Water quality system informatics (WQSI) is an emerging field that employs cybernetics to collect and digitize data associated with water quality. It involves monitoring the physical, chemical, and biological processes that affect water quality and the ecological impacts and interconnections within water quality systems. WQSI integrates theories and methods from water quality engineering, information engineering, and system control theory, enabling the intelligent management and control of water quality. This integration revolutionizes the understanding and management of water quality systems with greater precision and higher resolution. WQSI is a new stage of development in environmental engineering that is driven by the digital age. This work explores the fundamental concepts, research topics, and methods of WQSI and its features and potential to promote disciplinary development. The innovation and development of WQSI are crucial for driving the digital and intelligent transformation of national industry patterns in China, positioning China at the forefront of environmental engineering and ecological environment research on a global scale.
Hong Liu, Zhaoming Chen, Zhiwei Wang, Ming Xu, Yutao Wang, Jinju Geng, Fengjun Yin.
Water Quality System Informatics: An Emerging Inter-Discipline of Environmental Engineering.
Engineering, 2024, 43(12): 121-130 DOI:10.1016/j.eng.2024.03.018
Environmental engineering is a remarkable field that brings together various disciplines, including biology, chemistry, material science, civics, and hydraulics. In China, it has become an essential arena that plays a crucial role in promoting the balanced development of society, economy, and ecology [1]. The integration of artificial intelligence (AI) and big data disciplines has further enhanced the interdisciplinary collaboration within environmental engineering, providing new resources and momentum for its continuous growth and development.
In China’s modernization efforts to establish harmonious coexistence between humans and nature, systematic thinking has become a pivotal approach guiding the progress of environmental engineering. For example, Moser [2] introduced systems thinking in systems engineering practice, Rodenas and Jackson [3] applied systems thinking to water environment management, Nguyen and Bosch [4] used systems thinking to study the sustainable development of biosphere reserves, and Monat and Gannon [5] used systems thinking to solve complex engineering problems. However, research in environmental engineering often lacks a comprehensive perspective, as existing paradigms are mainly focused on describing and regulating individual components within a system. Consequently, dynamics and interconnections between individual components are often overlooked, hindering the thorough exploration and utilization of abundant system information and creating a gap between academic achievements and practical engineering. Adopting a systematic thinking approach and integrating the latest data science and AI techniques are crucial for reforming the existing research paradigms, pathways, and methods in environmental engineering. This reform will pave the way for the development of fundamental theories, technical methods, and innovative solutions, ushering in a new stage of progress in this field.
In the traditional research and application of water quality informatization, although sensor technology and remote transmission technology (e.g., network communication) can be adopted to realize the remote collection and transmission of water quality data (e.g., water temperature, pH, dissolved oxygen, and various pollutant concentrations [6], [7], [8]), it is difficult to determine the overall operational characteristics and laws of a water quality system by studying these individual water quality parameters or single indicators. With the rapid development of modern information technology, people have begun to explore the coupling relationship between multiple variables and parameters within water quality systems and to study the impact of changes in multiple parameters on the overall operational efficiency of water quality systems. For example, big data analysis techniques and AI algorithms [9], [10], [11] are used to mine the relationships between multi-source data, thereby revealing potential patterns in water quality data and predicting future water quality trends. By analyzing the interaction relationships between multi-process parameters within a water quality system, intelligent control equipment can be applied to realize the comprehensive regulation of water quality indicators. This makes it possible to improve the overall operation effect of water quality management [12], [13], provide a scientific basis for environmental engineering decision-making, and realize the intelligent and sustainable development of water quality management.
As water pollution control is a crucial aspect of environmental engineering, the focus of research in this area is on studying the rules of transformation and methods of controlling water quality. However, the current research paradigm lacks systematic thinking, failing to fully utilize system information. Therefore, initiating a transformation in the research paradigm within environmental engineering—beginning with the science of water pollution control—is an effective approach to tackle this issue. This initiative has led to the emergence of a new interdisciplinary field, termed water quality system informatics (WQSI).
2. Basic concepts in water quality system informatics
2.1. Water quality
Water quality refers to the quality of water that is necessary to satisfy a certain function; it is described by the parameter set Q, which characterizes the physical, chemical, biological, and ecological properties of water and their temporal variation trend. , where represents the ith time-varying water quality parameter. Q may also include spatial variables. In general, within an artificially defined physical space, various impurities are present for a certain number of water molecules, and the type and quantity of these impurities determine the water quality. Here, the introduction of the time dimension into the concept of water quality develops measures of water quality from the traditional isolated data into continuous or discrete functions of time, which not only endows the concept of water quality with systematic and dynamic implications but also makes it possible to mathematically describe these implications.
2.2. Water quality system
A water quality system is an interdependent entity composed of various parts and links in the existence or change process of water quality. It is represented as , where represents the ith part or link of the system, , and f1(∙) represents the system action function, describing the interdependence between these parts and links. A water quality system includes not only various experimental systems at small or micro scales, but also larger-scale engineering systems, such as municipal water supply networks and wastewater treatment systems, as well as urban and rural water quality systems such as artificial or natural rivers, ditches, ponds, reservoirs, and so forth. Several smaller simple water quality systems can be interconnected into larger and more complex water quality systems, or a large and complex system can be decomposed into several relatively small and simple subsystems.
2.3. Water quality system state and state variables
The water quality system state refers to the motion state of a water quality system described by a set of minimum number of time-domain variables that cannot be linearly represented between each other. It is influenced by both external input factors and internal process factors and can be expressed as . Here, represents the motion state of the water quality system at moment t. denotes external input factors, +, such as pollutant inputs, process operation parameters, and so forth. stands for internal process factors, +, such as the characteristic parameters of biodegradation, chemical reactions, decarbonization processes, nitrogen conversion processes, and so forth. α and β are the natural number. These variables that describe the motion state of a water quality system are called state variables.
2.4. Water quality information
The term water quality information refers to all the information that helps people to increase the certainty of understanding a water quality system and to ensure precise control of the dynamic status of the water quality system; it includes data, text, images, audio, and video. This information includes quantitative indicators of water quality; the physical, chemical, and biochemical reaction processes, mechanisms, and effects of water quality changes; and dynamic relationships that can describe interactions between different parts and links within a water quality system. It can be categorized into state information and process information, internal information and external information, and so forth.
2.5. Water quality signal
A water quality signal refers to a physical signal carrying water quality information, such as electricity, light, sound, and heat, as a function of the time variable, . If spatial variables are included, it is x, y, and z represent a point in spatial coordinate. Non-electrical water quality information must be converted into electrical signals for analysis and utilization. The water quality signal is the physical carrier of water quality information; it is the prerequisite and mathematical means to understand and control a water quality system. Water quality information is the specific content carried by water quality signals, reflecting the state of the water quality system. Water quality information and water quality signals occupy a fundamental core position in the water quality system.
Water quality signals usually include analog water quality signals and digital water quality signals. Analog water quality signals are time-continuous signals that can be recognized and acquired by various sensors and testing instruments; they include dissolved oxygen, pH, and turbidity, and change continuously with time. When water quality signals have values only at certain time-discrete points and are undefined at other moments, they are called discrete signals. Analog water quality signals are converted through sampling, holding, quantization, and encoding into a specific type of discrete signal, called digital water quality signals, which are usually represented as binary numbers to facilitate storage, transmission, and processing within computer systems.
2.6. Water quality system informatics
Guided by system cybernetics, the field of water quality system informatics (WQSI) involves the signaling and digitization of information related to the dynamic interdependencies among various components of the system, thereby realizing the intelligent management and control of a water quality system. WQSI is based on studying the survival status of water quality; the physical, chemical, and biological processes of water quality changes; and the ecological effects of water quality. Core components of WQSI include water quality information mining, water quality signal utilization, water quality system properties and models, and intelligent control of the water quality system status.
3. Regulating water quality systems
3.1. Input-output mapping
The water quality system is one of the fundamental concepts of WQSI, embodying a systematic approach to understanding and managing water quality. It serves as the mechanism for transmitting and transforming signals related to water quality. We can initiate the grasping and regulation of a water quality system by establishing a connection between the inputs (signals) and outputs (signals), as shown in Fig. 1.
Input signals originate from external factors such as materials, energy, or physical disturbances affecting the water quality system, whereas output signals represent the system’s response to these external factors, which may also encompass random disturbances unrelated to the initial inputs. Only signals that notably impact the water quality system, such as those influencing the thermodynamic state or the rate-controlling steps of chemical or biochemical reactions, are considered input signals. This includes temporal variations in process parameters such as the material dosage, influent flow rate, aeration rate, and reflux ratio [14]. The state parameters of the water quality system or other predetermined physical quantities triggered by the input signal can serve as output signals, encompassing various water quality parameters and physical quantities, such as the time-domain function of the redox potential [15].
Both the input and output signals are typically described using time-domain functions, which can be transformed into frequency-domain functions through time-frequency conversion for frequency-domain analysis. The relationship between the input and output of a water quality system—determined through methods such as functional relationships, artificially defined rules, or models—constitutes the input-output mapping of the water quality system. Direct input-output mapping can be established to provide an external description of the water quality system, while indirect mapping can also be established through the motion state of the water quality system to achieve an internal description. For any water quality system, the input-output mapping comprises both linear and nonlinear components.
3.2. Linear water quality systems
All water quality systems are fundamentally nonlinear; however, for the sake of simplicity, they are often treated as linear systems while still maintaining their primary physical significance. If a wastewater treatment reactor follows first-order reaction kinetics, it is considered a linear water quality system. In mathematical terms, a linear water quality system is defined as a system in which the physical, chemical, and biological processes reach a steady state, and the equations or models describing the system are linear equations that satisfy homogeneity and superposition simultaneously [16]. This includes both single-input single-output (SISO) linear water quality systems and multi-input multi-output (MIMO) linear water quality systems.
3.2.1. SISO linear water quality systems
When a water quality system contains only one input and one output and satisfies the linear property, it is called a SISO linear water quality system . Here, are all time–domain functions, as illustrated in Fig. 2. If the input and output water quality signals also satisfy time–invariance (i.e., the output does not change with the starting point of the input time), the system can be further simplified into a linear time–invariant water quality system.
For a time–continuous linear water quality system, the dynamic relationship between input and output is typically described by constant coefficient linear differential equations, as follows:where and are real coefficients. Here, denote the order of derivative of the function. The left side of Eq. (1) reflects the inherent properties of the water quality system as determined by physical, chemical, or biological processes and various physical parameters, while the right side of Eq. (1) represents the interaction between the input signals and the water quality system. Differential equations are vital for analyzing a linear water quality system and, in the case of discrete signals, the system is described by difference equations.
Eq. (1) indicates that, although there are different types of water quality systems, their basic laws are universal; that is, they can all be described by Eq. (1). For different system orders and different input excitation x(t), the system output responses will be different, thus distinguishing the type of water quality system and its internal structural properties. Therefore, the comprehensive representation of the physical, chemical, and biological mechanisms within a water quality system is reflected in this relationship between the input-output of the system and can be described by a unified mathematical equation such as Eq. (1). In the case of a linear water quality system, the relationship between the system’s process mechanisms and the external signals is the micro and macro or cause and result relationship. This relationship is the theoretical foundation for environmental engineering science evolving from technical science to engineering science, and further to system science. It is also the main feature that distinguishes the research paradigm of water quality informatics from conventional technical science.
As shown in Fig. 2, if the input signal is the unit impulse signal , then the ratio of and after a Laplace transformation is called the transfer function of the water quality system. Here =/, X(s) is the Laplace transform of , Y(s) is the Laplace transform of . This is a basic mathematical tool to describe the structure of a linear water quality system, and it is the main method to study a linear water quality system. Studying a linear water quality system via the method of water quality informatics may bring new discoveries. In Eq. (1), when , the system can be called a first-order water quality system; when , it is a second-order water quality system. Common water quality systems—especially experimental systems—can be simplified into first-order or second-order water quality systems. For a first-order water quality system, the equation can be rearranged as follows:where are constants, and the homogeneous solution is:
Eq. (3) is exactly the first-order reaction kinetic equation for a chemical reaction, where is the initial value of the reactants. By further considering the different forms of input signals on the right-hand side of this equation (e.g., sine signals, step signals, pulse signals), it is possible to simulate the output responses and optimize the input functions to obtain the optimal control mode of dosing, aeration, and so on. This method is beyond the reach of traditional environmental engineering research methods. The method can also be applied to a second-order water quality system, although the mathematical solution process is more complex. In addition to this time-domain method, a frequency-domain method can be used to obtain the pole-zero configuration of the water quality system and realize precise control. For a third-order and above higher-order water quality system, the state equation approach is utilized.
3.2.2. MIMO linear water quality systems
For higher-order water quality systems, state equations are used to describe the dynamic relationship between the state variables and the inputs of the system, while output equations are adopted to describe the output relationship between the state variables and inputs. Compared with the differential equations of a SISO water quality system, which only considers the external characteristics of the system, the combination of state equations and output equations fully considers the intrinsic properties of the water quality system. This is known as the state-space expression, state-space description, or state equations of the water quality system [17]. Together with the transfer function matrix [18], it constitutes an important tool for studying MIMO linear water quality systems. A state-space expression can be applied to analyze higher-order linear water quality systems as well as nonlinear and time-varying water quality systems.
The state–space expression of a MIMO linear water quality system can be expressed in a vector matrix form [17]:where are column vectors, is the state vector which is n dimensional, represents the first derivative of , is an system matrix (representing the coupling between state variables within the system), is an input matrix (indicating the effect of various inputs on state variables), is an output matrix (representing the transformation of state variables to outputs), and is an direct matrix (indicating the inputs’ direct action on the outputs, often a zero matrix). For time-varying water quality systems, some elements of are functions of time. Compared with SISO linear systems, the structure of MIMO linear water quality systems is more complex, requiring special attention to the study of their controllability and observability. Here, controllability refers to the possibility of controlling the state of the controlled system with input actions, and the controllability of the output is more meaningful. Observability refers to whether or not the measured value of the system output can determine the system state.
In a MIMO linear water quality system, when it is necessary to measure the effect of an input on the state variables of a system, the following methods can usually be used: ① sensitivity analysis, which is achieved by calculating the partial derivatives (i.e., sensitivity) of the state variables on each input; ② control volume analysis, which divides the system into different control volumes and then analyzes the effect of each input variable on the state variables of the system in each control volume—something that can be achieved through mathematical modeling and simulation to quantitatively evaluate the effect of each input on the state variables; and ③ the system identification approach, where the state space model of the system is estimated by analyzing the system input and output data to assess the effect of different inputs on the state variables of the system. Through these methods, the effects of a single input on the system state variables can be quantified and analyzed to help in understanding the system’s behavior and to carry out system control and optimization.
Although linear water quality system models have limitations when used for actual nonlinear water quality systems, they can still provide useful approximations and inspiration in certain situations. For example, under stable environmental conditions, a linear model may be able to better characterize changes in water quality parameters. Thus, in such cases, a linear model can be used to predict trends in water quality parameters. Another example is that, although a water quality system may be nonlinear, the effects of certain pollutants may exhibit linear relationships to some extent. In this case, linear models can be used to identify the main pollution sources and provide guidance for further measures. Aside from practical applications, when limited by data availability and computational resources, linear models may be used for parameter estimation or simplified models of water quality systems while modeling and analyzing water quality systems for rapid analysis and decision-making. Therefore, although linear models may not fully describe nonlinear system behavior, they are still feasible and useful in some cases.
3.3. Nonlinear water quality systems
The term nonlinear water quality system refers to a water quality system that has not yet reached a stable state in a specific aspect of its physical, chemical, or biological processes and cannot be accurately described using linear equations. In these systems, the output response of a wastewater treatment reactor is not simply a linear function of the input stimulus but may involve various nonlinear elements, such as power functions, trigonometric functions, exponential functions, and so on. Nonlinear water quality systems do not adhere to homogeneity and superposition, meaning that the overall effect is not equal to the sum of individual effects, and they cannot be described using transfer functions. Furthermore, the stability of a nonlinear water quality system is not only determined by its structural properties but also depends on the initial conditions. Within a nonlinear water quality system, there are often locally stable and unstable motions, both of which can be analyzed separately to understand local stability.
The current understanding of nonlinear water quality systems is quite limited. In engineering applications, it is challenging to accurately predict the output processes of these systems. The primary concerns are whether the system is stable and whether it produces self-excited oscillations. Analyzing nonlinear water quality systems is also much more complex, and traditional mathematical tools such as Laplace transforms and Fourier transforms, which are used for linear systems, are not suitable. The mathematical equations for these systems often lack analytical solutions, so they primarily rely on phase-plane analysis, descriptive function methods, and the Lyapunov method [19], [20].
3.4. Management and control of water quality systems
For a linear water quality system, using multiple sources of information, we can create equations that reflect the relationships between variables. These equations involve mass conservation, charge conservation, and energy conservation, as well as relevant boundary and initial condition constraints. However, when dealing with a nonlinear water quality system, it is often challenging to establish a comprehensive mathematical model, making control much more difficult.
The rapid advancement of big data science has provided the necessary data to build data-driven models for managing nonlinear water quality systems. By understanding the linear laws governing water quality systems, techniques such as machine/deep learning can be used to uncover nonlinear patterns and characteristics within extensive datasets, leading to the creation of relevant data-driven models. However, these models sometimes lack interpretability, and achieving a global optimum can be challenging. Therefore, combining a mechanism control model with a data-driven model to form a composite control model (Fig. 3) represents an important direction for the advancement of water quality system control technology.
Composite control models are usually constructed to combine the advantages of mechanistic models and data-driven models in order to better reveal the nonlinear characteristics of water quality systems and improve the predictability and adaptability of models. This combination can be constructed using different methods:
(1) Hybrid modeling: In this method, a mechanistic model is used to provide a deep understanding of the physical processes of the system, while a data-driven model is used to capture the complex nonlinear relationships in the system. Through a reasonable combination of the two, a more comprehensive and accurate control model can be obtained.
(2) Nested model: In this kind of composite model, a mechanism model is used as the main body to describe the overall behavior of the system, while a data-driven model is nested within it to capture the local changes or nonlinear relationships in the system.
(3) Parameter correction: In this model, a data-driven model is used to learn the real-time data to adjust and optimize the parameters of the mechanistic model, thereby improving the adaptability and accuracy of the model.
3.5. Information acquisition and simulation in water quality systems
Acquiring information about water quality is crucial for understanding and managing a water quality system. This involves studying the internal laws and mechanisms of the system from various perspectives, such as the sensing layer, data layer, and feature layer. It also involves representing, refining, converting, fusing, and utilizing this information [21]. Establishing a signal connection between sensing methods and devices is essential for functional signal utilization (Fig. 4). To address the integration challenges among the mechanism, signal, and equipment, interdisciplinary knowledge must be deeply integrated and organically fused. Water quality information consists of diverse data from multiple sources, which are overlapping, multidimensional, heterogeneous, and dynamic. Consequently, operations such as cleaning, fusion, and transmission are necessary to accurately assess water quality properties.
Traditional water quality information acquisition is limited to measuring the current water quality process and related parameters; thus, it lacks a comprehensive depiction of future changes in the water quality system. Digital twin technology, however, can simulate the evolution of water quality and predict the performance of a water quality system at a relatively small cost. It can also display changes in various state parameters through graphics and animations. This detailed and realistic information output makes digital twin technology an important tool for process design, operation management, comprehensive evaluation, and even the development of new technologies for water quality systems [22].
In recent years, the rapid growth of information fusion technologies such as network technology, cloud computing, and virtual simulation has set the stage for the application of digital twin technology in water quality systems. Through digital twin technology, it becomes possible to achieve synchronized operation, virtual-real interaction, and iterative optimization of water quality systems in a virtual space [23]. This allows for dynamic images to reflect the operational status and changing trends of water quality systems, enabling intuitive fine process control including status perception, analysis, decision-making, and execution (Fig. 5). In essence, digital twin technology for water quality is primarily centered on the comprehensive digitization and real-time visualization of water quality systems, thereby enabling intelligent control of the entire process within the water quality system.
However, because of the complex nature of water quality systems, the current simulation methods and professional software often oversimplify the connections between different parts of the system, leading to a need for improved accuracy in the results. The ongoing advancement of digital twin technology for water quality systems, along with the continuous enhancement of simulation precision and prediction efficiency, is crucial in order to meet the practical demands of both spatiotemporal and dynamic regulation simulations of water quality systems. This is a vital area that WQSI urgently needs to focus on in the future.
3.6. Standardization of water quality systems
When a unified technical standard system is implemented for the various aspects of water quality systems, such as information collection, transmission, storage, processing, and system control, it will enhance the sharing of information resources and prevent redundant development and construction. This will ultimately lead to a significant improvement in economic efficiency. Standardization is crucial for the engineering application and advancement of water quality systems.
Water quality system standards encompass not only information technology standards but also relevant standards in areas such as applied information technology and intelligent control. Currently, there is a lack of unified standards for multi-source information in water quality systems. It is important to integrate the existing information indicator system for water quality in order to establish common standards and key information technology standards for information storage, transmission, sharing, and software development. Strengthening the standardization of water quality systems and establishing an effective, advanced, and applicable standard system are also crucial aspects for the future development of WQSI.
4. Research topics within and methods of WQSI
The research topics within WQSI include five aspects, with a focus on how to obtain multi-source heterogeneous water quality information and accordingly construct an informatics expression and control model for a water quality system (Fig. 6).
4.1. Informatics expression
The area of informatics expression encompasses differential equations, difference equations, transfer functions (frequency response functions), pole-zero plots, and signal flow diagrams, all of which describe a SISO system. It also encompasses the state-space expression for describing MIMO water quality systems and nonlinear water quality systems. Furthermore, it includes analyses of the controllability and observability of linear water quality systems, as well as the proposal of optimal control methods.
4.2. Information acquisition
Information acquisition involves the utilization of various measurement technologies such as sensors and remote sensing to gather water quality information data, along with the use of modern analytical instruments to obtain the concentration data of various pollutants in the water quality system. Additionally, it encompasses simulation data derived from the simulation process of a water quality system. Moreover, it involves research on data transmission protocols, data cloud storage, and database management methods related to the transmission of the acquired massive water quality information data from the collection point to the analysis platform.
4.3. Information analysis and processing
Information analysis and processing encompass various data preprocessing techniques, such as data noise removal and data outlier detection, to ensure the accuracy and availability of the collected massive data. It also includes real-time data analysis and intelligent data processing methods such as data mining and machine learning to extract useful information from large amounts of data and deeply analyze the dynamic properties of the water quality system. In addition, it involves the visualization of water quality data to aid users in understanding water quality data more easily, thereby assisting decision-makers and researchers in quickly identifying water quality evolution trends.
4.4. Modeling and simulation
The area of modeling and simulation encompasses more than just the physical modeling of water quality systems. It primarily uses mathematical equations to describe processes such as water temperature distribution and water mixing. Moreover, it involves modeling chemical and biological reactions within water quality systems, taking into account factors such as chemical reaction kinetics and ecological models to describe how pollutants’ concentrations change and how microorganisms interact within the water system. Furthermore, data-driven control models can be developed using the extensive water quality data collected, enabling the prediction of future water quality changes through machine learning and time series analysis. Simulation techniques such as computational fluid dynamics, finite element analysis, and other numerical methods are also employed to simulate how water quality changes under different scenarios.
4.5. Standardization construction
The aspect of standardization construction involves establishing a standard system for describing signals within water quality systems, as well as standardized methods for collecting and storing water quality data. It also encompasses developing technical standards for intelligent control indicators and evaluating water quality systems. Furthermore, it includes creating standardized procedures for operating water quality systems and standards for sharing water quality data to ensure consistency in data collection and analysis processes, as well as accessibility to shared water quality data.
Traditional research methods in water quality science typically rely on principles such as Newton’s laws of motion, mass conservation, and charge conservation laws in chemical reactions, as well as biochemical reaction principles [24], which fall within the traditional framework of material science research. In contrast, the distinctiveness of the WQSI research methodology lies in its foundation on system theory, cybernetics, and information theory, among others. It incorporates research methods from data science, operations research, informatics, control science, and management science, along with relevant technologies in online sensing and equipment development. Through this approach, it continuously develops its own foundational theories, research methods, and technological framework, representing a significant breakthrough in the cognitive style, research paradigm, and control mode of water quality systems.
The research components and methods are interconnected and mutually influence each other to create a comprehensive research framework. Acquiring high-quality water information serves as the foundation for information analysis and modeling. Standardizing the expression of water quality information supports mathematical operations and information processing, contributing to data consistency in information acquisition and analysis. In turn, this enhances the accuracy of water quality system modeling and simulation. Analytical methods for water quality information can be utilized to calibrate models and improve simulation accuracy, ensuring data quality assurance for standardized construction. Standardization guarantees the comparability and consistency of water quality data, reducing the cost-effectiveness of information acquisition, analysis, and application.
5. Discipline characteristics and significance of WQSI
WQSI belongs to the discipline of environmental engineering. It not only consolidates existing theories and technical methods of water pollution control science but also integrates theories and technical methods from other disciplines such as system theory, cybernetics, and information theory. Its interdisciplinary nature is evident, as depicted in Fig. 7.
WQSI emphasizes a systematic and global understanding of water quality, taking into account the interaction and processes among various components within the system. This approach aids in understanding the dynamic behavior and interaction relationships of a water quality system, highlighting its significant systematic features. In addition, WQSI applies information theory technology to quantify and manage water quality data, including data collection, transmission, and processing, in order to enhance the efficiency and reliability of data utilization, showcasing its prominent informatization features. Furthermore, WQSI relies on a substantial amount of data, including monitoring data, experimental data, and simulation data. It leverages machine learning, data mining, and other data science techniques to analyze this data, demonstrating notable data-driven features. Finally, WQSI utilizes intelligent control methods such as feedback control, optimal control, and adaptive control to achieve real-time regulation and optimization of the water quality system and ensure that the water body meets specific quality standards. Therefore, it possesses prominent intelligent features.
The process of cross-integration between WQSI and other disciplines is shown in Fig. 7. Firstly, the data used in WQSI is obtained through sensing, testing, and other technologies; thus, WQSI is closely tied to testing and measuring, instrumentation, and other disciplines. Secondly, the transmission and processing process of water quality information is inseparable from signal analysis, information communication, and other technologies; thus, WQSI is closely related to the disciplines of information engineering and communication engineering. Again, through the analysis of water quality information, the regulation of process parameters within a water quality system can be carried out, integrating with control theory and automation disciplines. Finally, the results of water quality information processing are helpful for enhancing water quality monitoring and improving the management of water resources and the aquatic environment; thus, WQSI intersects with environmental engineering, ecology, and similar fields. The development of WQSI cannot be separated from the development and technical support of other related disciplines, and such development will also promote the development of other disciplines and lead to the advancement of related technologies. This is the significance of cross-integration development between different disciplines.
Currently, research on WQSI is still in its early stages. The concept of water quality informatics is mentioned in Ref. [25], but that study only expands on the calculation and evaluation techniques of water quality indicators without addressing the specific aspects outlined in this paper. The study also neglects to discuss the systematic and dynamic nature of water quality, as well as key issues such as mathematical description and system control. Therefore, a comprehensive study of WQSI can not only facilitate effective interdisciplinary collaboration but also guide the further development of the environmental engineering discipline.
After years of exploration and accumulation, the internal logic of the discipline of environmental engineering has entered a new stage of development, characterized by new cognitive perspectives, thinking methods, and technical tools. The proposal and establishment of WQSI embody the systematic, holistic, green, intelligent, and equipment-oriented values of environmental engineering. The increasing clarity of WQSI’s “informationization” characteristic reflects this evolution [1]. Consequently, the construction and development of WQSI in China are essential for driving the digital and intelligent transformation of national industrial forms and for seizing a leading position in the international environmental engineering discipline.
6. Prospects of WQSI
WQSI is still in its early stages, but it has great potential for development. At the theoretical level, concepts such as water quality information theory, water quality system cybernetics, and water quality system management science will form the foundation of WQSI within the logical framework of data, information, and knowledge. On the technical side, water quality information perception and equipment, water quality simulation techniques, and intelligent control technology for complex water quality systems will support the development of WQSI. In terms of engineering applications, WQSI will focus on intelligent decision-making and intelligent control for the optimized operation of complex water quality systems.
Furthermore, WQSI has distinct boundaries, rich content, and cross-disciplinary characteristics, making it a new growth point for environmental engineering under the influence of data and intelligence-driven research. China’s “dual carbon” goals present challenges to environmental engineering, and the purpose of WQSI is to deepen our understanding of carbon sources and sinks in the river and small-scale water systems. WQSI will also develop theoretical knowledge and technical methods for complex water quality systems—particularly in large-scale pollution and carbon reduction systems—and play a key role in their economic and reliable operation. For example, the use of high-resolution monitoring data provided by WQSI can help governments and enterprises to more accurately assess carbon emissions and monitor their changing trends. Emission sources can also be finely managed, and the treatment process can be optimized to reduce energy consumption and carbon emissions. WQSI can also help establish a regulatory and trading platform for carbon emission data through the informatics method in order to promote the development of the carbon trading market and provide an economic incentive mechanism for carbon emissions reduction. It is expected that WQSI will advance basic research in various subfields of environmental engineering and serve as a theoretical foundation for the development of other disciplines such as the management of water resources and aquatic environments.
7. Conclusions
WQSI is a new interdisciplinary field that has emerged from the deep integration and penetration of water quality engineering, information engineering, and control engineering, guided by systems cybernetics. It provides the scientific groundwork for creating new methods and technologies to intelligently control water quality systems. The development of WQSI will assist in the extraction and use of information from water quality systems and will address challenges such as nonlinearity, time-varying behavior, and uncertainty in real water quality systems. Thus, it will enable the high-level intelligent control of complex water quality systems.
To promote the development of WQSI, two key aspects need to be considered. Firstly, it is crucial to strengthen foundational theoretical research in this field, enhance the integration between relevant disciplines, and refine the theoretical and methodological system. Secondly, it is essential to deeply integrate the latest achievements in information technology with WQSI in order to meet the requirements for intelligent control in water quality systems. This involves making technological innovations, establishing standards, and developing applications to continually enhance the technical capabilities for intelligent control.
The establishment and growth of this new interdisciplinary field in China are essential for driving the digital and intelligent transformation of the national industrial sector. They will also help China to gain a leading international position in environmental engineering and in the broader field of ecological and environmental sciences, which holds significant strategic importance.
Acknowledgments
We would like to express our gratitude for the support provided by the National Natural Science Foundation of China (52327813, 52131003, and 52370058); the Outstanding Scientist of Chongqing Talent Plan (CQYC20210101288 and CQYC202101006). Special thanks to Qiuan Huang from Shanghai University and Chuan Wang from Guangzhou University for their constructive insights. Some of the contents have been presented at the 2023 Ecological and Environmental Systems Engineering and Risk Control Forum—Discipline Strategic Development Seminar (Nanjing), the 2023 5th China Urban Water Environment and Water Ecology Development Conference (Wuhan), and the 2023 National and International Water Summits for Key Green Water Tech Research and Development (Hong Kong).
Compliance with ethics guidelines
Hong Liu, Zhaoming Chen, Zhiwei Wang, Ming Xu, Yutao Wang, Jinju Geng, and Fengjun Yin declare that they have no conflict of interest or financial conflicts to disclose.
J.Yang. The development of environmental engineering disciplinary in context of National Natural Science Foundation of China. Chin J Environ Eng, 16 (4) (2022), pp. 1055-1062.
[2]
H.Moser. Systems engineering, systems thinking, and learning: a case study in space industry. Springer, Berlin (2013).
[3]
M.A.Rodenas, M.C.Jackson. Lessons for systems engineering from the Segura River reclamation project: a critical systems thinking analysis. Syst Res Behav Sci, 38 (3) (2021), pp. 368-376.
[4]
N.C.Nguyen, O.J.H.Bosch. A systems thinking approach to identify leverage points for sustainability: a case study in the Cat Ba Biosphere Reserve, Vietnam. Syst Res Behav Sci, 30 (2) (2013), pp. 104-115.
[5]
J.P.Monat, T.F.Gannon. Applying systems thinking to engineering and design. Systems, 6 (3) (2018), p. 34.
[6]
U.Ewuzie, N.O.Aku, S.U.Nwankpa. An appraisal of data collection, analysis, and reporting adopted for water quality assessment: a case of Nigeria water quality research. Heliyon, 7 (9) (2021), p. e07950.
[7]
H.M. CamposRibeiro, A.D.C.Almeida, B.R.P.D.Rocha, A.V.Krusche. Water quality monitoring in large reservoirs using remote sensing and neural networks. IEEE Lat Am T, 6 (5) (2008), pp. 419-423.
[8]
Q.Cao, G.Yu, Z.Qiao. Application and recent progress of inland water monitoring using remote sensing techniques. Environ Monit Assess, 195 (1) (2023), p. 125.
[9]
X.Zhu, H.Guo, J.Huang, S.Tian, W.Xu, Y.Mai. An ensemble machine learning model for water quality estimation in coastal area based on remote sensing imagery. J Environ Manage, 323 (2022), Article 116187.
[10]
L.Grbčić, S.Družeta, G.Mauša, T.Lipić, D. VukićLušić, M.Alvir, et al. Coastal water quality prediction based on machine learning with feature interpretation and spatio-temporal analysis. Environ Modell Softw, 155 (9) (2022), Article 105458.
[11]
H.Mohammed, H. MichelTornyeviadzi, R.Seidu. Emulating process-based water quality modelling in water source reservoirs using machine learning. J Hydrol, 609 (2022), Article 127675.
[12]
Y.Liu, X.Liu, B.Duan, Z.Yu, T.Cheng, L.Yu, et al. Polymer-water interaction enabled intelligent moisture regulation in hydrogels. J Phys Chem Lett, 12 (10) (2021), pp. 2587-2592.
[13]
Y.Du, Z.Zhou, J.Zhao. Multiregional building energy efficiency intelligent regulation strategy based on multi-objective optimization and model predictive control. J Clean Prod, 349 (2022), Article 131264.
[14]
X.Y.Fan, H.Q.Li, P.Yang, B.Lai. Effect of C/N ratio and aeration rate on performance of internal cycle MBR with synthetic wastewater. Desalination Water Treat, 54 (3) (2015), pp. 573-580.
[15]
M.Bourgin, E.Borowska, J.Helbing, J.Hollender, H.P.Kaiser, C.Kienle, et al. Effect of operational and water quality parameters on conventional ozonation and the advanced oxidation process O3/H2O2: kinetics of micropollutant abatement, transformation product and bromate formation in a surface water. Water Res, 122 (1) (2017), pp. 234-245.
[16]
D.Sravani. Watershed modelling using SWAT for hydrology and water quality: a review. J Indian Water Works Ass, 53 (2) (2021), pp. 123-127.
[17]
H.Hu, C.X.Wang, N.F.Ren. Modern control system. Tsinghua University Press, Beijing (2014). Chinese.
Y.C.Fang, G.Z.Lu. Nonlinear system theory. (2nd ed.), Tsinghua University Press, Beijing (2022). Chinese.
[20]
G.Kerschen, K.Worden, A.F.Vakakis, J.C.Golinval. Past, present and future of nonlinear system identification in structural dynamics. Mech Syst Signal Process, 20 (3) (2006), pp. 505-592.
[21]
S.Y.Kim, J.H.Kim, T.Jeong, K.B.Kim, H.J.Kim, K.M.Nam, et al. Accelerated decomposition of Bi2S3 nanorods in water under an electron beam: a liquid phase transmission electron microscopy study. Nanotechnology, 32 (19) (2021), Article 195702.
[22]
W.Yang, J.Song, Y.Higano, J.Tang. Exploration and assessment of optimal policy combination for total water pollution control with a dynamic simulation model. J Clean Prod, 102 (2015), pp. 342-352.
[23]
M.Grieves, J. Vickers. Digital twin: mitigating unpredictable, undesirable emergent behavior in complex systems. Transdisciplinary perspectives on complex systems: new findings and approaches, Springer, Berlin (2017), pp. 85-113.
[24]
Z.Li, E.Yu, K.Zhang, W.Gong, Y.Xia, J.Tian, et al. Water treatment effect, microbial community structure, and metabolic characteristics in a field-scale aquaculture wastewater treatment system. Front Microbiol, 11 (2020), p. 930.
[25]
N.R.Blessing, S.Benedict. Computing principles in formulating water quality informatics and indexing methods: an ample review. J Comput Theor Nanosci, 14 (4) (2017), pp. 1671-1681.
PDF
(2034KB)
