Abstract
Water Supply issues a formal retraction in relation to the above article by Shu Xu, Ching-Hsien Hsu and Carlos Enrique Montenegro-Marin. The publisher issues this retraction due to concerns regarding potential citation manipulation due to inappropriate references, as well as issues regarding the validity of data and figures used in the article. The authors did not provide a response to our queries during the investigation. Since the journal did not receive a response to these concerns, the Editors-in-Chief no longer have confidence in the integrity of the article.
HIGHLIGHT
Restoring ecosystems is a promising way to reverse human-caused habitat loss and provide new homes for a wide range of wildlife species.
ARTIFICIAL INTELLIGENCE-BASED ENVIRONMENTAL DECISION RESTORATION
IoT devices can enhance air quality, reduce energy use, and reduce waste. There are several applications, including smart water management and smart agriculture. Energy data may be collected and analyzed to help improve delivery and efficiency in a wide range of sectors by improving connectivity and reducing energy waste. The IoT provides more user control of energy demand through mobile devices. The ongoing urban policy has led to a considerable population increase in metropolitan areas and accelerated the development of new places. The risk of extending the growing regions and their penetration into natural ecosystems hindered sustainable development in cities (Dong et al. 2020; Colsoul et al. 2021). In particular, the role of urban communities in regional development has become more important. Environmental challenges such as decreased biodiversity land and water scarcity challenge the critical form of modern urban development (Sundarasekar et al. 2019). The expectation of structurally stable, sustainable, and high-quality urban cluster development in urban cluster financial management is becoming a global priority for scientists and policymakers (Gann et al. 2019; Reyes-García et al. 2019).
The organic retreat should describe natural ecosystems' integrity and well-being and their long-term potential to biologically preserve and restore the environment (Wen et al. 2019). Compared to the concepts, such as an urban boundary and the border of urban development, the theoretical basis for biological retreat configuration is the land-based environmental theory, which focuses on the spatial (or deliberate) relationship between critical reinforcements as a channel (Ma et al. 2019; Dar & Bhat 2020). The extraction of new elements and the definition of green channel constitute two essential phases in the growth of organic retreat structure (Jia et al. 2018). The ecological sources of natural habitat parks are classified using two critical strategies (Sheng et al. 2019). First, the national forests, scenic areas, and large landscapes are collected (Le et al. 2019). The second is based on the ‘import response communication paradigm,’ which is more fitting because it focuses on the environment's structural and functional aspects (Manogaran et al. 2021; Shalaginov et al. 2021). To identify the ecological channel, there are a variety of strategies needed.
Land loss is a reduction or absence of biological land activities, largely because of a confrontation with human participation in natural processes (Menouer et al. 2020; Yin et al. 2021). This deterioration is due to land decimation, soil degradation, water pollution, and limited economic output (Khelifi et al. 2020). The destruction will contribute to the ecosystem and habitat loss and ultimately place human well-being at risk (Usman et al. 2020; Shen et al. 2021). Consequently, protecting land resources and critical biodiversity has been important and the source of much environmental management (Lokesh et al. 2019). Ecological restoration is the return of a destroyed habitat to its original form (VE & Cho 2020; Xu et al. 2021). Land restoration includes socioeconomic and environmental experience to recognize causes and mechanisms for degradation and studies methods and alternatives to maintaining and regenerating the damaged ecosystem (Ezhilmaran & Adhiyaman 2017). The final objective is to secure the terrestrial environment and guarantee the economy's social and economic survival (Yu et al. 2021). No large-scale research was done on the restoration approaches that led to a blindness of the land mechanism and unexpected environmental mechanisms that involved land degradation (Meng et al. 2021).
Moreover, no local stakeholder engagement proposals for regeneration indicate that these initiatives are democratically acceptable (Giorio et al. 2019). The original ecosystem is frequently reconstructed ecologically, and the researchers have not properly portrayed local environmental degradation and public safety threats (Giorio et al. 2019; Yaqub et al. 2020). This study is a part of a larger effort to provide a comprehensive capacity for assessing regional hydroclimate data, emphasizing feature-specific indicators and stakeholder-relevant results. Building and adapting a standardized, objective, measurable, repeatable, and trustworthy framework is a key part of this endeavor (Srivastava et al. 2019).
Ecological regeneration risk factors include frequent changes in natural environments, imperfect interpretation of natural systems by humans, and a lack of knowledge of past successes and shortcomings. In this paper, artificial intelligence-based environmental decision restoration framework (AI-EDRF) has been proposed to strengthen the continuously evolving natural systems; people are deficient in biological systems and the insufficient knowledge about past achievements and failures.
IoT systems in power generation include smart grids, smart electric meters, home automation systems, and industrial automation. These activities aim to reduce the amount of energy used in useful ways. Energy efficiency in commercial and industrial buildings is part of a process that includes planning and managing energy consumption. Using IoT devices, including smart thermostats and lighting systems, facility managers can keep track of a building's real-time energy use and adjust various electrical equipment usage schedules to lessen demand during peak hours. The energy management solution gives the user comprehensive control over the energy data at the fundamental and granular levels while lowering energy costs.
The main contribution of the article:
Restoring ecosystems is a promising way to reverse human-caused habitat loss and provide new homes for many wildlife species.
AI applications might assist in constructing other energy-efficient structures, improving power storage, and optimizing the deployment of renewable energy sources.
A battery-powered sensor, an actuator, and a communication system make up the IoT device's hardware. A sensor's main role is to gather information from its environment.
IoT sensors may process data to learn paths that save a lot of time and energy while also saving a massive amount of time.
The flow of this research is organized in the following way. The literature survey details are discussed in Section 2. Section 3 details the ecological restoration of the water body sustainability, and Section 4 describes the proposed AI-EDRF. Section 5 provides an in-depth discussion of real-time analysis and its outcomes. Section 6 includes research outcomes and outlines the future scope.
LITERATURE SURVEY
AI-EDRF has been proposed to strengthen the continuously evolving natural systems; people are deficient in natural systems and insufficient knowledge available about past achievements and failures. The biological terrestrial collective analysis introduced to improve natural systems is rapidly evolving, and people are inadequately aware of natural systems. Stochastic water quality analysis is integrated with AI-EDRF to boost past achievements and failures.
Xu et al. (2020) used the linear programming (LP) method of optimization that is then applied to the resource allocation and sharing (RAS) restrictions. Genetic representations (GR) work around the limits to lessen the number of unfulfilled requests and misallocated assets. The LP model inherits the conventional genetic phases to address resource allocation and access problems, thereby decreasing latency. LP and GR break the impasse in cloud network MCA and resource allocation when applied together. In a nutshell, LP-GR accurately estimates response times, resource utilization, requests processed, and average latency.
Previous research has expanded on climate restoration, forestry, social society, and terrestrial ecological processes (Govindan et al. 2021). Land restoration includes recovery of trees, drainage of wastewater, degradation of surfaces and salt groundwater, and desertification management (Stefanakis 2019). Agricultural land reform, urban land recovery, forest management, and soil depletion prevention are needed for economic development (Lin et al. 2020). The state of terrestrial deterioration was assessed by mono-factor studies and cannot include incentives for taking action for complete restoration (Moore-O'Leary et al. 2017).
A terrestrial habitat regional hazard assessment can assess the risk for environmental loss from human and natural activity. It provides useful expertise for environmental decision-making to support climate change, strategic planning, and even regional market activities (Tibebe et al. 2019). This approach's main advantage is that it offers an enhanced connection between environmental and manufacturing industries (Perring et al. 2018) and reveals the evolving patterns of ecological damage to determine the system.
Riparian biodiversity acts as a barrier that extracts sediments, contaminants, and contaminants otherwise accessible to streams; it helps maintain banks, enhances flood protection, provides aquatic and terrestrial habitats, and acts as an ecological network between habitats (Reaser et al. 2021). Steep slopes are extremely landslip prone and lead to a rise in erosion and export of sediments to the water basin, so vegetations are vital to maintaining stabilization in the slopes due to the absorption of root water and, in particular, the strengthening of plant root natural ecosystems (Jiang et al. 2018).
Sustainable and environmental management and evaluation building application systems for a water supply will provide an operating platform and policy climate for unified water control and conservation of the environment (Li et al. 2017). Developing a water body sustainability assessment scheme allowed qualitative and quantitative water body quality measurements (Bustamante et al. 2019). The system will monitor the sustainability of the water body by the state-of-the-art automated data collection and data transfer systems, evaluate ecological impacts, and determine the risk of plant protection stress in mainstream areas in various incoming glasses of water by integrating an ecological assessment model system (Xu et al. 2019).
Liu et al. (2020) prepared that desertification seriously threatens global sustainable development. Effective monitoring and understanding of its driving causes are essential for prevention and rehabilitation. Ecological restoration programs have a critical role in halting desertification, especially in quantifying their impact. Based on this, we recommend that we continue to execute ecological restoration policies to preserve desertification restoration.
Guo et al. (2020) detailed that ecological restoration of terrestrial ecosystems is critical in safeguarding the environment and sustainably using land resources. Restoration of ecological functions is becoming increasingly difficult as land degradation worsens. Using a non-linear model of ecological interactions, the research area was separated into six sub-regions, and the ecological hazards were described. Multiple-criteria decision analysis was used to evaluate many stakeholders in the situation. An integrated evaluation utilizing the preference was completed to determine the best possible outcome.
Constant & Taylor (2020) explained that ecosystem service frameworks and planning procedures sometimes fail to incorporate indigenous perspectives, requiring knowledge co-production based on various perspectives. Rural and urban communities have a shared goal of restoring biodiversity and ecosystem services and working together to strengthen existing partnerships and create new educational opportunities that rely on indigenous and scientific understanding of forest ecosystems.
Rey et al. (2019) discussed that water and the goal of soil bioengineering is to maximize the advantages to both people and the environment via the integrated use of scientific knowledge and practical experience in ecosystem management. Strategies that include living plants as structural components are incorporated. First, they define bioengineering and its historical development while highlighting the challenges inherent in the field's application to action planning, implementation, and evaluation.
Zhang et al. (2021) explained that dry and environmental degradation was made more difficult by a lack of available water in arid environments. Water consumption in the instrument that measures the League of China's Inner Mongolia Autonomous Region was estimated. A system dynamic model incorporating economic, sociological, and ecological elements was built to mimic the water-use structure in this region. Plantation sites must be carefully selected based on their water-balance state, the environmental land size must be adequately controlled, and efficient water-saving measures in industrial and agricultural output must be implemented.
The condition-based monitoring (CBM) method was developed by Yodo et al. (2023) to reduce unscheduled system downtime by tracking a system's current state of health and anticipating when problems might occur. The paper provides a thorough assessment of CBM's value as a strong technique to improve the resilience of energy infrastructure. The results of this poll will also help those who work in the energy sector prepare for and respond to disasters, such as power outages caused by extreme weather.
Evidenced by the recent funding of 19 research projects under the Horizon 2020 program, the work of Kalampokis et al. (2023), in developing a concept-centric analysis framework (C-CAF), sheds light on how the public sector is using and deploying new technologies. To the best of our knowledge, this is the first effort to collect such data from the perspective of research projects; the results may shed light on how these technologies are really used and implemented in pilot programs.
Focusing on the optimization, power quality enhancement, and fault outbreak analyses of hybrid renewable microgrids, Zulu et al. (2023) introduced Artificial Intelligence Optimization Technique Applications in a Hybrid Microgrid (AIOTA-HM). In microgrids, fault outbreaks limit performance and call for adaptive approaches to energy management and efficiency. In recent years, artificial intelligence (AI) has emerged as a crucial factor. Microgrid applications of AI are discussed, along with their strengths, cons, and future directions.
Ecological regeneration will improve air quality, reverse forest clearing and wilderness, slow biodiversity loss, enhance urban ecosystems and probably improve the livelihoods and ties between mankind and nature.
A comprehensive literature review of AI-based environmental decision restoration is conducted, and the results of the most important studies are presented and discussed. The report details how the authors used AI-EDRF to support their study's dynamic natural systems.
ECOLOGICAL RESTORATION OF THE WATER BODY SUSTAINABILITY
Comprehensive environmental assessment framework for the sustainability of the water body.
Comprehensive environmental assessment framework for the sustainability of the water body.
Assessment of new natural ecosystem mainstream water body sustainability situations; ecological environment assessment module; ecology risk assessment module; total basin ecological water application review module have been discussed. An integrated water body environmental assessment framework includes data from dynamic remote sensing tracking, comprehensive databases, and real-time ecological and environmental monitoring data. The measurement will be conducted quantitatively to provide a qualitative approximation of the ecological demand for water, using the framework and the model to conduct a detailed ecological and environmental assessment in a mainstream area. Vegetation water scarcity can be estimated. The findings are transmitted to the water regulatory system to control the environmental drainage system through a mainstream ecological gate utilizing a water regulatory model through the environmental water requirement model. All the module findings can be transferred to the local area network (LAN) discussion center, printed, and available for consultation.
Technical, environmental monitoring framework evaluating the water sources' biodiversity is a robust geo-information system composed of the quality assurance module for the ecological environment, the environmental risk assurance module, and the ecosystem demand module.
Ecological and environmental quality assessment features include ecological, environmental assessment index systems evaluation factor weight assessment determinations to display the ecological quality assessment chart, administrative divisions environmental quality assessments, and polygon ecological assessment.
Recognizing environmental risk factors, the desertification risk assessment model, the sedimentation risk assessment model, and the vegetation stress risk assessment model. A review of risk factors is part of the risk evaluation module. Input data from vegetation water demand knowledge quotas through water supply assessment subsystems and remote sensing complicated land use monitoring; research of global demand in the water of various ecological systems; geographical information system (GIS)-assisted ecological water demands spatial analyses.
Internet of Things (IoT) technologies are a potential option for achieving efficient and effective monitoring. First, this article introduces the idea of using the IoT to keep tabs on the working conditions at a factory. The water quality sensor is a broad name for various sensors that detect residual chlorine, turbidity, suspended particles, conductivity, and dissolved oxygen. Water quality does not refer to a certain day characteristic. It comprises various components to measure water conditions. The operating idea of the water level sensor is that when it is inserted into a given depth in the liquid to be measured, the pressure on the sensor's front surface is transformed into the liquid level height. The water flow sensor comprises a copper body, a water rotor, and a hall-effect sensor. When water runs through the rotor, the rotor rolls its speed changes with different flow rates. And the hall-effect sensor outputs the matching pulse signal. A database is an information for easy access, administration, and updating. Computer databases often hold data records or files aggregating sales transactions, customer data, financials, and product information. There is the interaction between a user and software operating on a web server. The user interface is the web browser, and the web page is downloaded and presented (as shown in Figure 4).
ARTIFICIAL INTELLIGENCE-BASED ENVIRONMENTAL DECISION RESTORATION FRAMEWORK (AI-EDRF)
























































The bth input , if the statistical mean and standard deviation are
and
, respectively. The water body sustainability can be derived and expressed in Equations (11) and (12). The Gaussian Process method has been successfully implemented in many domains, such as image and signal processing and broad speed prediction, to manage challenging prediction tasks because of its obvious benefits, such as satisfactory generalization ability, high robustness, and low computational complexity. Motivated by the promising results of the Gaussian Process technique, this research introduces a new technique for abstracting the mapping between location details and temperature readings.













Equation (17) denotes B for the power supply system using temperature sensor and web interface
in the logarithmic function for controlling water
supply and
the mathematical function for
monitoring water control.
The reaction evaluation depends on the degree of environmental damage that varies according to harmful concentrations and has a non-linear impact. The relationship variables assessed the relationships between hazardous states and the vulnerability of environments. Scientific data screenings, laboratory studies, historical history, direct findings, and literature evaluation findings have been used to forecast the benchmarks to ensure a good relationship between danger and the environment's vulnerability (as shown in Figure 6a and 6b).
An environmental decision restoration analysis was conducted using the AI technique to determine sufficient environmental restoration times in the terrestrial ecosystem. The process is described as follows: the first was to use three aspects, for example, provincial biohazard, municipal-financial eminence, and evaluation by various decision-makers, for the treatment and normalization of the results. The second solution was to have positive values and negative values. Each alternative's likeness to the ideal answer was determined, and distances between each option and the best and worst alternatives were also estimated.
EXPERIMENTAL ANALYSIS OF THE ENVIRONMENTAL DECISION RESTORATION
(a) Water quality and measurement of water level, (b) energy storage with IoT device.
(a) Water quality and measurement of water level, (b) energy storage with IoT device.
Accuracy analysis

Performance ratio

(a) Performance ratio (fog nodes), (b) performance ratio (user systems).
Reaction time analysis


Data deployment analysis

Outcome table for monitoring intervals
Metrics . | FL . | NN. . | EC . | PM . | AI-EDRF . | Discoveries . |
---|---|---|---|---|---|---|
Accuracy (%) | 57.00 | 59.82 | 71.09 | 40.09 | 94.73 | 39.83% High |
Data deployment (%) | 82.58 | 82.73 | 80.91 | 84.09 | 96.14 | 14.11% High |
Metrics . | FL . | NN. . | EC . | PM . | AI-EDRF . | Discoveries . |
---|---|---|---|---|---|---|
Accuracy (%) | 57.00 | 59.82 | 71.09 | 40.09 | 94.73 | 39.83% High |
Data deployment (%) | 82.58 | 82.73 | 80.91 | 84.09 | 96.14 | 14.11% High |
Outcome table for fog nodes
Metrics . | FL . | NN. . | EC . | PM . | AI-EDRF . | Discoveries . |
---|---|---|---|---|---|---|
Accuracy (%) | 62.70 | 65.80 | 78.20 | 44.10 | 93.20 | 32.72% High |
Performance ratio (%) | 66.03 | 54.10 | 65.80 | 78.20 | 96.35 | 31.46% High |
Reaction time (%) | 60.55 | 57.30 | 66.20 | 58.15 | 19.10 | 68.46% Low |
Data deployment (%) | 90.83 | 91.00 | 89.00 | 92.50 | 94.75 | 4.14% High |
Metrics . | FL . | NN. . | EC . | PM . | AI-EDRF . | Discoveries . |
---|---|---|---|---|---|---|
Accuracy (%) | 62.70 | 65.80 | 78.20 | 44.10 | 93.20 | 32.72% High |
Performance ratio (%) | 66.03 | 54.10 | 65.80 | 78.20 | 96.35 | 31.46% High |
Reaction time (%) | 60.55 | 57.30 | 66.20 | 58.15 | 19.10 | 68.46% Low |
Data deployment (%) | 90.83 | 91.00 | 89.00 | 92.50 | 94.75 | 4.14% High |
Outcome table for users systems
Metrics . | FL . | NN. . | EC . | PM . | AI-EDRF . | Discoveries . |
---|---|---|---|---|---|---|
Performance ratio (%) | 60.03 | 49.18 | 59.82 | 71.09 | 97.59 | 38.48% High |
Reaction time (%) | 55.05 | 52.09 | 60.18 | 52.86 | 17.36 | 68.46% Low |
Metrics . | FL . | NN. . | EC . | PM . | AI-EDRF . | Discoveries . |
---|---|---|---|---|---|---|
Performance ratio (%) | 60.03 | 49.18 | 59.82 | 71.09 | 97.59 | 38.48% High |
Reaction time (%) | 55.05 | 52.09 | 60.18 | 52.86 | 17.36 | 68.46% Low |
(a) Data deployment (monitoring intervals), (b) data deployment (fog nodes).
Table 1 shows that the developed framework (AI-EDRF) for monitoring intervals produces much better accuracy and data deployment performance. It explores better accuracy and data deployment results by 94.73 and 96.14%, respectively. Simultaneously, the traditional (FL) monitoring intervals show poor accuracy and data deployment results. The comparative study results between the developed framework (AI-EDRF) and traditional methods explore improving accuracy and data deployment by 39.83 and 14.11%, respectively.
Table 2 shows that the developed framework (AI-EDRF) for fog nodes explores better performance in all outcome metrics than traditional approaches like FL, NN, EC, and PM. The comparative analysis between developed and traditional approaches explores improving accuracy, performance ratio, and data deployment by 32.72, 31.46, and 4.14%, respectively, and decreases the reaction time by 68.46%.
Table 3 shows that the developed framework (AI-EDRF) for user systems produces a better performance ratio and reaction time. It explores better performance ratio and reaction time results by 97.59 and 17.36%, respectively. Simultaneously, the traditional method (FL) for user systems shows poor performance and reaction time results. The comparative study results between the developed framework (AI-EDRF) and traditional methods explore the betterment in performance ratio by 38.48% and a decrease in reaction time by 68.46%.
When electricity is created, transferred, and dispersed, it is used. Electricity from transmission and distribution networks is turned into useable work, light, heat, or a mix at utilization points. Energy is required for many everyday activities, including keeping our houses warm and cool, lighting our workplaces, moving our personal belongings, and creating things we utilize.
Energy waste is wasted power, water, and gas usage that does not accomplish any beneficial purpose. One of the most common causes of energy waste is inadequate management and maintenance of energy equipment. The three main categories of energy waste are permanent difficulties from long-term energy squandering.
Customers can already stay updated on the energy they use due to the proliferation of IoT devices, smart meters, and other intelligent electronic devices. Sensor data may be collected and analyzed in real-time thanks to the IoT infrastructure. Designers may expect a major transformation they interact with the physical environment due to IoT devices. An IoT system may track and manage any connected object, machine, or even live organism using sensors and actuators linked via networks to a central database (as shown in Figures 11, 12 and 13).
This research analyzes the efficacy of an AI-EDRF for restoring aquatic ecosystems. For the purpose to gather information from sensors and issue alerts in the event of abnormal operations, the framework employs a network of fog nodes and user systems. The effectiveness of the framework is measured in terms of quality criteria including accuracy, performance ratio, and response time. As the number of monitors and fog nodes in the system grows, so does its precision. As the number of fog nodes and users in the system grows, so does the performance ratio. The response time is decreased because of the use of an AI system to categorize sensor data as either normal or abnormal. Conventional approaches such as FL, NN, EC, and PM are outperformed by the suggested AI-EDRF. The research shows that with precise and timely warnings, the AI-EDRF can help the ecological restoration process.
CONCLUSION AND FUTURE ANALYSIS
Human interference in developed countries strains biological processes, primarily due to rapid urbanization. The paper uses the combination of source and argument of environmental decision restoration to identify the three-dimensional significance of environmental sources, environmental paths, and strategic points as machine learning, minimum cumulative resistant models, or the circuit's principle. In this paper, AI-EDRF has been proposed to strengthen the continuously evolving natural systems; people are deficient in natural systems and the insufficient knowledge available about past achievements and failures. The biological terrestrial collective analysis introduced to improve natural systems is rapidly evolving, and people are inadequately aware of natural systems. Stochastic water quality analysis is integrated with AI-EDRF to boost past achievements and failures. The computational analysis is executed based on accuracy, performance ratio, reaction time, and data deployment, verifying the developed framework's reliability. As a result, energy efficiency will gain greatly from the IoT. For utilities, smart grid sensors may help them operate more efficiently by keeping tabs on their power network. It will give them improved distribution resources to meet actual demand, cutting down on energy wastage. As a result of advances in AI design, the accuracy, performance ratio, and workload of consumption of environmental planning and urban management have been vastly improved. Compared to prior approaches, experimental testing shows the new method to be more efficient and accurate. Future analysis will be conducted in the field of geographical visualization, incorporated with other machine learning techniques and the potential for a large-scale model of multiple variables under the provincial risk management framework. Better statistical methods must be developed to analyze responses with personalized criteria and risk assessments. The outcomes resulted in an accuracy ratio of 94.73%, data deployment of 96.14%, a performance ratio and reaction time of 97.59%, a workload of consumption in IoT ratio of 92.5%, and an energy wastage ratio of 85.7%, an average power utilization ratio 97.3%.
ACKNOWLEDGEMENTS
This work was supported by Key scientific research projects of Hunan Provincial Department of Education(17A117).
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.
CONFLICT OF INTEREST
The authors declare there is no conflict.