Optimal Sensor Placement Research Articles

Seismic Damage Identification Using a Combination of Mixture of Gaussian and Harris Hawks Optimization

ABSTRACTThe present study aims to identify damage in structures using seismic responses and a two‐stage method based on mixture of Gaussian (MOG) and Harris hawks optimization (HHO). Two‐dimensional (2‐D) frame structures under seismic loads are simulated using the finite element method (FEM). The placement of sensors is optimized using a combination of an iterated improved reduced system (IIRS) and the binary differential evolution (BDE) algorithm. The MOG classifier is trained to find the damaged story and damaged element type (i.e., whether it is a beam or a column) using the nodal acceleration responses at the optimal sensor placement of the seismically loaded structure. Thus, as the first step, the possibly damaged elements are located through the classifier. Then, in the second step, the damage is accurately located and quantified by HHO algorithm. The performance of the proposed method is assessed using the numerical results of 2‐D frames of 7 and 14 stories in different damage scenarios with and without considering noise. As a result, the efficiency of the proposed method for the seismic damage identification of 2‐D frames is revealed. Sensor positions are well optimized, and it causes the method to be highly effective. Moreover, MOG correctly finds a damaged story and identifies whether a damaged element is a beam or a column. Damage in the presence of noise is also localized and quantified precisely by HHO.

A value-of-information-based optimal sensor placement design framework for cost-effective structural health monitoring (with application to miter gate monitoring)

Structural health monitoring (SHM) involves collecting information to assess the health of a structure, typically to guide risk-informed maintenance decision-making or predict limit state behavior throughout its lifespan. Although the value of information (VoI) obtained from an SHM system can facilitate improved decision-making, it is important to estimate its overall utility by considering the costs involved in designing, developing, installing, maintaining, and operating the system. A feasible SHM system provides greater expected returns resulting from data-informed lifecycle management decisions than the cost of the SHM system design, fabrication, deployment, operation, and maintenance. That is, since data acquisition is a precursor to data-informed decision-making, the design of an SHM system governs its feasibility. Such cost-benefit analyses are a current topic of research in the SHM community. One approach that has been proposed for these analyses is a preposterior decision analysis using the VoI metric. In this paper, we propose a sensor optimization framework that maximizes the VoI over the structure’s lifecycle, constrained by a pre-decided maintenance policy. We use two different VoI metrics: the traditional expected VoI and a gambling-theory-inspired normalized expected-savings-to-investmentrisk ratio. We introduce three time-normalized, unitless VoI metrics that are valuable for evaluating the performance of an SHM design over an extended period. Furthermore, we consider the effect of different risk profiles on the overall optimal sensor design, recognizing that the cost of decision-making depends on the utility and risk perception of the decision-maker. We conduct a detailed analysis on the marginal utility gain of gathered information as we add more sensors, observing the utility to diminish in line with the law of diminishing returns as the information content increases. This framework is applied to the design of an SHM system used for monitoring miter gates.

Sensor Placement Optimization in Sewer Networks: Machine Learning–Based Source Identification Approach

Sensor Placement Optimization in Sewer Networks: Machine Learning–Based Source Identification Approach

Data-driven parallel optimization method for sensor placement and state estimation based on generative adversarial principle in unobservable systems

Abstract Data-Driven State Estimation (DDSE) is characterized by low measurement requirements and high estimation accuracy, which provides the possibility of stable and efficient estimation for unobservable distribution systems with low measurement quantities. However, the lack of an efficient and flexible sensor placement method with DDSE at this stage hinders the development and application of DDSE in distribution networks. To address this issue, a data-driven parallel optimization method for sensor placement and state estimation based on the generative adversarial principle in unobservable systems is proposed in this paper. This method transforms sensor placement and state estimation into an adversarial learning model, solvable through gradient descent. The sensor placement optimization strategy generates a cost-effective placement scheme, while DDSE evaluates the sensor scheme, provides improvement suggestions, and enhances its estimation accuracy. Through adversarial training, the proposed scheme yields a significantly reduced number of sensor placement schemes compared to traditional methods, while maintaining high precision of DDSE. Simulation results on the IEEE-33 node system validate the effectiveness of the proposed approach.

Underwater Line Monitoring Using Optimally Placed Inclinometers

Underwater monitoring presents challenges related to maintaining a continuous power supply and communication, necessitating the use of a smaller number of sensors to effectively cover the entire line. An underwater line tracking method is proposed to evaluate global behaviors and stresses in real time. The method employs angles at several points on the line, as well as displacements and curvatures at both ends. In this method, any line displacement, angle, and curvature are expressed as Fourier series, and Fourier coefficients are obtained by utilizing sensor data. Then, the behavior of any line location is assessed. In addition, to reduce the number of sensors and improve accuracy, optimal inclinometer locations are determined by a genetic algorithm. The proposed line tracking algorithm was validated through two numerical examples; one with an inclined tunnel and one with a marine steel catenary riser attached to a Floating Production Storage and Offloading (FPSO) vessel. Through these examples, the proposed algorithm was proven to capture global behaviors accurately when optimally located sensors are used. In the riser monitoring case, the optimized sensor placement with eight intermediate sensors achieved an average mean distance error of 1.91 m, which is lower than the 2.65 m error obtained with ten intermediate sensors without optimization.

Uncertainty propagation of initial conditions in thermal models

The operation of machine tools often demands a highly accurate knowledge of the tool center point’s (TCP) position. The displacement of the TCP over time can be inferred from coupled thermal models, which comprise a set of geometrically coupled heat equations. Each of these equations represents the temperature in part of the machine, and they are often formulated on complicated geometries. The accuracy of the TCP prediction depends highly on the accuracy of the model parameters, such as heat exchange parameters, and the initial temperature. Thus it is of utmost interest to determine the influence of these parameters on the TCP displacement prediction. In turn, the accuracy of the parameter estimate is essentially determined by the measurement accuracy and the sensor placement. Determining the accuracy of a given sensor configuration is a key prerequisite of optimal sensor placement. In this paper, we develop a thermal model for the discretization of realistic machine tool. On top of this model we propose two numerical algorithms to evaluate any given thermal sensor configuration with respect to its accuracy. We compute the posterior variances from the posterior covariance matrix with respect to an uncertain initial temperature field. The full matrix is dense and potentially very large, depending on the size of the discretized model. We first present a way to compute this approximation, which relies on the computation of the model sensitivities with respect to the initial values. Additionally, we present a low-rank tensor method which exploits the underlying system structure. We compare the efficiency of both algorithms with respect to runtime and memory requirements and discuss their respective advantages with regard to optimal sensor placement problems.

Coverage optimization for IoT-based network using Monarch Butterfly Optimization with greedy strategy and self-adaptive crossover operator

In the dynamic development of technology, the Internet of Things (IoT) stands as a driving force of innovation and opening new frontiers. Data collection and processing in IoT systems provide services to decision making in various fields including home automation, disaster management, healthcare, and so on. In such services, the IoT has certain limitations to perform the required tasks and it can cause extreme consequences in the network. One of the main challenge in IoT is to ensure maximum coverage by optimally placing the sensors. To address the coverage problem, a meta-heuristic algorithm Monarch Butterfly Optimization with greedy strategy and self-adaptive crossover operator (GCMBO) is proposed for Optimal Sensor Placement (OSP) in 3D monitoring region. The performance of proposed algorithm is evaluated by a series of simulation and it provide efficient results when compared to other similar existing optimization algorithms. Additionally, a statistical analysis (ANOVA and LSD post-hoc) is exhibited from the comparative results.

Assessing the design of integrated methane sensing networks

Abstract While methane is the second largest contributor to global warming after carbon dioxide, it has a larger warming effect over a much shorter lifetime. Despite accelerated technological efforts to radically reduce global carbon dioxide emissions, rapid reductions in methane emissions are needed to limit near-term warming. Being primarily emitted as a byproduct from agricultural activities and energy extraction, methane is currently monitored via bottom-up (i.e., activity level) or top-down (via airborne or satellite retrievals) approaches. However, significant methane leaks remain undetected and emission rates are challenging to characterize with current monitoring frameworks. In this paper, we study the design of a layered monitoring approach that combines bottom-up and top-down approaches as an integrated sensing network. By recognizing that varying meteorological conditions and emission rates impact the efficacy of bottom-up monitoring, we develop a probabilistic approach to optimal sensor placement in its bottom-up network. Subsequently, we derive an inverse Bayesian framework to quantify the improvement that a design-optimized integrated framework has on emission-rate quantifications and their uncertainties. We find that under realistic meteorological conditions, the overall error in estimating the true emission rates is approximately 1.3 times higher, with their uncertainties being approximately 2.4 times higher, when using a randomized network over an optimized network, highlighting the importance of optimizing the design of integrated methane sensing networks. Further, we find that optimized networks can improve scenario coverage fractions by more than a factor of 2 over experimentally-studied networks, and identify a budget threshold beyond which the rate of optimized-network coverage improvement exhibits diminishing returns, suggesting that strategic sensor placement is also crucial for maximizing network efficiency.

Data-Driven Sparse Sensor Placement Optimization on Wings for Flight-By-Feel: Bioinspired Approach and Application.

Flight-by-feel (FBF) is an approach to flight control that uses dispersed sensors on the wings of aircraft to detect flight state. While biological FBF systems, such as the wings of insects, often contain hundreds of strain and flow sensors, artificial systems are highly constrained by size, weight, and power (SWaP) considerations, especially for small aircraft. An optimization approach is needed to determine how many sensors are required and where they should be placed on the wing. Airflow fields can be highly nonlinear, and many local minima exist for sensor placement, meaning conventional optimization techniques are unreliable for this application. The Sparse Sensor Placement Optimization for Prediction (SSPOP) algorithm extracts information from a dense array of flow data using singular value decomposition and linear discriminant analysis, thereby identifying the most information-rich sparse subset of sensor locations. In this research, the SSPOP algorithm is evaluated for the placement of artificial hair sensors on a 3D delta wing model with a 45° sweep angle and a blunt leading edge. The sensor placement solution, or design point (DP), is shown to rank within the top one percent of all possible solutions by root mean square error in angle of attack prediction. This research is the first to evaluate SSPOP on a 3D model and the first to include variable length hairs for variable velocity sensitivity. A comparison of SSPOP against conventional greedy search and gradient-based optimization shows that SSPOP DP ranks nearest to optimal in over 90 percent of models and is far more robust to model variation. The successful application of SSPOP in complex 3D flows paves the way for experimental sensor placement optimization for artificial hair-cell airflow sensors and is a major step toward biomimetic flight-by-feel.

Framework for Data‒Driven Fault Diagnosis of Numerical Spacecraft Propulsion Systems

The increasing complexity of deep space missions introduces significant challenges in maintaining spacecraft health, particularly in the propulsion systems, due to the inherent communication delays with Earth. This research proposes a novel framework for the autonomous, data-driven fault diagnosis of spacecraft propulsion systems. Leveraging data generated from a spacecraft propulsion system simulation model. The study addresses the limitations imposed by computational resources and sensor installation constraints through Sequential Forward Selection (SFS) for optimized sensor placement and feature selection. The framework's effectiveness is demonstrated through implementation on a microcomputer, showing promising results in terms of diagnostic accuracy and processing speed, thus highlighting its potential for onboard spacecraft application. This study not only advances the autonomous capabilities of spacecraft in deep space but also contributes to the broader field of Prognostics and Health Management (PHM) by providing a scalable, efficient approach to fault diagnosis in critical spacecraft systems. The proposed framework demonstrates a promising approach to optimizing diagnostic tasks for spacecraft systems. However, the trade-offs observed necessitate a careful consideration of task-specific requirements and the potential need for adjustments to maintain a high level of accuracy alongside computational efficiency.

Analyzing Optimal Wearable Motion Sensor Placement for Accurate Classification of Fall Directions.

Falls represent a significant risk factor, necessitating accurate classification methods. This study aims to identify the optimal placement of wearable sensors-specifically accelerometers, gyroscopes, and magnetometers-for effective fall-direction classification. Although previous research identified optimal sensor locations for distinguishing falls from non-falls, limited attention has been given to the classification of fall direction across different body regions. This study assesses inertial measurement unit (IMU) sensors placed at 12 distinct body locations to determine the most effective positions for capturing fall-related data. The research was conducted in three phases: first, comparing classifiers across all sensor locations to identify the most effective; second, evaluating performance differences between sensors placed on the left and right sides of the body; and third, exploring the efficacy of combining sensors from the upper and lower body regions. Statistical analyses of the results for the most effective classifier model demonstrate that the support vector machine (SVM) is more effective than other classifiers across all sensor locations, with statistically significant differences in performance. At the same time, the comparison between the left and right sensor locations shows no significant performance differences within the same anatomical areas. Regarding optimal sensor placement, the findings indicate that sensors positioned on the pelvis and upper legs in the lower body, as well as on the shoulder and head in the upper body, were the most effective results for accurate fall-direction classification. The study concludes that the optimal sensor configuration for fall-direction classification involves strategically combining sensors placed on the pelvis, upper legs, and lower legs.

Sensor placement method for water distribution networks based on sampling of non-bandlimited graph signals

Monitoring water distribution networks (WDNs) requires careful consideration of sensor placement, which is essential for obtaining comprehensive information about the network. A natural graphical structure underlies WDN, making graph sampling theory advantageous for selecting monitoring nodes. However, graph sampling theory is only applied only to restrictive band-limited signals, while the pressure data of WDN is a restrictive non-band-limited signal. To address this issue, this paper presents an approximate conversion method for transforming non-band-limited signals into band-limited signals, accompanied by an optimal spectrum threshold formula. This formula is used to perform spectral screening in the graph frequency domain, effectively converting non-band-limited signals into band-limited signals that preserve the major frequency components while ignoring smaller-value frequency components. By sampling band-limited signal, we identify sampling nodes that perfectly recover the signal. These sampling nodes act as monitoring nodes that can perform a comprehensive inspection of the WDN and accurately locate leaks. The accuracy of our method in recovering the signal and locating the leak is demonstrated by comparing it with two existing sensor placement optimization methods.

Strategic deployment in the deep: Principled underwater sensor placement optimization with three-dimensional acoustic map.

Underwater acoustic sensors are vital for monitoring marine environments and detecting targets, but their optimal placement presents challenges, particularly in deep-sea environments. This paper addresses the question of determining the optimal sensor placement in a specific ocean region through a principled optimization approach. While previous studies mainly utilized heuristic algorithms without exploiting problem-specific structures, this work explores leveraging the complex three-dimensional acoustic environment through principled modeling and tailored optimization. Specifically, intricate three-dimensional multi-directional acoustic maps are constructed for each sensor. Based on these maps, the sensor placement problem is then cast as an integer linear programming, allowing the study to leverage established theoretical results from operations research. Additionally, an alternative algorithm with its performance indicator is presented to find near-optimal solutions efficiently and can empirically reach over 99% coverage of the optimal solution. Experimental results using real-life data from the South China Sea demonstrate the effectiveness of the proposed approach in achieving much larger detection coverage compared to random and empirical strategies. Notably, the alternative fast algorithm approaches the optimal solution in significantly less time. Furthermore, experiments show that any further simplification of this approach leads to the performance degradation.

A physics-driven sensor placement optimization methodology for temperature field reconstruction

Perceiving the global field from sparse sensors has been a grand challenge in the monitoring, analysis, and design of physical systems. In this context, sensor placement optimization is a crucial issue. Most existing works require large and sufficient data to construct data-based criteria, which are intractable in data-free scenarios without numerical and experimental data. To this end, we propose a novel physics-driven sensor placement optimization (PSPO) method for temperature field reconstruction using a physics-based criterion to optimize sensor locations. In our methodological framework, we firstly derive the theoretical upper and lower bounds of the reconstruction error under noise scenarios by analyzing the optimal solution, proving that error bounds correlate with the condition number determined by sensor locations. Furthermore, the condition number, as the physics-based criterion, is used to optimize sensor locations by the genetic algorithm. Finally, the best sensors are validated by reconstruction models, including non-invasive end-to-end models, non-invasive reduced-order models, and physics-informed models. Experimental results, both on a numerical and an application case, demonstrate that the PSPO method significantly outperforms random and uniform selection methods, improving the reconstruction accuracy by nearly an order of magnitude. Moreover, the PSPO method can achieve comparable reconstruction accuracy to the existing data-driven placement optimization methods.

Sparse regularized graph pooling network optimal sensor placement method for diesel engine vibration fault perception system

Vibration sensor network optimization increases monitoring effectiveness and reduces sensor quantity and transmission burden. However, the traditional model-driven methods depend on precise finite element models, challenging for complex machinery. This paper presents a data-driven approach using the Sparse Regularized Graph Pooling Network (SRGPN), which conceptualizes sensor networks as graphs and uses graph pooling to identify optimal sensor combinations. A sparsity regularization term related to the number of sensors is included in the loss function, aiming for the minimal yet effective sensor combination. Additionally, a monitoring capability metric suited for diesel engines with multi-source impulse signals is proposed, reflecting the sensors’ monitoring capacity. Validated through simulations and tests on a diesel engine test bench, the results show that SRGPN optimally places sensors near excitation sources, balancing sensor count and monitoring needs. This approach shows potential for optimizing sensor placements in condition monitoring.

Greedy selection of optimal location of sensors for uncertainty reduction in seismic moment tensor inversion

We address an optimal sensor placement problem through Bayesian experimental design for seismic full waveform inversion for the recovery of the associated moment tensor. The objective is that of optimally choosing the location of the sensors (stations) from which to collect the observed data. The Shannon expected information gain is used as the objective function to search for the optimal network of sensors. A closed form for such objective is available due to the linear structure of the forward problem, as well as the Gaussian modeling of the observational errors and prior distribution. The resulting problem being inherently combinatorial, a greedy algorithm is deployed to sequentially select the sensor locations that form the best network for learning the moment tensor. Numerical results are presented to display the optimal network of sensors and how the uncertainty in the inferred seismic moment tensor contracts. The scenario of full three-dimensional velocity models or unknown earthquake source locations is treated as nuisance uncertainty, contributing to the overall uncertainty without being the focus of the inversion. This is addressed using a consensus approach over a set of realizations of the nuisance parameter. We analyzed the resulting network of stations for the moment tensor inversion under model misspecification, which reflects realistic data-generating processes.

Assessment of postural sway with accelerometers: A comparison with force plate measures

Introduction & Purpose Postural control is essential for maintaining balance during static and dynamic tasks. It is typically measured by centre of pressure (CoP) deviations using a ground reaction force plate (FP). Alternatively, more affordable and accessible accelerometer-based measures have been proposed. However, no consensus on the best protocol exists (Pinho et al., 2019). This study investigates the feasibility of assessing postural sway using linear acceleration data, considering sensor placement, task difficulty and analysed parameters. Methods The University of Vienna Ethics Committee approved this study (Ref. Nr. 01108). Ten healthy adults (6 M/4 F; 37.9 ± 14.8 years; BMI = 23.3 ± 2.3) performed eleven balance tasks, seven easy tasks from the Berg Balance Scale and four harder tasks with eyes closed, each lasting 20 seconds. Postural sway was assessed using a FP and five inertial measurement units (IMUs) attached to the upper spine (C7), lower spine (L1), sacrum and both thighs. Data were synchronously collected at 1500 Hz, zeroed, band pass filtered (3rd order Butterworth, 0.3-10 Hz; Martinez-Mendez et al., 2012), and cropped to 19 s. Outcomes included CoP path length, root mean square of sway amplitude in the anteroposterior (AP) and mediolateral (ML) directions for the FP, and RMS of the linear acceleration in AP and ML directions, along with their sum, for the ACC. Combinations of the two AP, two ML and two overall parameters were analysed by winsorised Pearson correlations and Bland Altman limits of agreement of z-scores for each task and for each sensor site. Fisher's Z-transformed correlation coefficients and limits of agreement were compared across task difficulty, parameter combinations, and sensor placement sites using a 3-way ANOVA. Results Pearson coefficients correlating the ACC parameters with the FP parameters averaged 0.60 (95% CI [0.54, 0.66]) for the easy tasks and 0.78 (95% CI [0.726, 0.84]) for the hard tasks, with no significant differences between sensor sites. A significant interaction (p < 0.001) was found between task difficulty and parameter combination. Simple main effect analysis showed no difference between correlations of AP, ML and overall parameters for hard tasks but for easy tasks, where overall parameters showed the highest correlation, followed by ML and then AP parameters, all differing significantly from each other. The Bland Altman mean difference of z-scores between FP and ACC parameters was on average -0.002 (95% CI [-0.027, 0.024]). The lower and upper limits of agreement were on average -0.72 (95% CI [-0.84, -0.62]) and 0.73 (95% CI [0.62, 0.83]) respectively. Discussion The correlations between FP and ACC were higher during more challenging tasks, while the limits of agreement were broader and less precise in these tasks (data not shown). No optimal sensor placement site was identified as correlation and agreement between the two methods did not differ significantly between sites. The broad limits of agreement suggest relatively low precision of ACC measures compared to FP. Overall parameters often outperformed ML and AP parameters in terms of correlation and precision. Conclusion Accelerometer-based measures of postural sway showed high correlations with force plate measures but low precision when tasks where difficult enough to challenge participants’ balance. Sensor placement site did not influence correlation and precision. References Pinho, A. S., Salazar, A. P., Hennig, E. M., Spessato, B. C., Domingo, A., & Pagnussat, A. S. (2019). Can we rely on mobile devices and other gadgets to assess the postural balance of healthy individuals? A systematic review. Sensors, 19(13), Article 13. https://doi.org/10.3390/s19132972 Martinez-Mendez, R., Sekine, M., & Tamura, T. (2012). Postural sway parameters using a triaxial accelerometer: Comparing elderly and young healthy adults. Computer Methods in Biomechanics and Biomedical Engineering, 15(9), 899–910. https://doi.org/10.1080/10255842.2011.565753

Optimal Sensor Placement and Multimodal Fusion for Human Activity Recognition in Agricultural Tasks

This study examines the impact of sensor placement and multimodal sensor fusion on the performance of a Long Short-Term Memory (LSTM)-based model for human activity classification taking place in an agricultural harvesting scenario involving human-robot collaboration. Data were collected from twenty participants performing six distinct activities using five wearable inertial measurement units placed at various anatomical locations. The signals collected from the sensors were first processed to eliminate noise and then input into an LSTM neural network for recognizing features in sequential time-dependent data. Results indicated that the chest-mounted sensor provided the highest F1-score of 0.939, representing superior performance over other placements and combinations of them. Moreover, the magnetometer surpassed the accelerometer and gyroscope, highlighting its superior ability to capture crucial orientation and motion data related to the investigated activities. However, multimodal fusion of accelerometer, gyroscope, and magnetometer data showed the benefit of integrating data from different sensor types to improve classification accuracy. The study emphasizes the effectiveness of strategic sensor placement and fusion in optimizing human activity recognition, thus minimizing data requirements and computational expenses, and resulting in a cost-optimal system configuration. Overall, this research contributes to the development of more intelligent, safe, cost-effective adaptive synergistic systems that can be integrated into a variety of applications.

Variation of Seismocardiogram-Derived Cardiac Time Intervals and Heart Rate Variability Metrics Across the Sternum

Abstract Cardiac time intervals (CTIs) are vital indicators of cardiac health and can be noninvasively assessed using a combination of electrocardiography (ECG) and seismocardiography (SCG), a method of capturing cardiac-induced chest vibrations via accelerometers. SCG signals can be measured from different chest locations. However, more investigations are needed to evaluate the impact of sensor placement on SCG-derived cardiac parameters. This study investigates the effect of accelerometer placement along the sternum on SCG-derived CTI estimations and heart rate variability (HRV) parameters. A semi-automated algorithm was developed to detect SCG fiducial points and seven CTIs from thirteen healthy individuals. Comparative analysis with manually selected peaks and gold-standard ECG was conducted to assess fiducial point detection accuracy. Results indicate the highest recall and precision in aortic valve opening (0.84–1.00 and 0.96–1.00, respectively) and mitral valve closure (0.77–1.00 and 0.93–1.00, respectively) detection. Aortic valve closure (0.43–1.00 and 0.61–1.00, respectively) and mitral valve opening (0.64–1.00 and 0.91–1.00, respectively) detection, although slightly less accurate due to signal intensity variations, demonstrated overall effectiveness compared to manually selected peaks. Furthermore, SCG-derived heart rates showed a high correlation coefficient (r > 0.9) with the gold-standard ECG heart rates. Single-factor ANOVA revealed significant differences (p < 0.05) in SCG-derived CTI estimations based on sensor locations on the sternum, highlighting the importance of sensor placement for accurate assessments.

Sensor placement for optimal aerodynamic data fusion

Aircraft design is recently evolving towards a digital twin representation that involves many heterogeneous data sources. The aerodynamic development of aircraft usually incorporates data from computational fluid dynamics simulations, wind tunnel testing, and flight tests. All of these data sources have their advantages and disadvantages, which can optimally be combined using data fusion methods. However, the quality of the data fusion result strongly depends on the experimental design, i.e. the placement of discrete sensors. Therefore, an optimized sensor placement is essential for data fusion applications, as the number of sensors is limited. This work presents a sensor placement strategy for the widely used Gappy proper orthogonal decomposition data fusion methodology. The sensor placement relies on a Bayesian formulation of the data fusion, allowing accurate error estimates. Based on the Bayesian posterior, a utility function characterizes the quality of the fused result by quantifying the expected information gain for the proper orthogonal decomposition coefficients. As the optimization of the sensor locations involves a complex combinatorial problem, we introduce an efficient genetic algorithm for this task. The method is demonstrated on a two-dimensional airfoil and the NASA Common Research Model with synthetic measurement errors. For both test cases, an optimal sensor placement results in smaller reconstruction errors than a conventional layout. The Bayesian approach leads, in most cases, to more accurate reconstructions and is more versatile than other well-established sensor placement methods. The proposed genetic algorithm finds better optima with significantly fewer function evaluations than the widely used greedy algorithms.