Things system; The feeding system is connected to the charged processing module and is used to evenly feed the material into the electrostatic separation system; The Internet of Things system is connected to the artificial intelligence optimization system through network communication, and is used to collect characteristic data of the material and operation data of the electrostatic separation system and perform data processing and analysis on the characteristic data and the operation data; The artificial intelligence optimization system is used to generate the optimal electric field adjustment instruction and live control strategy according to the received characteristic data and the operating data; The charging processing module is connected to the electric field control module and connected to the artificial intelligence optimization system through a communication bus, and is used to adjust the charging mode and charging parameters of the material according to the received charging control strategy; The electric field control module is connected to the artificial intelligence optimization system via a communication bus, and is used to optimize the electric field distribution according to the received electric field adjustment instructions; The material receiving system is connected to the electric field control module and is used to receive materials after electrostatic separation.

2. The electrostatic separation system based on Internet of Things and artificial intelligence according to claim 1 is characterized in that: The Internet of Things system includes a sensor module, an Internet of Things communication module and a cloud platform; The sensor module is used to collect characteristic data of the material and operation data of the electrostatic separation system; The Internet of Things communication module is used to receive the feature data and the operation data and send them to the artificial intelligence optimization system and the cloud platform; The cloud platform is used to process and analyze the received feature data and operation data.

3. The electrostatic separation system based on Internet of Things and artificial intelligence according to claim 1 is characterized in that: The artificial intelligence optimization system is specifically used for: Receiving the characteristic data from the Internet of Things system; Determining a charging mode according to the characteristic data and a pre-trained decision tree model; generating charging parameters according to the feature data and a pre-trained reinforcement learning model corresponding to the charging mode; A charging control strategy is generated using the charging mode and the charging parameters and sent to the charging processing module.

4. The electrostatic separation system based on Internet of Things and artificial intelligence according to claim 1 is characterized in that: The artificial intelligence optimization system is specifically used for: Receiving the characteristic data and the operation data from the Internet of Things system; Performing simulation modeling on the electric field of the electrostatic separation system according to the operating data to obtain a numerical model of the electric field distribution; generating an electric field distribution state according to the electric field distribution numerical model; Generate electric field distribution control parameters according to the characteristic data, the electric field distribution state and a pre-built reinforcement learning agent that calls the electric field distribution numerical model; An electric field adjustment instruction is generated using the electric field distribution control parameter and sent to the electric field control module.

5. The electrostatic separation system based on Internet of Things and artificial intelligence according to claim 1 is characterized in that: Also includes: Safety intelligent management system; The safety intelligent management system is connected to the Internet of Things system and the artificial intelligence optimization system, and is used to identify the abnormal state and failure trend of the electrostatic sorting system based on the received operation data, and send the identification results to the artificial intelligence optimization system to adjust the operation strategy.

6. The electrostatic separation system based on Internet of Things and artificial intelligence according to claim 5 is characterized in that: The security intelligent management system is specifically used for: Receiving the operation data from the Internet of Things system; Identifying an abnormal state of the electrostatic separation system according to the operating data and a pre-trained anomaly detection model; The anomaly detection model is obtained by performing unsupervised training on the historical operating data of the electrostatic separation system under normal operating conditions; Identifying the failure trend of the electrostatic separation system according to the operation data and the pre-trained fault prediction model; The fault prediction model is obtained by conducting supervised training on the historical operation data and fault records of the electrostatic separation system in different operation cycles; The recognition result is sent to the artificial intelligence optimization system to adjust the operation strategy.

7. The electrostatic separation system based on Internet of Things and artificial intelligence according to claim 1 is characterized in that: Also includes: Energy efficiency optimization system; The energy efficiency optimization system is connected to the Internet of Things system and the electric field control module, and is used to generate an optimization control strategy based on the received operating data to perform adaptive optimization and regulation on the electric field power.

8. The electrostatic separation system based on Internet of Things and artificial intelligence according to claim 7 is characterized in that: The energy efficiency optimization system is specifically used for: Receiving the operation data from the Internet of Things system; generating an optimized control strategy through dynamic analysis according to the operating data; The optimization control strategy is sent to the electric field control module to perform adaptive optimization adjustment on the electric field power.

9. An electrostatic separation method based on the Internet of Things and artificial intelligence, applied to the electrostatic separation system based on the Internet of Things and artificial intelligence as claimed in any one of claims 1 to 8, characterized in that: include: Receiving the characteristic data from the Internet of Things system; Determining a charging mode according to the characteristic data and a pre-trained decision tree model; generating charging parameters according to the feature data and a pre-trained reinforcement learning model corresponding to the charging mode; A charging control strategy is generated using the charging mode and the charging parameters and sent to the charging processing module.

10. The electrostatic separation method based on Internet of Things and artificial intelligence according to claim 9, characterized in that: Also includes: Receiving the characteristic data and the operation data from the Internet of Things system; Performing simulation modeling on the electric field of the electrostatic separation system according to the operating data to obtain a numerical model of the electric field distribution; generating an electric field distribution state according to the electric field distribution numerical model; Generate electric field distribution control parameters according to the characteristic data, the electric field distribution state and a pre-built reinforcement learning agent that calls the electric field distribution numerical model; An electric field adjustment instruction is generated using the electric field distribution control parameter and sent to the electric field control module.

11. An electrostatic separation device based on the Internet of Things and artificial intelligence, applied to the electrostatic separation system based on the Internet of Things and artificial intelligence as claimed in any one of claims 1 to 8, characterized in that: include: A first data receiving unit, configured to receive the characteristic data from the Internet of Things system; A charging mode determination unit, used to determine the charging mode according to the characteristic data and a pre-trained decision tree model; A charging parameter generating unit, used for generating charging parameters according to the characteristic data and a pre-trained reinforcement learning model corresponding to the charging mode; The charging control strategy output unit is used to generate a charging control strategy using the charging mode and the charging parameters and send the strategy to the charging processing module.

12. The electrostatic separation device based on Internet of Things and artificial intelligence according to claim 11, characterized in that: Also includes: A second data receiving unit, configured to receive the feature data and the operation data from the Internet of Things system; A model building unit, used to simulate and model the electric field of the electrostatic separation system according to the operating data to obtain a numerical model of the electric field distribution; An electric field distribution state generating unit, used for generating an electric field distribution state according to the electric field distribution numerical model; A control parameter generating unit, configured to generate electric field distribution control parameters according to the characteristic data, the electric field distribution state and a pre-built reinforcement learning agent that calls the electric field distribution numerical model; The electric field adjustment instruction output unit is used to generate an electric field adjustment instruction using the electric field distribution control parameter and send the electric field adjustment instruction to the electric field control module.

13. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 9 to 10 is implemented.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 9 to 10 is implemented.

15. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 9 to 10 is implemented.

Images