Main Modules Research Articles

Design and Implementation of Beam Control System Based on FPGA

As the use of phased array technology becomes more widespread, it is receiving increased attention. Among them, the excellent performance of the beam control system will directly affect the performance of the whole phased array. FPGA is the preferred control chip for developing a beam control system. Therefore, this paper designs a beam control system architecture based on FPGA, designs the main functional modules of the system in detail, and conducts simulation experiments on its important functions. The results show that the system can quickly and accurately calculate the phase shift code required by the antenna unit to achieve fast beam response.

Design of Home Environment Monitoring System Based on a Cloud Platform for the Elderly Living Alone

The phenomenon of elderly people living alone was becoming increasingly common in the new era, so the demand for monitoring the living environment of elderly people living alone was becoming increasingly strong. The design of home environment monitoring for elderly people living alone was designed based on a cloud platform. The system mainly consists of an STM32 main control module, a temperature and humidity detection module, an Internet of Things module, a security module, a relay control module, and a user App terminal. Then the fuzzy PID algorithm was designed to remotely control indoor temperature and humidity. After practical operation verification, the system can perform local and remote monitoring of the living environment of elderly people living alone and has the characteristics of reliability and stability.

Augmented Reality Based Indoor Navigation System

Abstract: Every people, at some point in their life, may get lost inside a large auditorium, struggled to find their exit point at the airport, or may get late to a lecture because they could not find the right lecture hall at the university. This paper addresses the development of an Augmented Reality Based Indoor Navigation Application. It is based on four main modules, 3D model building, shortest path calculation, QR code scanning and path visualization. The application is developed in Unity using AR Core XR plugin management and AR Foundation plugin. ZXing library is used for scanning and decoding the QR codes. We aim at developing the front end in the simplest way possible so that the users can easily reach their destination by just opening the camera where the directions are shown as animations in their surroundings.

METTL3-Mediated STING Upregulation and Activation in Kupffer Cells Contribute to Radiation-Induced Liver Disease via Pyroptosis

Purpose: Radiation therapy is a vital adjuvant treatment for liver cancer, although the challenge of radiation-induced liver diseases (RILDs) limits its implementation. Kupffer cells (KCs) are a crucial cell population of the hepatic immune system and their biological function can be modulated by multiple epigenetic RNA modifications, including N6-methyladenosine (m6A) methylation. However, the mechanism for m6A methylation in KC-induced inflammatory response in RILD remains unclear. The present study investigated the function of m6A modification in KCs contributing to RILD. Methods and Materials: Methylated RNA-immunoprecipitation sequencing (MeRIP-seq) and RNA transcriptome sequencing were used to explore the m6A methylation profile of primary KCs isolated from mice after irradiation with 3 × 8 Gy. Western blotting and quantitative real-time polymerase chain reaction were used to evaluate gene expression. DNA pull-down and CHIP assays were performed to verify target gene binding and identify binding site. Results: MeRIP-seq revealed a significantly increased m6A modification level in human KCs after irradiation, suggesting the potential role of upregulated m6A in RILD. In addition, the study results corroborated that methyltransferase-like 3 (METTL3) acts as a main modulator to promote the methylation and gene expression of TEAD1, leading to STING-NLRP3 signaling activation. Importantly, it was shown that IGF2BP2 functions as an m6A “reader” to recognize methylated TEAD1 mRNA and promote its stability. METTL3/TEAD1 knockdown abolished the activation of STING-NLRP3 signaling and protected against RILD in addition to suppressing inflammatory cytokines and hepatocyte apoptosis. Moreover, clinically collected human normal liver tissue samples post-irradiation showed increased expression of STING and IL-1β in KCs compared to the non-irradiation group. Notably, STING pharmacological inhibition alleviated irradiation-induced liver injury in mice, indicating its potential therapeutic role in RILD. Conclusions: The results of our study reveal that TEAD1-STING-NLRP3 signaling activation contributes to RILD via METTL3-dependent m6A modification.

Smart City Information Systems: Research on Information Published for Citizens and Design of Effective Content in the Czech Republic

The concept of Smart Cities integrates innovative technologies to improve citizens’ quality of life in towns and cities worldwide. Crisis management is a separate section directly managed by the leadership of municipalities, cities and counties in cooperation between police, fire and municipal police to ensure the safety of residents and safety in public spaces. The purpose of this study is to investigate to which extent publicly available information related to the field of crisis management is unavailable to residents in municipalities, towns and cities through online information systems. The primary aim is to provide suggestions for a general information system structure and content that would highlight and satisfy the need to address the crisis management issue, especially in providing immediate information to the population through an innovative online form. The achievement of this goal is methodologically based on qualitative research analysing and comparing the information published for residents through Smart City information systems in selected towns and municipalities. Document analysis or conceptual design was applied, and evaluation criteria for objective assessment of Smart City information systems were appropriately determined. The comparative analysis based on this set of criteria enabled the development of the proposals of information systems’ content that can be used to keep the information systems for Smart Cities in cities, municipalities and regions, actual and beneficial. From the available resources, two main modules that focused either on citizens or cities were synthesised. Moreover, SWOT analysis or the Smart Regions Rapid Response structure was derived. Acquired results outline generic structures and contents that support the development of the concept of Smart Cities and can be suitably implemented for the development of the modification of information systems containing relevant information for residents, cities and municipalities, focusing on citizen safety.

Decision support in engineering design: the ELIGERE open source software platform

In engineering design, the selection of the optimal design solution represents a critical phase for the development of successful products. In this paper, we present ELIGERE, an open source decision support system targeted at engineering design applications. It allows to rank multiple design solutions with respect to different evaluation criteria according to the evaluations provided by a group of experts. ELIGERE is composed by three main modules: (1) a distributed web application, for generation and participation to the decision making session; (2) a mathematical engine, based on the fuzzy analytical hierarchy process, to quantify the results of the decision making session according to the evaluation of the experts; (3) a relational database, to collect and store data. The most important contribution of this paper is introducing a practical and effective software tool that facilitates decision-making analysis based on the fuzzy analytical hierarchy process, thereby allowing better-informed choices on concept selection, as it has been designed with a specific focus on the engineering field. In this paper we describe the key concepts of ELIGERE and its modalities of use in several real use cases. Finally, we compare ELIGERE with the widely used general purpose decision support software based on the fuzzy analytical hierarchy process.

MATT-DDI: Predicting multi-type drug-drug interactions via heterogeneous attention mechanisms

The joint use of multiple drugs can result in adverse drug-drug interactions (DDIs) and side effects that harm the body. Accurate identification of DDIs is crucial for avoiding accidental drug side effects and understanding potential mechanisms underlying DDIs. Several computational methods have been proposed for multi-type DDI prediction, but most rely on the similarity profiles of drugs as the drug feature vectors, which may result in information leakage and overoptimistic performance when predicting interactions between new drugs. To address this issue, we propose a novel method, MATT-DDI, for predicting multi-type DDIs based on the original feature vectors of drugs and multiple attention mechanisms. MATT-DDI consists of three main modules: the top k most similar drug pair selection module, heterogeneous attention mechanism module and multi‑type DDI prediction module. Firstly, based on the feature vector of the input drug pair (IDP), k drug pairs that are most similar to the input drug pair from the training dataset are selected according to cosine similarity between drug pairs. Then, the vectors of k selected drug pairs are averaged to obtain a new drug pair (NDP). Next, IDP and NDP are fed into heterogeneous attention modules, including scaled dot product attention and bilinear attention, to extract latent feature vectors. Finally, these latent feature vectors are taken as input of the classification module to predict DDI types. We evaluated MATT-DDI on three different tasks. The experimental results show that MATT-DDI provides better or comparable performance compared to several state-of-the-art methods, and its feasibility is supported by case studies. MATT-DDI is a robust model for predicting multi-type DDIs with excellent performance and no information leakage.

RF-RFE-SMOTE: A DoS And DDoS Attack Detection Framework

Denial of service and Distributed denial of service (Dos/DDos) attacks continue to be one of the most significant dangers in cybersecurity. Many efforts are being put into developing defenses against these types of attacks. The tools used by attackers to perform these types of attacks increase day-to-day. Thus, a countermeasure is necessary. For this reason, this thesis utilized one of the most recent datasets (CSE-CICIDS2018 and CIC-DDoS2019) containing most Dos/DDoS attacks. This study proposed a framework based on Machine Learning for detecting denial-of-service (DoS) and distributed denial-of-service (DDoS) attacks. The framework comprises three main modules: feature selection method using Random Forest—Recursive Feature Elimination (RF-RFE), handling the Imbalanced class distributions using Synthetic Minority Oversampling Technique (SMOTE), and classification. This study used five classifiers to make comparisons that include Random Forest (RF), Naive Bayes (NB), Logistic Regression (LR), and Linear and Quadratic Discriminant Analysis (LDA, QDA)”. Framework empirical findings reveal that the RF-RFE_SMOTE_RF outperformed all other models by obtaining an accuracy of 100% for CSE-CIC-IDS2018 and 0.99% for CIC-DDoS2019.

Deep image clustering with contrastive learning and multi-scale graph convolutional networks

Deep clustering has shown its promising capability in joint representation learning and clustering via deep neural networks. Despite the significant progress, the existing deep clustering works mostly utilize some distribution-based clustering loss, lacking the ability to unify representation learning and multi-scale structure learning. To address this, this paper presents a new deep clustering approach termed Image clustering with contrastive learning and multi-scale Graph Convolutional Networks (IcicleGCN), which bridges the gap between convolutional neural network (CNN) and graph convolutional network (GCN) as well as the gap between contrastive learning and multi-scale structure learning for the deep clustering task. Our framework consists of four main modules, namely, the CNN-based backbone, the Instance Similarity Module (ISM), the Joint Cluster Structure Learning and Instance reconstruction Module (JC-SLIM), and the Multi-scale GCN module (M-GCN). Specifically, the backbone network with two weight-sharing views is utilized to learn the representations for the two augmented samples (from each image). The learned representations are then fed to ISM and JC-SLIM for joint instance-level and cluster-level contrastive learning, respectively, during which an auto-encoder in JC-SLIM is also pretrained to serve as a bridge to the M-GCN module. Further, to enforce multi-scale neighborhood structure learning, two streams of GCNs and the auto-encoder are simultaneously trained via (i) the layer-wise interaction with representation fusion and (ii) the joint self-adaptive learning. Experiments on multiple image datasets demonstrate the superior clustering performance of IcicleGCN over the state-of-the-art. The code is available at https://github.com/xuyuankun631/IcicleGCN.

Learn to Few-Shot Segment Remote Sensing Images from Irrelevant Data

Few-shot semantic segmentation (FSS) is committed to segmenting new classes with only a few labels. Generally, FSS assumes that base classes and novel classes belong to the same domain, which limits FSS’s application in a wide range of areas. In particular, since annotation is time-consuming, it is not cost-effective to process remote sensing images using FSS. To address this issue, we designed a feature transformation network (FTNet) for learning to few-shot segment remote sensing images from irrelevant data (FSS-RSI). The main idea is to train networks on irrelevant, already labeled data but inference on remote sensing images. In other words, the training and testing data neither belong to the same domain nor category. The FTNet contains two main modules: a feature transformation module (FTM) and a hierarchical transformer module (HTM). Among them, the FTM transforms features into a domain-agnostic high-level anchor, and the HTM hierarchically enhances matching between support and query features. Moreover, to promote the development of FSS-RSI, we established a new benchmark, which other researchers may use. Our experiments demonstrate that our model outperforms the cutting-edge few-shot semantic segmentation method by 25.39% and 21.31% in the one-shot and five-shot settings, respectively.

The interplay between kisspeptin and endocannabinoid systems modulates male hypothalamic and gonadic control of reproduction in vivo.

Male reproduction is under the control of the hypothalamus-pituitary-gonadal (HPG) axis. The endocannabinoid system (ECS) and the kisspeptin system (KS) are two major signaling systems in the central and peripheral control of reproduction, but their possible interaction has been poorly investigated in mammals. This manuscript analyzes their possible reciprocal modulation in the control of the HPG axis. Adolescent male rats were treated with kisspeptin-10 (Kp10) and endocannabinoid anandamide (AEA), the latter alone or in combination with the type 1 cannabinoid receptor (CB1) antagonist rimonabant (SR141716A). The hypothalamic KS system and GnRH expression, circulating sex steroids and kisspeptin (Kiss1) levels, and intratesticular KS and ECS were evaluated by immunohistochemical and molecular methods. Non-coding RNAs (i.e., miR145-5p, miR-132-3p, let7a-5p, let7b-5p) were also considered. Circulating hormonal values were not significantly affected by Kp10 or AEA; in the hypothalamus, Kp10 significantly increased GnRH mRNA and aromatase Cyp19, Kiss1, and Kiss1 receptor (Kiss1R) proteins. By contrast, AEA treatment affected the hypothalamic KS at the protein levels, with opposite effects on the ligand and receptor, and SR141716A was capable of attenuating the AEA effects. Among the considered non-coding RNA, only the expression of miR145-5p was positively affected by AEA but not by Kp10 treatment. Localization of Kiss1+/Kiss1R+ neurons in the arcuate nucleus revealed an increase of Kiss1R-expressing neurons in Kp10- and AEA-treated animals associated with enlargement of the lateral ventricles in Kp10-treated animals. In the brain and testis, the selected non-coding RNA was differently modulated by Kp10 or AEA. Lastly, in the testis, AEA treatment affected the KS at the protein levels, whereas Kp10 affected the intragonadal levels of CB1 and FAAH, the main modulator of the AEA tone. Changes in pubertal transition-related miRNAs and the intratesticular distribution of Kiss1, Kiss1R, CB1, and CB2 following KP and AEA treatment corroborate the KS-ECS crosstalk also showing that the CB1 receptor is involved in this interplay. For the first time in mammals, we report the modulation of the KS in both the hypothalamus and testis by AEA and revealed the KP-dependent modulation of CB1 and FAAH in the testis. KP involvement in the progression of spermatogenesis is also suggested.

Association of intestinal dysbiosis with susceptibility to multiple sclerosis: Evidence from different population studies (Review).

Understanding the relationship between microorganisms that live in our intestines and neuroinflammatory and neurodegenerative pathologies of the central nervous system (CNS) is essential, since they have been shown to have an immunomodulatory effect in neurological disorders, such as multiple sclerosis (MS). The gut microbiota can be affected by several environmental factors, including infections, physical and emotional stress and diet, the latter known as the main modulator of intestinal bacteria. An abrupt shift in the gut microbiota composition and function is known as dysbiosis, a state of local and systemic inflammation produced by pathogenic bacteria and its metabolites responsible for numerous neurological symptoms. It may also trigger neuronal damage in patients diagnosed with MS. Intestinal dysbiosis affects the permeability of the intestine, allowing chronic low-grade bacterial translocation from the intestine to the circulation, which may overstimulate immune cells and cells resident in the CNS, break immune tolerance and, in addition, alter the permeability of the blood-brain barrier (BBB). This way, toxins, inflammatory molecules and oxidative stress molecules can pass freely into the CNS and cause extensive damage to the brain. However, commensal bacteria, such as the Lactobacillus genus and Bacteroides fragilis, and their metabolites (with anti-inflammatory potential), produce neurotransmitters such as γ-aminobutyric acid, histamine, dopamine, norepinephrine, acetylcholine and serotonin, which are important for neurological regulation. In addition, reprogramming the gut microbiota of patients with MS with a healthy gut microbiota may help improve the integrity of the gut and BBB, by providing clinically protective anti-inflammatory effects and reducing the disease's degenerative progression. The present review provides valuable information about the relationship between gut microbiota and neuroinflammatory processes of the CNS. Most importantly, it highlights the importance of intestinal bacteria as an environmental factor that may mediate the clinical course of MS, or even predispose to the outbreak of this disease.

Detailed Energy Analysis of a Sheet-Metal-Forming Press from Electrical Measurements

This paper presents a methodology that allows for the detection of the state of a sheet-metal-forming press, the parts being produced, their cadence, and the energy demand for each unit produced. For this purpose, only electrical measurements are used. The proposed analysis is conducted at the level of the press subsystems: main motor, transfer module, cushion, and auxiliary systems, and is intended to count, classify, and monitor the production of pressed parts. The power data are collected every 20 ms and show cyclic behavior, which is the basis for the presented methodology. A neural network (NN) based on heuristic rules is developed to estimate the press states. Then, the production period is determined from the power data using a least squares method to obtain normalized harmonic coefficients. These are the basis for a second NN dedicated to identifying the parts in production. The global error in estimating the parts being produced is under 1%. The resulting information could be handy in determining relevant information regarding the press behavior, such as energy per part, which is necessary in order to evaluate the energy performance of the press under different production conditions.

Mathematical Simulation in Nuclear Medicine for Optimization Diagnostic Accuracy of SPECT/CT

Purpose: To develop the method of mathematical simulation in the field of nuclear medicine and its
 practical application in research aimed at improving the diagnostic accuracy of the SPECT/CT method.
 The basic principles of the method of mathematical simulation in nuclear medicine and the main modules of the software package “Virtual Platform for Simulation Tests of the SPECT/CT Method”, created
 at the laboratory for simulation in nuclear medicine of Novosibirsk State University and the Institute of
 Theoretical and Applied Mechanics.
 Material and methods: The main principles of the developed software package are the realism of the ‘virtual patient’ mathematical model, the accuracy in modeling of physical processes in visualization, as
 well as simulation conducting close to clinical practice. All studies were carried out in collaboration
 with clinicians. The software complex was verified by comparison with clinical data. Numerical experiments have shown a close correspondence between the results of clinical and virtual studies. In simulation tests, the same errors were obtained on images that were observed on clinical images. Simulation
 computer tests were carried out using the software package “Virtual Platform...” in the field of nuclear
 oncology, cardiology and neurology, aimed at studying the accuracy of images of pathological foci to assess the possibility of switching to quantitative SPECT, as well as optimizing the protocol for examining
 patients.
 Results: Simulation tests have shown that in order to move to quantitative SPECT, it is not enough to be
 limited to the development of a methodology for calibrating SPECT systems for a particular radionuclide. There are problems without understanding and solving which it is impossible to talk about the diagnostic accuracy of the quantitative SPECT method. Such problems are edge artifacts that appear on
 the images of pathological foci, and the rule of stopping the iterative algorithm when the values of activity in the area of interest (pathological focus) that are closest to the true values are reached. A stopping
 criterion based on Pearson's modified chi-square test of goodness of fit is proposed. There is a significantly good correlation between the proposed stop criterion and the minimum RMS error of the image
 reconstruction. The results of simulation modeling with the aim of optimizing SPECT examinations of
 brain perfusion demonstrated the possibility of reducing the data recording time compared to the standard protocol by at least two times.
 Conclusion: The computer simulation method presented in this paper is a practical technology that
 contributes to the optimization and development of a quantitative SPECT method to achieve the best
 possible results in the imaging of pathological foci. Further development of the software package is focused on applications in radionuclide therapy and theranostics.

P3SNet: Parallel Pyramid Pooling Stereo Network

In autonomous driving and advanced driver assistance systems (ADAS), stereo matching is a challenging research topic. Recent work has shown that high-accuracy disparity maps can be obtained with end-to-end training with the help of deep convolutional neural networks from stereo images. However, many of these methods suffer from long run-time for real-time studies. Therefore, in this paper, we introduce P3SNet, which can generate both real-time results and competitive disparity maps to the state-of-the-art. P3SNet architecture consists of two main modules: parallel pyramid pooling and hierarchical disparity aggregation. The parallel pyramid pooling structure makes it possible to obtain local and global information intensively from its multi-scale features. The hierarchical disparity aggregation provides multi-scale disparity maps by using a coarse-to-fine training strategy with the help of the costs obtained from multi-scale features. The proposed approach was evaluated on several benchmark datasets. The results on all datasets showed that the proposed P3SNet achieved better or competitive results while having lower runtime. The code is available at https://github.com/aemlek/P3SNet.

High impedance fault detection device based on edge artificial intelligence

The current high impedance fault (HIF) detection devices are mainly based on threshold method, the hardware cannot meet the computing requirements of artificial intelligence (AI), and the software structure is difficult to adapt to the upgrade requirements of AI. These AI detection algorithms are mainly based on simulation analysis and laboratory testing and have not yet been applied practically. The paper develops an edge AI-based HIF detection device, it combines the main body and independent AI module. The AI module embeds an edge identification algorithm and shares the sampling data with the main body. The detection result is achieved from intelligent classifier in edge AI module and fed back to the main device. Finally, the device is tested on the Luohe distribution network true test platform, the detection accuracy for multiple types of fault is 98.6%.

Question answering over knowledge graphs using BERT based relation mapping

AbstractA knowledge graph (KG) is a structured form of knowledge describing real‐world entities, properties and relationships as a graph. Question answering over knowledge graphs (KGQA) allows people to ask questions in natural language and extract answers from KG accurately and more quickly. The main task of a KGQA is to convert a natural language query to the corresponding structured query form like SPARQL. However, generating the precise SPARQL query from a question is challenging and highly error‐prone. Here we propose a question‐answering framework that uses KG to answer simple questions without using SPARQL. Question classification, dependency parsing, entity linking, BERT‐based relation finding and answer extraction constitute the main modules of the approach. We have used the DBpedia as the KG and tested the end‐to‐end system with a subset of QALD‐4, LC‐QuAD and SimpleQuestions datasets. Results show considerable improvement compared to other approaches in terms of F1‐score.

A Dual-branch Learning Model with Gradient-balanced Loss for Long-tailed Multi-label Text Classification

Multi-label text classification has a wide range of applications in the real world. However, the data distribution in the real world is often imbalanced, which leads to serious long-tailed problems. For multi-label classification, due to the vast scale of datasets and existence of label co-occurrence, how to effectively improve the prediction accuracy of tail labels without degrading the overall precision becomes an important challenge. To address this issue, we propose A Dual-Branch Learning Model with Gradient-Balanced Loss (DBGB) based on the paradigm of existing pre-trained multi-label classification SOTA models. Our model consists of two main long-tailed module improvements. First, with the shared text representation, the dual-classifier is leveraged to process two kinds of label distributions; one is the original data distribution and the other is the under-sampling distribution for head labels to strengthen the prediction for tail labels. Second, the proposed gradient-balanced loss can adaptively suppress the negative gradient accumulation problem related to labels, especially tail labels. We perform extensive experiments on three multi-label text classification datasets. The results show that the proposed method achieves competitive performance on overall prediction results compared to the state-of-the-art methods in solving the multi-label classification, with significant improvement on tail-label accuracy.

Engagement of atrial fibrillation patients with the AF-EduApp, a new mobile application to support AF management.

A multidisciplinary approach is needed for the management of atrial fibrillation (AF) in which the patient has a central role. Smart devices create opportunities to improve AF management. This paper aimed to evaluate the in-house developed AF-EduApp application on its usability, satisfaction, and communication effectiveness with the care team. During a multicenter, prospective randomized controlled trial, 153 AF patients were included in the AF-EduApp study, with a minimum follow-up of 12 months and a maximum follow-up of 15 months if taking oral anticoagulation (OAC). The AF-EduApp contains six main modules: Questionnaires, Education, Measurement data entry, Medication overview with reminders, Appointments, and Communication with the care team. The App focuses on four main goals: (1) to improve AF knowledge, (2) to increase self-care capabilities, (3) electronic monitoring to improve therapy adherence to OAC, and (4) communication with the care team. Patients unable to use the AF-EduApp were assigned to a no-App control group (n = 41) without intervention comparable to the standard care group (SC, n = 346) of the AF-EduCare study. A total of 152 patients effectively used the App during a mean follow-up of 386.8 ± 108. 1 days (one included patient could not install the application due to an iPhone from the United States). They opened the application on average on 130.1 ± 144.7 days. Of the 109 patients still in follow-up after 12 months (i.e. patients who did not withdraw and on OAC), 90 patients (82.6%) actively used the application at least one day in the next 41 days. The Measurement module was the most used, with a median of used days over the total available days of 6.4%. A total of 75 App patients (49.3%) asked questions, mostly clinical-related questions (e.g. medication use, or actionability on clinical entered parameters). A mean score of 8.1 ± 1.7 about the "perceived quality of follow-up in the past year" was given by the App ITT patients, compared to a score of 7.7 ± 2.0 by the SC group (P = .072). Patients who used the App were more attracted to future follow-up with an application compared to patients who would be capable of using the application of the SC group (31.6% vs. 12.5%; P < .001). This study showed a positive attitude towards using a mobile application, with AF patients using the application one-third of the available days. Patients used the App most for entering measured parameters, and to contact the care team.

Application of Multivariate Reinforcement Learning Engine in Optimizing the Power Generation Process of Domestic Waste Incineration

Garbage incineration is an ideal method for the harmless and resource-oriented treatment of urban domestic waste. However, current domestic waste incineration power plants often face challenges related to maintaining consistent steam production and high operational costs. This article capitalizes on the technical advantages of big data artificial intelligence, optimizing the power generation process of domestic waste incineration as the entry point, and adopts four main engine modules of Alibaba Cloud reinforcement learning algorithm engine, operating parameter prediction engine, anomaly recognition engine, and video visual recognition algorithm engine. The reinforcement learning algorithm extracts the operational parameters of each incinerator to obtain a control benchmark. Through the operating parameter prediction algorithm, prediction models for drum pressure, primary steam flow, NOx, SO2, and HCl are constructed to achieve short-term prediction of operational parameters, ultimately improving control performance. The anomaly recognition algorithm develops a thickness identification model for the material layer in the drying section, allowing for rapid and effective assessment of feed material thickness to ensure uniformity control. Meanwhile, the visual recognition algorithm identifies flame images and assesses the combustion status and location of the combustion fire line within the furnace. This real-time understanding of furnace flame combustion conditions guides adjustments to the grate and air volume. Integrating AI technology into the waste incineration sector empowers the environmental protection industry with the potential to leverage big data. This development holds practical significance in optimizing the harmless and resource-oriented treatment of urban domestic waste, reducing operational costs, and increasing efficiency.