Abstract. This study presents a methodology for the optimization of a monitoring network of sensors measuring the polluting substances in an urban environment with a view to estimate an unknown emission source. The methodology was presented by coupling the Simulated Annealing algorithm with the renormalization inversion technique and the Computational Fluid Dynamics (CFD) modeling approach. Performance of a network was analyzed by reconstructing the unknown continuous point emissions using the concentration measurements from the sensors in that optimized network. This approach was successfully emission rates with the 10 and 13 sensors networks were estimated within a factor of two which are also comparable to 75% 10 from the original network. This study presents the first application of the renormalization data-assimilation approach for the optimal network design to estimate a continuous point source emission in an urban-like environment.

The aim of this study is to optimize sensor networks for fast deployment in order to reconstruct an unknown source of intentional or accidental release in local urban topography. In such emergency circumstances, only the meteorological conditions are available in real time and the network deployed must be efficient enough regardless of a source's position and intensity. To determine the optimal positions to be instrumented by the sensors, an adequate cost function is defined based on the renormalization inversion method. This function, named the entropic criterion, quantifies the amount of information contained in a network of the sensors to estimate the intensity and the location of an unknown source. The optimal design is approached as combinatorial optimization (NP‐Hard) and a stochastic algorithm (simulated annealing, SA) is employed to solve this problem. The computation is performed by coupling the CFD adjoint fields in an urban environment, the renormalization algorithm and the SA. The optimization is evaluated with 20 trials of the Mock Urban Setting Test (MUST) tracer field experiment for the reconstruction of a continuous point release in an idealized urban geometry using optimal networks of sizes 10 and 13 sensors. The process is achieved successfully and the results showed that the reduction of an original network of 40 sensors to one third (13) and one quarter (10) does not degrade the performance of this network. Also, a comparison of the optimal design efficiency based on apriori information and without apriori information about the source showed that the present entropic criterion leads to network design and performance that can accurately retrieve an unknown emission source in an urban environment.

Abstract. This study presents an optimization methodology for reducing the size of an existing monitoring network of the sensors measuring polluting substances in an urban-like environment in order to estimate an unknown emission source. The methodology is presented by coupling the simulated annealing (SA) algorithm with the renormalization inversion technique and the computational fluid dynamics (CFD) modeling approach. This study presents an application of the renormalization data-assimilation theory for optimally reducing the size of an existing monitoring network in an urban-like environment. The performance of the obtained reduced optimal sensor networks is analyzed by reconstructing the unknown continuous point emission using the concentration measurements from the sensors in that optimized network. This approach is successfully applied and validated with 20 trials of the Mock Urban Setting Test (MUST) tracer field experiment in an urban-like environment. The main results consist of reducing the size of a fixed network of 40 sensors deployed in the MUST experiment. The optimal networks in the MUST urban region are determined, which makes it possible to reduce the size of the original network (40 sensors) to ∼1/3 (13 sensors) and 1∕4 (10 sensors). Using measurements from the reduced optimal networks of 10 and 13 sensors, the averaged location errors are obtained as 19.20 and 17.42 m, respectively, which are comparable to the 14.62 m obtained from the original 40-sensor network. In 80 % of the trials with networks of 10 and 13 sensors, the emission rates are estimated within a factor of 2 of the actual release rates. These are also comparable to the performance of the original network, whereby in 75 % of the trials the releases were estimated within a factor of 2 of the actual emission rates.

This study describes a process to design a sensor network. This network could include: wireless mobile sensors deployed by first responders in hazardous material operations, stationary sensors used to protect an area against accidental, or intentional, contaminations or stationary air quality monitoring stations. The objective of the network is the estimation (localization -quantification) of releases sources. The design of such a network has an important issue in determining the optimal placement of sensors. This paper presents the first application of the renormalized data assimilation method to address this issue. It is associated with a classical optimization algorithm (simulate annealing) to solve the combinatory optimization problem consisting of finding the optimal configuration of sensors among a set of potential positions. Three scenarios, corresponding with three different cost functions, are proposed. The first one consists of optimizing the design of a network deployed in emergency situations. Experimental data from a wind tunnel experiment are used. The objective is to characterize the source to minimize error in measurement forecasts. The second one is to optimize the design of the same network but in a situation where the source can be anywhere in the domain. To that end, an entropic criterion is used. The last one consists of optimizing the design of a stationary network. The objective is to characterize the source with varying meteorological conditions (experimental meteorological data are used).

No abstract