A digital monitoring management method and system for solid waste treatment

Abstract

The application discloses a kind of digital monitoring management method and system for solid waste treatment, including, by real-time acquisition solid waste treatment each link data information, and using digital twin model carries out data analysis and efficiency prediction, finally optimizes each link efficiency in solid waste treatment process, to improve overall processing effect;The application is combined by digital twin model and multi-objective learning framework, not only can accurately predict each link efficiency in solid waste treatment process, but also can maintain high robustness under different interference conditions, significantly improve the intelligent level and resource utilization efficiency of solid waste treatment system, with good economy, environmental adaptability and stability, applicable to various scales of solid waste treatment scene.

Description

Technical Field

[0001] This invention relates to the field of solid waste monitoring technology under machine learning, and in particular to a digital monitoring and management method and system for solid waste treatment. Background Technology

[0002] In recent years, solid waste management has received significant attention from countries worldwide. With accelerated urbanization and deepening industrialization, the amount of solid waste generated has increased dramatically, exacerbating its treatment challenges and environmental impact. Traditional solid waste management methods typically rely on manual operation and static process design, but this approach has significant limitations in addressing the complexity of waste types and the dynamic changes in treatment processes. With the rapid development of technologies such as the Internet of Things (IoT), big data analytics, and artificial intelligence (AI), solid waste management is gradually transforming towards digitalization and intelligentization. Especially with the support of sensor networks, edge computing, and cloud platforms, real-time data collection, transmission, and analysis across the entire waste management process have become possible.

[0003] Despite some progress in existing technologies, numerous technical bottlenecks remain in practical applications. First, current sensor networks and edge computing solutions suffer from insufficient real-time performance and inadequate data integrity when processing solid waste data, particularly in the optimization of high-dimensional and heterogeneous data processing. Second, the application of digital twin models in solid waste treatment is still in its early stages; model training often fails to adequately consider the complex relationships within the solid waste treatment process, resulting in predictions that do not meet actual needs. Furthermore, in optimizing the efficiency of solid waste treatment, traditional single-objective optimization methods struggle to handle the trade-offs and dynamic adjustments between multiple objectives, failing to achieve continuous dynamic optimization of treatment efficiency. These shortcomings not only limit the widespread application of existing technologies in complex solid waste treatment scenarios but also impact the overall treatment efficiency and resource utilization rate of solid waste. Summary of the Invention

[0004] The purpose of this section is to outline some aspects of embodiments of the present invention and to briefly describe some preferred embodiments. Simplifications or omissions may be made in this section, as well as in the abstract and title of this application, to avoid obscuring the purpose of these documents; however, such simplifications or omissions should not be construed as limiting the scope of the invention.

[0005] In view of the aforementioned existing problems, this invention is proposed. Therefore, this invention provides a digital monitoring and management method for solid waste treatment to solve the problems mentioned in the background art.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a digital monitoring and management method for solid waste treatment, comprising:

[0008] Data on solid waste is collected in real time at each stage of the treatment process by sensors, and the data is transmitted to the cloud platform through the edge computing unit to form a real-time archive of data information in the solid waste treatment process.

[0009] A digital twin model is created based on the real-time archives of the data information, the digital twin model is trained, and the efficiency prediction results of solid waste in each treatment stage are output.

[0010] Based on the efficiency prediction results of solid waste in each treatment stage, a multi-objective learning framework is constructed. Through the multi-objective learning framework, the efficiency prediction results are continuously monitored and adjusted to achieve efficiency management of solid waste in each treatment stage.

[0011] As a preferred embodiment of the digital monitoring and management method for solid waste treatment described in this invention, the method includes: real-time collection of data information on solid waste at each treatment stage via sensors, and transmission of the data information to a cloud platform via a wireless network, including:

[0012] All sensors are organized into multiple sensor networks in the form of a network topology diagram. Each sensor network corresponds to a solid waste treatment stage. Each sensor in each sensor network represents a node of the sensor network. Nodes in the current sensor network are added or deleted according to the type of solid waste treatment stage.

[0013] The data collected by the sensor network is transmitted to the cloud platform through the edge computing unit. During the transmission process, the data collected by the sensor network is processed. The cloud platform stores the processed data in a structured or semi-structured form to form a real-time archive of data information in the solid waste treatment process.

[0014] As a preferred embodiment of the digital monitoring and management method for solid waste treatment described in this invention, the data information collected by the sensor network is processed, including:

[0015] An autoencoder is constructed to arrange the collected data information into a high-dimensional input matrix. Multiple convolutional layers are used to map the input data information to a low-dimensional space and output low-dimensional data information.

[0016] The encoder-decoder pairing performs backpropagation on the low-dimensional data information. The mean square error formula is used to check whether there are errors in the processed data information. If there are no errors, the processed data information is output through the bottleneck layer of the encoder, and the low-dimensional features of each input data are extracted. Otherwise, the mapping is performed again.

[0017] As a preferred embodiment of the digital monitoring and management method for solid waste treatment described in this invention, the method includes: creating a digital twin model based on real-time archives of the data information, training the digital twin model, and outputting efficiency prediction results for solid waste in each treatment stage, including:

[0018] The digital twin model receives data information from real-time archives through the input layer and sends the data information into the embedding layer. The embedding layer classifies the data information and models the temporal dependencies of each stage in the solid waste treatment process. Through stacked recurrent neural network layers, it records the duration and amount of solid waste in each treatment process. The results of the stacked recurrent neural network layers are output through a fully connected layer.

[0019] Using the modeled data, the stacked recurrent neural network layers are trained until each layer in the stacked recurrent neural network is trained.

[0020] As a preferred embodiment of the digital monitoring and management method for solid waste treatment described in this invention, the stacked recurrent neural network layer includes:

[0021] The stacked recurrent neural network is divided into long short-term memory network layers and gated recurrent unit layers with equal number of layers. The output of each layer of the stacked recurrent neural network is used as the input of the next layer to form a data recursive transmission mechanism.

[0022] During training, if the amount of data processed by the current layer exceeds the capacity of the layer, if the current layer is a Long Short-Term Memory (LSTM) network layer, it will be replaced by a gated recursive unit layer; otherwise, no processing will be performed.

[0023] As a preferred embodiment of the digital monitoring and management method for solid waste treatment described in this invention, a multi-objective learning framework is constructed based on the efficiency prediction results of output solid waste in each treatment stage, including:

[0024] The real-time archive of data information during the solid waste treatment process is invoked, and the data in the real-time archive is used as a Markov decision process to construct a multi-objective learning framework in terms of state space and action space respectively.

[0025] The efficiency prediction results of the solid waste in each treatment stage are used as the input set of the multi-objective learning framework, and a reward function is designed. The reward factor is increased if the efficiency of the solid waste in each treatment stage reaches the expected target, and decreased if it is lower than the expected target.

[0026] As a preferred embodiment of the digital monitoring and management method for solid waste treatment described in this invention, the method includes: continuously monitoring and adjusting the efficiency prediction results through the multi-objective learning framework, comprising:

[0027] The efficiency prediction results of solid waste in each treatment stage are calculated by weighted average method. The weighted average is monitored in real time. If the efficiency of solid waste in any treatment stage is lower than the weighted average, the reward factor of the reward function in the multi-objective learning framework is adjusted.

[0028] Secondly, the present invention provides a digital monitoring and management system for solid waste treatment, comprising:

[0029] The solid waste information digitization module is configured to collect data information of solid waste in each treatment stage in real time through sensors, and transmit the data information to the cloud platform through the edge computing unit to form a real-time archive of data information in the solid waste treatment process;

[0030] The solid waste efficiency prediction module is configured to create a digital twin model based on the real-time archive of the data information, train the digital twin model, and output the efficiency prediction results of solid waste in each treatment stage.

[0031] The solid waste efficiency monitoring module is configured to construct a multi-objective learning framework based on the efficiency prediction results of the output solid waste in each treatment stage. Through the multi-objective learning framework, the efficiency prediction results are continuously monitored and adjusted to achieve efficiency management of solid waste in each treatment stage.

[0032] Thirdly, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement any step of the above-described method.

[0033] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any step of the above-described method.

[0034] Compared with existing technologies, the beneficial effects of the invention are as follows:

[0035] 1. This invention constructs a digital twin model and trains it using supervised learning methods, which can accurately simulate the operating status of solid waste at each stage of treatment. In addition, based on historical data and real-time archives, it outputs efficiency prediction results for each stage. Compared with the traditional prediction methods used in the prior art, this invention can better reflect the dynamic changes in the solid waste treatment process and provide more accurate prediction and evaluation.

[0036] 2. By combining Long Short-Term Memory (LSTM) networks and Gated Recurrent Units (GRUs), a stacked recurrent neural network (RNN) is constructed for time series modeling. This can better capture the temporal dependencies between various stages in the solid waste processing process. When the amount of data processed exceeds the computing power of a certain layer, the model can automatically adjust the network layer structure, improve processing power and prediction accuracy, and enhance the model's adaptability and robustness in complex scenarios.

[0037] 3. By constructing a multi-objective learning framework, the efficiency indicators in the solid waste treatment process can be continuously monitored and adjusted. Through Markov decision process modeling and reward function design, this invention can achieve collaborative optimization of multiple treatment links (i.e., when the efficiency of a certain treatment link is lower than expected, the optimization scheme is quickly adjusted to avoid the negative impact of inefficient links on the entire treatment process). It effectively balances the possible conflicts between various objectives and ensures the comprehensive optimization of various parameters (such as equipment processing capacity, equipment energy consumption, etc.) in different treatment links, which greatly improves the overall efficiency of solid waste treatment. Attached Figure Description

[0038] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0039] Figure 1 This is a flowchart illustrating the overall process of a digital monitoring and management method for solid waste treatment according to an embodiment of the present invention.

[0040] Figure 2 This is a simplified digital twin model diagram of a digital monitoring and management method for solid waste treatment according to an embodiment of the present invention;

[0041] Figure 3 This is a comparison chart of the efficiency of each processing stage in the digital monitoring and management method for solid waste treatment according to an embodiment of the present invention. Detailed Implementation

[0042] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0043] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0044] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0045] This invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of this invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not adhering to the usual scale. Furthermore, the schematic diagrams are merely examples and should not be construed as limiting the scope of protection of this invention. In actual fabrication, the three-dimensional spatial dimensions of length, width, and depth should be included.

[0046] Furthermore, in the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used solely for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In addition, the terms "first," "second," or "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0047] Unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" in this invention should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; similarly, they can refer to mechanical connections, electrical connections, or direct connections, or indirect connections through an intermediate medium, or internal connections between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0048] Example 1

[0049] Reference Figure 1 and Figure 2This is the first embodiment of the present invention, which provides a digital monitoring and management method for solid waste treatment, including:

[0050] S1. Real-time data information of solid waste in each treatment stage is collected by sensors, and the data information is transmitted to the cloud platform through the edge computing unit to form a real-time archive of data information in the solid waste treatment process;

[0051] Specifically, sensors can be categorized based on their application scenarios in various treatment stages of solid waste (such as transportation, storage, sorting, and incineration), including temperature sensors, humidity sensors, gas sensors (such as those monitoring methane, ammonia, and carbon dioxide), and weight sensors.

[0052] Specifically, the data information includes physical parameters (temperature, humidity, pressure, weight of solid waste, etc.), chemical parameters (gas composition, pH value, dissolved oxygen concentration, etc.), environmental monitoring data (air quality, noise, etc.), process monitoring data (time and speed of the current processing stage, equipment operating status, transportation volume, etc.), and waste classification information (recyclable, hazardous waste, etc.).

[0053] Furthermore, all sensors are organized into multiple sensor networks in the form of a network topology diagram. Each sensor network corresponds to a solid waste treatment stage. In each sensor network, the sensors represent a node of the sensor network. Nodes in the current sensor network are added or deleted according to the type of solid waste treatment stage.

[0054] Specifically, the types of solid waste treatment processes refer to each treatment process (transportation, storage, sorting, incineration, etc.);

[0055] It should be noted that adding or deleting nodes in the current sensor network aims to disable the corresponding solid waste treatment process and enable it when needed, rather than physically adding or deleting them. For example, the incineration process involves a high-temperature, high-pressure environment, and solid waste may undergo complex physical and chemical reactions during incineration, such as changes in temperature, gas composition, and smoke concentration. In this case, it is necessary to enable the temperature and gas sensors and disable other unrelated sensors (such as humidity and weight sensors). In contrast, the composting process has a more stable environment, focusing mainly on the biodegradation process of solid waste under aerobic conditions. Therefore, it is only necessary to enable the temperature and humidity sensors and disable other unrelated sensors (such as weight and gas sensors).

[0056] Furthermore, the data collected by the sensor network is transmitted to the cloud platform through the edge computing unit. During the transmission process, the data collected by the sensor network is processed. The cloud platform stores the processed data in a structured or semi-structured form to form a real-time archive of data information in the solid waste treatment process.

[0057] It should be noted that, due to the inherent characteristics of edge computing units, data redundancy and network transmission pressure can be reduced during transmission;

[0058] It should be explained that in the solid waste treatment scenario, the data collected by the sensors usually contains a lot of redundant information and noise. Although the characteristics of the edge computing unit itself can reduce data redundancy during transmission, it cannot completely eliminate data redundancy. At the same time, since multiple sensors work at different locations at the same time, a large amount of high-dimensional data is generated. Therefore, in order to meet the efficiency prediction of solid waste in each stage of treatment, the data needs to be processed in advance.

[0059] Furthermore, an autoencoder is constructed to arrange the collected data information into a high-dimensional input matrix, and then multiple convolutional layers are used to map the input data information into a low-dimensional space to output low-dimensional data information.

[0060] Furthermore, the low-dimensional data information is backpropagated through the encoder-decoder pairing. The mean square error formula is used to check whether there is an error in the processed data information. If there is no error, the processed data information is output through the bottleneck layer of the encoder, and the low-dimensional features of each input data are extracted. Otherwise, the mapping is re-performed.

[0061] It should be noted that backpropagation refers to reconstructing low-dimensional data information into high-dimensional data information, in order to indicate whether the original data information will be shifted when mapped to a low-dimensional space. The mean square error formula is used to determine whether a shift will occur.

[0062] Specifically, the mean squared error formula is expressed as:

[0063]

[0064] Where, x i For the i-th data information collected, Let n be the i-th data information after backpropagation, n be the total amount of data information, and L be the mean square error value.

[0065] Specifically, the condition for no error is L = 0;

[0066] It should be noted that the numerical data in the processed data is stored as structured data, and the text data is stored as semi-structured data because structured data usually requires rapid querying, statistics, and analysis, while semi-structured data usually requires additional parsing and processing to obtain useful information and is not suitable for direct storage in relational (structured) databases. Therefore, combining structured and unstructured data into a real-time archive of data information in the solid waste treatment process conforms to the data organization method of archives.

[0067] S2. Create a digital twin model based on real-time archives of data information, train the digital twin model, and output the efficiency prediction results of solid waste in each treatment stage.

[0068] It should be explained that a digital twin model is a virtual model used to map and simulate real-world physical objects, systems, or processes. Through digital twins, real-world physical entities can be monitored, predicted, and optimized in a virtual environment in real time. In the present invention, in order to realize the digital monitoring process of solid waste treatment, a digital twin model is used as the basic model of the solution.

[0069] For details, please refer to Figure 2 The digital twin model includes an input layer, an embedding layer, stacked recurrent neural network layers, and a fully connected layer;

[0070] It should be noted that by building a digital twin model, testing and optimization can be carried out without actually operating physical equipment, thereby saving actual equipment costs and maintenance costs.

[0071] Furthermore, the processing steps for each layer in the digital twin model are as follows:

[0072] The system receives data from real-time archives through the input layer and sends the data to the embedding layer. The embedding layer classifies the data and models the temporal dependencies of each stage in the solid waste treatment process. Through stacked recurrent neural network layers, it records the duration and amount of solid waste in each treatment process. The system outputs the efficiency prediction results of solid waste in each treatment stage through the stacked recurrent neural network layers through the fully connected layer.

[0073] It should be noted that different types of data may have different dependencies and processing requirements. Therefore, it is necessary to classify the data to ensure that each type of data can find a suitable time series model to accurately simulate the behavior of solid waste in each processing stage. For example, the data information for the solid waste compression stage includes pressure, temperature, processing time, and processing volume. The pressure of the compression equipment gradually increases over time, reaching its optimal effect within a certain period, and then may enter a stable phase. The temperature may rise with the increase of compression force, thus affecting the processing volume and processing time. Therefore, these data are classified as time series data for the compression stage. Time series modeling, on the basis of the compression stage, means finding the changing trends, dependencies, and patterns of the time series data in the time dimension. Based on the above example, it can be predicted that the pressure data in the compression stage has periodic fluctuations, and the pressure will gradually increase or decrease each time compression depending on the type of solid waste. At this time, time series modeling can identify the pressure changes of solid waste at different time points during processing, thereby adjusting the working strategy of the processing equipment to accurately simulate the behavior of solid waste in the compression stage.

[0074] It should be explained that, since the solid waste treatment process involves multiple stages and complex behavioral changes, and these behavioral changes are continuous in time and have significant temporal dependencies, it is necessary to use the data information after time-series modeling to train the stacked recurrent neural network layers, so as to more accurately predict the efficiency of solid waste in each treatment stage.

[0075] By utilizing the data information after time-series modeling, the stacked recurrent neural network layers are trained until each layer in the stacked recurrent neural network is trained.

[0076] Specifically, the stacked recurrent neural network layer structure and number of layers are as follows:

[0077] The stacked recurrent neural network is divided into long short-term memory network layers and gated recurrent unit layers with equal number of layers. The output of each layer of the stacked recurrent neural network is used as the input of the next layer to form a data recursive transmission mechanism.

[0078] Specifically, stacked recurrent neural networks are divided into equal layers; for example, if the stacked recurrent neural network is set to 50 layers, then the number of layers equally divided into long short-term memory network layers and gated recurrent unit layers is 25:25.

[0079] Specifically, during training, if the amount of data processed by the current layer exceeds the capacity of that layer (i.e., each layer has an initial set data processing capacity), if the current layer is a Long Short-Term Memory (LSTM) network layer, it will be replaced by a gated recursive unit layer; otherwise, no processing will be performed.

[0080] It should be noted that since LSTM is suitable for processing time series data with strong long-term dependencies, while GRU is suitable for data with higher computational efficiency and weaker long-term dependencies, the model will automatically select a more suitable data processing mechanism (using GRU instead of LSTM) based on the size of the data, thereby avoiding the occurrence of computational bottlenecks and improving the adaptability of neural network layers.

[0081] S3. Based on the efficiency prediction results of solid waste in each treatment stage, a multi-objective learning framework is constructed. Through the multi-objective learning framework, the efficiency prediction results are continuously monitored and adjusted to achieve efficiency management of solid waste in each treatment stage.

[0082] It should be explained that a multi-objective learning framework is an optimization strategy used to solve complex problems with multiple objectives or constraints, and is suitable for situations where a balance and trade-off need to be made between multiple objectives.

[0083] Furthermore, by calling up real-time archives of data information during the solid waste treatment process, and treating the data in the real-time archives as a Markov decision process, a multi-objective learning framework is constructed using state space and action space methods respectively.

[0084] It should be noted that the state space refers to all possible states in the solid waste treatment process (such as the current working state of the treatment stage, equipment load, etc.); the action space refers to all possible operations that can be taken in each state (such as adjusting equipment settings, changing the processing order, etc.). Therefore, in the Markov decision process, each stage of solid waste treatment can be regarded as a state, and the model's operation corresponds to an action.

[0085] Specifically, the Markov decision process is expressed mathematically as (S,A,T,R,π), where S is the state space, A is the action space, T is the state transition function, R is the reward function, and π is the adjustment policy.

[0086] Furthermore, the efficiency prediction results of solid waste in each treatment stage are used as the input set of the multi-objective learning framework, and a reward function is designed. The reward factor is increased if the efficiency of solid waste in each treatment stage reaches the expected target, and decreased if it is lower than the expected target.

[0087] Specifically, the reward function is generally represented as R(s, a), where R is the reward value, s is the current state including information such as the working status of the current stage, equipment load, and processing time; and a is the action selected in state s, such as adjusting equipment settings, starting / stopping the stage, etc.

[0088] Specifically, the improved reward function is expressed as:

[0089]

[0090] Among them, E actual E represents the actual efficiency of solid waste treatment at each stage. target α represents the expected target efficiency of solid waste in each treatment stage; α is the reward factor.

[0091] Furthermore, the efficiency prediction results of solid waste in each treatment stage are calculated by weighted average method, and the weighted average is monitored in real time. If the efficiency of solid waste in any treatment stage is lower than the weighted average, the reward factor of the reward function in the multi-objective learning framework is adjusted.

[0092] Specifically, the formula for the weighted average method is expressed as follows:

[0093]

[0094] Among them, W avg This is a weighted average, representing the overall efficiency prediction of solid waste across all treatment stages; E i The efficiency prediction result for the i-th processing stage; w i Let be the weight of the i-th processing step. The larger the weight, the greater the impact of the processing step on the overall processing steps. m is the total number of solid waste processing steps.

[0095] It should be noted that by calculating the weighted average, the overall efficiency prediction results of solid waste can be monitored. Based on the monitoring results, adjustments can be made in a timely manner to the treatment links that deviate from the expectations, thus balancing the potential conflicts caused by deviations in the treatment links and achieving efficient management of solid waste.

[0096] Furthermore, this embodiment also provides a digital monitoring and management system for solid waste treatment, including:

[0097] The solid waste information digitization module is configured to collect data information of solid waste in each treatment stage in real time through sensors, and transmit the data information to the cloud platform through the edge computing unit to form a real-time archive of data information in the solid waste treatment process;

[0098] The solid waste efficiency prediction module is configured to create a digital twin model based on the real-time archive of the data information, train the digital twin model, and output the efficiency prediction results of solid waste in each treatment stage.

[0099] The solid waste efficiency monitoring module is configured to construct a multi-objective learning framework based on the efficiency prediction results of the output solid waste in each treatment stage. Through the multi-objective learning framework, the efficiency prediction results are continuously monitored and adjusted to achieve efficiency management of solid waste in each treatment stage.

[0100] This embodiment also provides a computer device applicable to a digital monitoring and management method for solid waste treatment, including:

[0101] The system includes a memory and a processor. The memory stores computer-executable instructions, and the processor executes these instructions to implement a digital monitoring and management method for solid waste treatment as described in the above embodiments.

[0102] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0103] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements a digital monitoring and management method for solid waste treatment as proposed in the above embodiments.

[0104] The storage medium proposed in this embodiment and the data storage method proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0105] Example 2

[0106] Reference Figure 3This is the second embodiment of the present invention, which provides a digital monitoring and management method for solid waste treatment, including: This embodiment aims to verify the advantages of predicting the efficiency of the solid waste treatment process through a digital twin model and a multi-objective learning framework; To achieve this goal, the experiment will set up two sets of solid waste treatment systems: one set is a traditional treatment system (existing technology), and the other set is a treatment system based on the scheme of the present invention (using a digital twin model and a multi-objective learning framework); The configuration of the experimental environment will ensure that the actual solid waste treatment process can be simulated and the performance can be compared under the same conditions;

[0107] The experimental environment is configured as follows:

[0108] The sensor types include: Temperature sensor: Model XPT-2000, range: -40℃ to 100℃, accuracy: ±0.5℃, used to monitor temperature during stacking, compression, and incineration; Humidity sensor: Model HYG-100, range: 0-100%, accuracy: ±3%, used to monitor humidity changes during composting; Gas sensor: Model GAS-500, supports monitoring gases such as methane, ammonia, and carbon dioxide, range: 0-1000ppm, accuracy: ±2%FS, used for monitoring gas composition during incineration and composting; Pressure sensor: Model PS-600, range: 0-2000kPa, accuracy: ±0.5%FS, used to monitor pressure changes during compression; Weight sensor: Model WEIGHT-800, range: 0-1000kg, accuracy: ±0.1%FS, used to monitor the weight of waste during transportation and stacking.

[0109] Edge computing unit: Model EdgeX3000, has strong data processing capabilities, supports real-time data transmission and preprocessing, and has built-in data compression and noise reduction algorithms to reduce data redundancy during transmission;

[0110] Cloud platform: Deployed on AWS EC2, equipped with virtual servers with 8 vCPUs, 32GB RAM, and 500GB SSD storage, supporting large-scale data processing and real-time analysis, ensuring the efficiency of model training and data storage;

[0111] The experiment was designed with the following solid waste treatment stages: Transportation: Solid waste is transported from the source to the treatment plant via conveyor belt; sensors include temperature, humidity, and weight sensors. Stacking: Solid waste is stacked in a designated area where humidity and temperature fluctuate significantly; humidity and temperature are monitored to ensure the stability of the stacking process. Sorting: Solid waste is sorted according to type; gas sensors are used to monitor for potential hazardous gas leaks. Incineration: Solid waste is treated by incineration; gas and temperature sensors are used to monitor environmental conditions in real time during the treatment process.

[0112] pass Figure 3 (Starting from day 1), it can be seen that the efficiency of the present invention in the transportation process is improved compared to the traditional solution. Figure 3 (Transportation stage) Firstly, it can be seen that the efficiency of the present invention decreases between day 4 and day 5, but improves during the remaining time (except day 18). Furthermore, the efficiency of the present invention is higher than that of the traditional solution for most of the time, especially between day 5 and day 17, showing the most significant improvement in transport efficiency during this period. Secondly, (stacking stage) it can be seen that although the peak efficiency of the present invention is better than that of the traditional solution for most of the time, the efficiency decreases between day 10 and day 11 and between day 28 and day 30. Additionally, (sorting and incineration stages) it is clear that the present invention is superior to the traditional solution.

[0113] In summary, the present invention improves efficiency in all stages of solid waste treatment (for most of the time) compared to traditional solutions, which indirectly demonstrates that the present invention can effectively reduce efficiency decline caused by external interference or equipment fluctuations.

[0114] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take 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. The solutions in the embodiments of this application can be implemented using various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0115] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0116] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0117] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0118] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0119] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A digital monitoring and management method for solid waste treatment, characterized in that, include: Data on solid waste is collected in real time at each stage of the treatment process by sensors, and the data is transmitted to the cloud platform through the edge computing unit to form a real-time archive of data information in the solid waste treatment process. Data on solid waste is collected in real time through sensors at each stage of its treatment process, and this data is transmitted to a cloud platform via a wireless network, including: All sensors are organized into multiple sensor networks in the form of a network topology diagram. Each sensor network corresponds to a solid waste treatment stage. Each sensor in each sensor network represents a node of the sensor network. Nodes in the current sensor network are added or deleted according to the type of solid waste treatment stage. The data collected by the sensor network is transmitted to the cloud platform through the edge computing unit. During the transmission process, the data collected by the sensor network is processed. The cloud platform stores the processed data in a structured or semi-structured form to form a real-time archive of data information in the solid waste treatment process. A digital twin model is created based on the real-time archives of the data information, the digital twin model is trained, and the efficiency prediction results of solid waste in each treatment stage are output. A digital twin model is created based on the real-time archives of the data information. The digital twin model is trained and outputs the efficiency prediction results of solid waste in each treatment stage, including: The digital twin model receives data information from real-time archives through the input layer and sends the data information into the embedding layer. The embedding layer classifies the data information and models the temporal dependencies of each stage in the solid waste treatment process. Through stacked recurrent neural network layers, it records the duration and amount of solid waste in each treatment process. The results of the stacked recurrent neural network layers are output through a fully connected layer. Using the modeled data, the stacked recurrent neural network layers are trained until each layer in the stacked recurrent neural network is trained. The stacked recurrent neural network layers include: The stacked recurrent neural network is divided into long short-term memory network layers and gated recurrent unit layers with equal number of layers. The output of each layer of the stacked recurrent neural network is used as the input of the next layer to form a data recursive transmission mechanism. During training, if the amount of data processed by the current layer exceeds the capacity of the layer, if the current layer is a Long Short-Term Memory network layer, it will be replaced by a gated recursive unit layer; otherwise, no processing will be performed. Based on the efficiency prediction results of solid waste in each treatment stage, a multi-objective learning framework is constructed. Through the multi-objective learning framework, the efficiency prediction results are continuously monitored and adjusted to achieve efficiency management of solid waste in each treatment stage. Based on the efficiency prediction results of output solid waste in each treatment stage, a multi-objective learning framework is constructed, including: The real-time archive of data information during the solid waste treatment process is invoked, and the data in the real-time archive is used as a Markov decision process to construct a multi-objective learning framework in terms of state space and action space respectively. The efficiency prediction results of the solid waste in each treatment stage are used as the input set of the multi-objective learning framework, and a reward function is designed. The reward factor is increased if the efficiency of the solid waste in each treatment stage reaches the expected target, and decreased if it is lower than the expected target.

2. The digital monitoring and management method for solid waste treatment as described in claim 1, characterized in that, Data processing is performed on the data collected by the sensor network, including: An autoencoder is constructed to arrange the collected data information into a high-dimensional input matrix. Multiple convolutional layers are used to map the input data information to a low-dimensional space and output low-dimensional data information. The encoder-decoder pairing performs backpropagation on the low-dimensional data information. The mean square error formula is used to check whether there are errors in the processed data information. If there are no errors, the processed data information is output through the bottleneck layer of the encoder, and the low-dimensional features of each input data are extracted. Otherwise, the mapping is performed again.

3. The digital monitoring and management method for solid waste treatment as described in claim 1, characterized in that, The efficiency prediction results are continuously monitored and adjusted using the multi-objective learning framework, including: The efficiency prediction results of solid waste in each treatment stage are calculated by weighted average method. The weighted average is monitored in real time. If the efficiency of solid waste in any treatment stage is lower than the weighted average, the reward factor of the reward function in the multi-objective learning framework is adjusted.

4. A digital monitoring and management system for solid waste treatment, based on the digital monitoring and management method for solid waste treatment according to any one of claims 1 to 3, characterized in that, include: The solid waste information digitization module is configured to collect data information of solid waste in each treatment stage in real time through sensors, and transmit the data information to the cloud platform through the edge computing unit to form a real-time archive of data information in the solid waste treatment process; The solid waste efficiency prediction module is configured to create a digital twin model based on the real-time archive of the data information, train the digital twin model, and output the efficiency prediction results of solid waste in each treatment stage. The solid waste efficiency monitoring module is configured to construct a multi-objective learning framework based on the efficiency prediction results of the output solid waste in each treatment stage. Through the multi-objective learning framework, the efficiency prediction results are continuously monitored and adjusted to achieve efficiency management of solid waste in each treatment stage.

5. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 3.

Images