Pollution Source Identification Problems Research Articles

Two-step AI-aided Bayesian source identification of urban-scale pollution

Poor air quality produces detrimental effects worldwide; hence, it is vital to thoroughly characterize air pollution sources and effectively address and mitigate such effects. Due to the complexity of the underlying physical processes, and uncertainties in the available observations, the air pollution source identification problem is typically cast within a Bayesian inversion framework. The latter incorporates prior knowledge and observations to characterize a release event through the posterior distribution of the source parameters. In this study, we rely on two-dimensional (2D) pollutant concentration distributions as observations, and adopt the Wasserstein (W2) distance to model the likelihood probability distribution for given emission parameters. Since the posterior distribution is estimated via random sampling that involves many forward model runs, the Bayesian framework can be computationally prohibitive for realistic urban air pollution problems that are driven by computationally demanding micro-scale flow simulations. Furthermore, computing the W2 distance is resource-intensive . In this context, we develop a computationally efficient Bayesian framework by following (i) a two-stage approach that reduces the cost of the Bayesian inversion, and (ii) an artificial intelligence (AI) approximation of the W2 distance. In the two-stage approach, a low-resolution dispersion model is run in the first stage to propose representative samples of emission parameters for final selection by the original high-resolution model in the second stage. In addition, we approximate the W2 distance using a deep neural network (DNN) to achieve an appreciable reduction in the computational cost with negligible loss in the inversion performance. We design numerical experiments to test the sensitivity of the inverse solution to the characteristics of the approximative model. The results indicate that pairing the two-stage approach with the DNN approximation of the W2 distance preserves the quality of the inverse solution, while achieving at least threefold reductions in the computational cost.

Study on source identification and source-sink relationship of LNAPLs pollution in groundwater by the adaptive cyclic improved iterative process and Monte Carlo stochastic simulation

• Biodegradation is considered in the transport of LNAPLs by adding a biodegradation model. • The VDGS method is used to optimize the hyperparameters of ResNet-18 surrogate model. • The adaptive cyclic improved iterative process is constructed to improve the solution effect of groundwater LNAPLs pollution source identification problem. • The groundwater pollution source-sink relationship is studied by using the Monte Carlo method. It is important to study the source-sink relationship of pollution. However, research on the groundwater pollution sources and pollution sinks has often been conducted separately in the past, which makes it difficult to study the source-sink relationship of pollutants. In this study, the source and sink of pollution in a site contaminated by a light non-aqueous phase liquid (LNAPL) is considered to study the relationship between them. The simulation model needs to be called repeatedly in the research process. To save time and reduce the calculation load, the ResNet-18 network structure is first utilized to create a substitute of simulation model. However, the performance of ResNet-18 surrogate model largely depends on the choice of hyperparameters. The variable-density grid search method is thus utilized to optimize the hyperparameters. In addition, the quality of training samples is an important factor to determine the accuracy of surrogate models. And the accuracy of the surrogate model will affect the effect of groundwater pollution identification. Therefore, the ResNet-18 surrogate model and the ensemble smoother algorithm are combined to construct an adaptive cyclic improved iterative process. Through the adaptive cyclic improved iterative process, the training samples, surrogate model and identification results of variables to be solved can be improved and updated together with the iterative process, to improve the solution effect of groundwater LNAPLs pollution source identification problem. Biodegradation is an important sink of LNAPL pollutants. Therefore, the biodegradation on the LNAPL pollutants in groundwater is considered by adding a biodegradation model to the original model of multi-phase flow. Finally, based on the numerical simulation models of multiphase flow and biodegradation, the Monte Carlo stochastic simulation are used to study the source-sink relationship of LNAPL pollutants. A hypothetical case based on real conditions was used to illustrate the validity of the methods. The results implied that the proposed methods can not only improve the effect of LNAPL pollution identification, but also study the source-sink relationship of LNAPL pollutant.

Groundwater pollution source identification problems with unknown aquifer parameters by ADGA approach

High-cost remediation of groundwater pollution makes it important to obtain exact information about the source. This is quite difficult to achieve in naturally ill-posed inverse problems of this kind. If the aquifer parameters are also unknown, the problem becomes even more challenging. To address this difficulty, we propose the alternating direction genetic algorithm (ADGA) approach, together with modification of the order of magnitude of the decision variables, to increase the accuracy of the results and computational efficiency. Seven scenarios were designed to test the accuracy of the proposed approach in aquifer with different properties, number of pollution sources, parameters and measurement errors. The results show that combining the ADGA approach with modification of the order of magnitude of the decision variables for identifying both groundwater pollution source and aquifer parameters significantly increases the accuracy of estimated results. The NE value for the estimated results decreased from 9.81% to 58.44% for different cases, and computation time is about half decreased. In addition, the approach is applicable in situations where concentrations of observational data with measurement error, and for multiple source locations and non-uniform media.

Simultaneous identification of groundwater pollution source location and release concentration using Artificial Neural Network

Groundwater pollution source identification problem plays an important role in designing the groundwater remediation measures. This study aims at identifying the groundwater pollutant source location and its strength given a set of concentration breakthrough curves at different locations. Feed forward three layered artificial neural network (ANN) has been used to identify these two source parameters: location, i.e., the distance of source from the observation point and strength, i.e., the release concentration. A simplified ANN architecture is achieved by presenting the available concentration breakthrough curves in a specific manner as input to the model. The database of 1750 patterns for training and testing of the model is generated employing analytical solution of one dimensional steady flow and transient contaminant transport in a homogeneous aquifer. The performance of the ANN model was evaluated using standard statistical methods. A network with architecture 22-16-2 was found to be optimum and resulted in a normalized root mean square error (NRMSE) of 0.014 for identifying the source location and NRMSE of 0.044 for identifying the source release concentration. The results show that ANN can be a very efficient tool for locating pollution sources and estimating the release concentration at source. A good ANN model performance was obtained even with a simple architecture and with a small number of input variables.

Wavelet denoising and cubic spline interpolation for observation data in groundwater pollution source identification problems

Abstract As the identified results of groundwater pollution source identification (GPSI) can influence the cost for the polluter in paying for remediating groundwater resources, it is important that the accuracy of the estimated result should be as high as possible. However, many factors can influence the result, such as noisy concentration data and incomplete concentration data. Thus, this paper is aimed at studying the difference between using the observation data before and after denoising and interpolating for solving GPSI problems. Four kinds of noise level and 20 groups of missing data were designed to test the performance of wavelet denoising and cubic spline interpolation, respectively. The results show that the denoising process can improve the estimated result for the GPSI problem, and the higher the noise level, the stronger this effect. In terms of interpolation, more accurate results can be made after interpolating if the missing data belong to the period after the source releases the pollutant. If the missing data are from when the pollution source is active, interpolation cannot help increase the estimated performance.

An Effective Kalman Filter-Based Method for Groundwater Pollution Source Identification and Plume Morphology Characterization

The identification of unknown groundwater pollution sources and the characterization of pollution plume remains a challenging problem. In this study, we addressed this problem by a linked simulation-optimization approach. This approach couples a contaminant transport simulation model with a Kalman filter-based method to identify groundwater pollution source and characterize plume morphology. In the proposed methodology, the concentration field library, the covariance reduction with a Kalman filter, an alpha-cut technique of fuzzy set, and a linear programming model are integrated for solving this inverse problem. The performance of this methodology is evaluated on an illustrative groundwater pollution source identification problem. The evaluation considered the random hydraulic conductivity filed, erroneous monitoring data, a prior information shortage of potential pollution sources, and an unexpected and unknown pumping well. The identified results indicate that, under these conditions, the proposed Kalman filter-based optimization model can give satisfactory estimations to pollution sources and plume morphology for domains with small and moderate heterogeneity but cannot validate the transport in the relatively high heterogeneous field.

Mixed integer optimization approach to groundwater pollution source identification problems

ABSTRACTIn this study, we consider an approach that employs 0–1 type integer programming in a simulation-optimization model to identify the characteristics of groundwater pollutant sources. The advantage of the proposed solution model is that it reduces the dimensions of the unknown variable, where 0–1 type integer variables denote the source position. If the estimated location variable is equal to 0, this means that the corresponding location is not the actual source location and there is no need to estimate its source flux. If the estimated location variable is equal to 1, this means that the corresponding location is the actual source location and the source flux needs to be identified. We evaluated the performance of the proposed model based on four scenarios with different potential source location sets, incomplete data conditions, and noisy data conditions. The results obtained indicated that the proposed solution model is an effective approach for solving groundwater pollution source identification problems.

Location and release time identification of pollution point source in river networks based on the Backward Probability Method

The pollution of rivers due to accidental spills is a major threat to environment and human health. To protect river systems from accidental spills, it is essential to introduce a reliable tool for identification process. Backward Probability Method (BPM) is one of the most recommended tools that is able to introduce information related to the prior location and the release time of the pollution. This method was originally developed and employed in groundwater pollution source identification problems. One of the objectives of this study is to apply this method in identifying the pollution source location and release time in surface waters, mainly in rivers. To accomplish this task, a numerical model is developed based on the adjoint analysis. Then the developed model is verified using analytical solution and some real data. The second objective of this study is to extend the method to pollution source identification in river networks. In this regard, a hypothetical test case is considered. In the later simulations, all of the suspected points are identified, using only one backward simulation. The results demonstrated that all suspected points, determined by the BPM could be a possible pollution source. The proposed approach is accurate and computationally efficient and does not need any simplification in river geometry and flow. Due to this simplicity, it is highly recommended for practical purposes.

A hybrid simulation–optimization approach for solving the areal groundwater pollution source identification problems

Summary In this study, a new simulation–optimization approach is proposed for solving the areal groundwater pollution source identification problems which is an ill-posed inverse problem. In the simulation part of the proposed approach, groundwater flow and pollution transport processes are simulated by modeling the given aquifer system on MODFLOW and MT3DMS models. The developed simulation model is then integrated to a newly proposed hybrid optimization model where a binary genetic algorithm and a generalized reduced gradient method are mutually used. This is a novel approach and it is employed for the first time in the areal pollution source identification problems. The objective of the proposed hybrid optimization approach is to simultaneously identify the spatial distributions and input concentrations of the unknown areal groundwater pollution sources by using the limited number of pollution concentration time series at the monitoring well locations. The applicability of the proposed simulation–optimization approach is evaluated on a hypothetical aquifer model for different pollution source distributions. Furthermore, model performance is evaluated for measurement error conditions, different genetic algorithm parameter combinations, different numbers and locations of the monitoring wells, and different heterogeneous hydraulic conductivity fields. Identified results indicated that the proposed simulation–optimization approach may be an effective way to solve the areal groundwater pollution source identification problems.

Identification of Unknown Groundwater Pollution Sources and Determination of Optimal Well Locations Using ANN-GA Based Simulation-Optimization Model

The linked simulation-optimization model can be used for solving a complex groundwater pollution source identification problem. Advanced simulators have been developed and successfully linked with numerous optimization algorithms for identification of groundwater pollution sources. However, the identification of pollution sources in a groundwater aquifer using linked simulation-optimization model has proven to be computationally expensive. To overcome this computational burden, an approximate simulator, the artificial neural network (ANN) model can be used as a surrogate model to replace the complex time-consuming numerical simulation model. However, for large-scale aquifer system, the performance of the ANN-based surrogate model is not satisfactory when a single ANN model is used to predict the concentration at different observation locations. In such a situation, the model efficiency can be enhanced by developing separate ANN model for each of the observation locations. The number of ANN models is equal to the number of observation wells in the aquifer. As a result, the complexity of the ANN-based simulation-optimization model will be related to the number of observation wells. Thus, this study used a modified formulation to find out the optimal numbers of observation wells which will eventually reduce the computational time of the model. The performance of the ANN-based simulation-optimization model is evaluated by identifying the groundwater pollutant sources of a hypothetical study area. The limited evaluation shows that the model has the potential for field application.

Development of Unknown Pollution Source Identification Models Using GMS ANN–Based Simulation Optimization Methodology

AbstractThe detection of groundwater pollution sources is an important but a very difficult task. The location and magnitude of groundwater pollution sources can be identified using inverse optimization technique. The technique is also known as simulation-optimization approach where the aquifer simulation model is incorporated with the optimization model for finding the unknown pollution sources in an aquifer. The efficiency of the simulation-optimization model is highly related to the performance of the simulation model. This study develops three improved methodologies for identification of unknown groundwater pollution sources. In the first approach, the groundwater modeling system (GMS) is linked with the optimization model for solving source identification problem. The optimization model is solved using the direct-search method. The incorporation of GMS with the optimization model allows for the solving of a bigger real-world pollution source identification problem. The challenge of this approach is t...

Encapsulating the Role of Solution Response Space Roughness on Global Optimal Solution: Application in Identification of Unknown Groundwater Pollution Sources

A major challenge of any optimization problem is to find the global optimum solution. In a multi-dimensional solution space which is highly non-linear, often the optimization algorithm gets trapped around some local optima. Optimal Identification of unknown groundwater pollution sources poses similar challenges. Optimization based methodology is often applied to identify the unknown source characteristics such as location and flux release history over time, in a polluted aquifer. Optimization based models for identification of these characteristics of unknown ground-water pollution sources rely on comparing the simulated effects of candidate solutions to the observed effects in terms of pollutant concentration at specified sparse spatiotemporal locations. The optimization model minimizes the difference between the observed pollutant concentration measurements and simulated pollutant concentration measurements. This essentially constitutes the objective function of the optimization model. However, the mathematical formulation of the objective function can significantly affect the accuracy of the results by altering the response contour of the solution space. In this study, two separate mathematical formulations of the objective function are compared for accuracy, by incorporating different scenarios of unknown groundwater pollution source identification problem. Simulated Annealing (SA) is used as the solution algorithm for the optimization model. Different mathematical formulations of the objective function for minimizing the difference between the observed and simulated pollutant concentration measurements show different levels of accuracy in source identification results. These evaluation results demonstrate the impact of objective function formulation on the optimal identification, and provide a basis for choosing an appropriate mathematical formulation for unknown pollution source identification in contaminated aquifers.

An almost-parameter-free harmony search algorithm for groundwater pollution source identification

The spatiotemporal characterization of unknown sources of groundwater pollution is frequently encountered in environmental problems. This study adopts a simulation-optimization approach that combines a contaminant transport simulation model with a heuristic harmony search algorithm to identify unknown pollution sources. In the proposed methodology, an almost-parameter-free harmony search algorithm is developed. The performance of this methodology is evaluated on an illustrative groundwater pollution source identification problem, and the identified results indicate that the proposed almost-parameter-free harmony search algorithm-based optimization model can give satisfactory estimations, even when the irregular geometry, erroneous monitoring data, and prior information shortage of potential locations are considered.

A Regularized Integral Equation Scheme for Three-Dimensional Groundwater Pollution Source Identification Problems

A Regularized Integral Equation Scheme for Three-Dimensional Groundwater Pollution Source Identification Problems

A linked simulation–optimization model for solving the unknown groundwater pollution source identification problems

This study proposes a linked simulation–optimization model for solving the unknown groundwater pollution source identification problems. In the proposed model, MODFLOW and MT3DMS packages are used to simulate the flow and transport processes in the groundwater system. These models are then integrated with an optimization model which is based on the heuristic harmony search (HS) algorithm. In the proposed simulation–optimization model, the locations and release histories of the pollution sources are treated as the explicit decision variables and determined through the optimization model. Also, an implicit solution procedure is proposed to determine the optimum number of pollution sources which is an advantage of this model. The performance of the proposed model is evaluated on two hypothetical examples for simple and complex aquifer geometries, measurement error conditions, and different HS solution parameter sets. Identified results indicated that the proposed simulation–optimization model is an effective way and may be used to solve the inverse pollution source identification problems.