Toward the autonomous flushing of building plumbing: Characterizing oxidation-reduction potential and temperature sensor dynamics.

Manual flushing of building plumbing is commonly used to address water quality issues that arise from water stagnation. Autonomous flushing informed by sensors has the potential to aid in the management of building plumbing, but a number of knowledge gaps hinder its application. This study evaluates autonomous flushing of building plumbing with online sensor and actuator nodes deployed under kitchen sinks in five residential houses. Online oxidation-reduction potential (ORP) and temperature data were collected for nine weeks during the winter and summer in houses with both free chlorine and chloramine. ORP levels in houses with free chlorine residuals decreased after overnight stagnation. The overnight decrease in ORP was not observed when tap water was automatically flushed for five minutes at 6:00 h every morning. ORP levels in houses with chloramine residuals did not decrease consistently after overnight stagnation, and daily automated flushes did not have an observable effect on the ORP signals. Additional laboratory experiments were carried out to evaluate ORP signals during chlorine decay and after incremental changes in chlorine, as would be expected in building plumbing conditions. Results from the lab and field deployments suggest on-line ORP sensors may be used to detect free chlorine decay due to stagnating water, but are not as effective in detecting chloramine decay. However, field results also suggest ORP may not respond as expected on a timely manner after free chlorine or chloramine have been restored, hindering their applicability in developing control algorithms. In this paper we tested twice-daily five-minute automatic flushing and found that it counteracts water quality degradation associated with overnight stagnation in free chlorine systems. An automatic sensor-based flushing is proposed using online temperature sensor data to determine when flushing has reached water from the main. The results suggest that flushing informed by temperature sensors can reduce the flushing time by 46 % compared to the preset five-minute static flush.

Machine Learning-Assisted, Process-Based Quality Control for Detecting Compromised Environmental Sensors.

Machine learning (ML) techniques promise to revolutionize environmental research and management, but collecting the necessary volumes of high-quality data remains challenging. Environmental sensors are often deployed under harsh conditions, requiring labor-intensive quality assurance and control (QAQC) processes. The need for manual QAQC is a major impediment to the scalability of these sensor networks. Existing techniques for automated QAQC make strong assumptions about noise profiles in the data they filter that do not necessarily hold for broadly deployed environmental sensors, however. Toward the goal of increasing the volume of high-quality environmental data, we introduce an ML-assisted QAQC methodology that is robust to low signal-to-noise ratio data. Our approach embeds sensor measurements into a dynamical feature space and trains a binary classification algorithm (Support Vector Machine) to detect deviation from expected process dynamics, indicating whether a sensor has become compromised and requires maintenance. This strategy enables the automated detection of a wide variety of nonphysical signals. We apply the methodology to three novel data sets produced by 136 low-cost environmental sensors (stream level, drinking water pH, and drinking water electroconductivity), deployed by our group across 250,000 km2 in Michigan, USA. The proposed methodology achieved accuracy scores of up to 0.97 and consistently outperformed state-of-the-art anomaly detection techniques.

An Automated Toolchain for Camera-Enabled Sensing of Drinking Water Chlorine Residual.

Chlorine residual concentration is an important parameter to prevent pathogen growth in drinking water. Disposable color changing test strips that measure chlorine in tap water are commercially available to the public; however, the color changes are difficult to read by eye, and the data are not captured for water service providers. Here we present an automated toolchain designed to process digital images of free chlorine residual test strips taken with mobile phone cameras. The toolchain crops the image using image processing algorithms that isolate the areas relevant for analysis and automatically white balances the image to allow for use with different phones and lighting conditions. The average red, green, and blue (RGB) color values of the image are used to predict a free chlorine concentration that is classified into three concentration tiers (<0.2 mg/L, 0.2-0.5 mg/L, or >0.5 mg/L), which can be reported to water users and recorded for utility use. The proposed approach was applied to three different phone types under three different lighting conditions using a standard background. This approach can discriminate between concentrations above and below 0.5 mg/L with an accuracy of 90% and 94% for training and testing data sets, respectively. Furthermore, it can discriminate between concentrations of <0.2 mg/L, 0.2-0.5 mg/L, or >0.5 mg/L with weighted-averaged F1 scores of 79% and 88% for training and testing data sets, respectively. This tool sets the stage for tap water consumers and water utilities to gather frequent measurements and high-resolution temporal and spatial data on drinking water quality.

Stationary and portable multipollutant monitors for high-spatiotemporal-resolution air quality studies including online calibration.

The distribution and dynamics of atmospheric pollutants are spatiotemporally heterogeneous due to variability in emissions, transport, chemistry, and deposition. To understand these processes at high spatiotemporal resolution and their implications for air quality and personal exposure, we present custom, low-cost air quality monitors that measure concentrations of contaminants relevant to human health and climate, including gases (e.g., O3, NO, NO2, CO, CO2, CH4, and SO2) and size-resolved (0.3-10 µm) particulate matter. The devices transmit sensor data and location via cellular communications and are capable of providing concentration data down to second-level temporal resolution. We produce two models: one designed for stationary (or mobile platform) operation and a wearable, portable model for directly measuring personal exposure in the breathing zone. To address persistent problems with sensor drift and environmental sensitivities (e.g., relative humidity and temperature), we present the first online calibration system designed specifically for low-cost air quality sensors to calibrate zero and span concentrations at hourly to weekly intervals. Monitors are tested and validated in a number of environments across multiple outdoor and indoor sites in New Haven, CT; Baltimore, MD; and New York City. The evaluated pollutants (O3, NO2, NO, CO, CO2, and PM2.5) performed well against reference instrumentation (e.g., r = 0.66-0.98) in urban field evaluations with fast e-folding response times (≤1 min), making them suitable for both large-scale network deployments and smaller-scale targeted experiments at a wide range of temporal resolutions. We also provide a discussion of best practices on monitor design, construction, systematic testing, and deployment.

Extracting useful signals from flawed sensor data: Developing hybrid data-driven approaches with physical factors.

Increased availability and affordability of sensors, especially water quality sensors, is poised to improve process control and modelling in water and wastewater systems. Sensor measurements are often flawed by unavoidable influent complexity and sensor instability, making extraction of useful signals problematic. Although a natural reaction is to put extra effort into sensor maintenance to achieve more reliable measurements, useful signals can be extracted from those unqualified signals by appropriate usage of available data-driven tools instructed by physical factors (e.g. prior process knowledge, physical constraints, phenomenal observations). Such methodology is herein defined as hybrid approaches. While the concept of coupling physical factors into data-driven tools is not new in downstream applications such as process modelling and control, little literature has explicitly applied it in the first and equally important step - signal processing. With flawed influent five-day biochemical oxygen demand (BOD5) sensor measurements as an example, this paper provides a comprehensive case study demonstrating how physical factors were incorporated throughout the procedures of processing a flawed signal for its maximum value. Results showed that useful signals were extracted and validated via an assembly of well-established machine learning tools, whose performance was improved with physical factors. An Improved Standard Signal Processing Architecture (ISSPA) is also proposed based on the results of this research.

Windshield wipers on connected vehicles produce high-accuracy rainfall maps.

Connected vehicles are poised to transform the field of environmental sensing by enabling acquisition of scientific data at unprecedented scales. Drawing on a real-world dataset collected from almost 70 connected vehicles, this study generates improved rainfall estimates by combining weather radar with windshield wiper observations. Existing methods for measuring precipitation are subject to spatial and temporal uncertainties that compromise high-precision applications like flash flood forecasting. Windshield wiper measurements from connected vehicles correct these uncertainties by providing precise information about the timing and location of rainfall. Using co-located vehicle dashboard camera footage, we find that wiper measurements are a stronger predictor of binary rainfall state than traditional stationary gages or radar-based measurements. We introduce a Bayesian filtering framework that generates improved rainfall estimates by updating radar rainfall fields with windshield wiper observations. We find that the resulting rainfall field estimate captures rainfall events that would otherwise be missed by conventional measurements. We discuss how these enhanced rainfall maps can be used to improve flood warnings and facilitate real-time operation of stormwater infrastructure.

Field and Laboratory Evaluations of the Low-Cost Plantower Particulate Matter Sensor.

Due to the rapid development of low-cost air-quality sensors, a rigorous scientific evaluation has not been conducted for many available sensors. We evaluated three Plantower PMS A003 sensors when exposed to eight particulate matter (PM) sources (i.e., incense, oleic acid, NaCl, talcum powder, cooking emissions, and monodispersed polystyrene latex spheres under controlled laboratory conditions and also residential air and ambient outdoor air in Baltimore, MD). The PM2.5 sensors exhibited a high degree of precision and R2 values greater than 0.86 for all sources, but the accuracy ranged from 13 to >90% compared with reference instruments. The sensors were most accurate for PM with diameters below 1 µm, and they poorly measured PM in the 2.5-5 µm range. The accuracy of the sensors was dependent on relative humidity (RH), with decreases in accuracy at RH > 50%. The sensors were able to produce meaningful data at low and high temperatures and when in motion, as it would be if utilized for outdoor or personal monitoring applications. It was most accurate in environments with polydispersed particle sources and may not be useful in specialized environments or experiments with narrow distributions of PM or aerosols with a large proportion of coarse PM.

Are all data useful? Inferring causality to predict flows across sewer and drainage systems using directed information and boosted regression trees.

As more sensor data become available across urban water systems, it is often unclear which of these new measurements are actually useful and how they can be efficiently ingested to improve predictions. We present a data-driven approach for modeling and predicting flows across combined sewer and drainage systems, which fuses sensor measurements with output of a large numerical simulation model. Rather than adjusting the structure and parameters of the numerical model, as is commonly done when new data become available, our approach instead learns causal relationships between the numerically-modeled outputs, distributed rainfall measurements, and measured flows. By treating an existing numerical model - even one that may be outdated - as just another data stream, we illustrate how to automatically select and combine features that best explain flows for any given location. This allows for new sensor measurements to be rapidly fused with existing knowledge of the system without requiring recalibration of the underlying physics. Our approach, based on Directed Information (DI) and Boosted Regression Trees (BRT), is evaluated by fusing measurements across nearly 30 rain gages, 15 flow locations, and the outputs of a numerical sewer model in the city of Detroit, Michigan: one of the largest combined sewer systems in the world. The results illustrate that the Boosted Regression Trees provide skillful predictions of flow, especially when compared to an existing numerical model. The innovation of this paper is the use of the Directed Information step, which selects only those inputs that are causal with measurements at locations of interest. Better predictions are achieved when the Directed Information step is used because it reduces overfitting during the training phase of the predictive algorithm. In the age of "big water data", this finding highlights the importance of screening all available data sources before using them as inputs to data-driven models, since more may not always be better. We discuss the generalizability of the case study and the requirements of transferring the approach to other systems.

Shaping Streamflow Using a Real-Time Stormwater Control Network.

"Smart" water systems are transforming the field of stormwater management by enabling real-time monitoring and control of previously static infrastructure. While the localized benefits of active control are well-established, the potential for system-scale control of watersheds is poorly understood. This study shows how a real-world smart stormwater system can be leveraged to shape streamflow within an urban watershed. Specifically, we coordinate releases from two internet-controlled stormwater basins to achieve desired control objectives downstream-such as maintaining the flow at a set-point, and generating interleaved waves. In the first part of the study, we describe the construction of the control network using a low-cost, open-source hardware stack and a cloud-based controller scheduling application. Next, we characterize the system's control capabilities by determining the travel times, decay times, and magnitudes of various waves released from the upstream retention basins. With this characterization in hand, we use the system to generate two desired responses at a critical downstream junction. First, we generate a set-point hydrograph, in which flow is maintained at an approximately constant rate. Next, we generate a series of overlapping and interleaved waves using timed releases from both retention basins. We discuss how these control strategies can be used to stabilize flows, thereby mitigating streambed erosion and reducing contaminant loads into downstream waterbodies.

An automated toolchain for the data-driven and dynamical modeling of combined sewer systems.

The recent availability and affordability of sensors and wireless communications is poised to transform our understanding and management of water systems. This will enable a new generation of adaptive water models that can ingest large quantities of sensor feeds and provide the best possible estimates of current and future conditions. To that end, this paper presents a novel data-driven identification/learning toolchain for combined sewer and stormwater systems. The toolchain uses Gaussian Processes to model dry-weather flows (domestic wastewater) and dynamical System Identification to represent wet-weather flows (rainfall runoff). By using a large and high-resolution sensor dataset across a real-world combined sewer system, we illustrate that relatively simple models can achieve good forecasting performance, subject to a finely-tuned and continuous re-calibration procedure. The data requirements of the proposed toolchain are evaluated, showing sensitivity to spatial heterogeneity and unique time-scales across which models of individual sites remain representative. We identify a near-optimal time record, or data "age," for which historical measurements must be available to ensure good forecasting performance. We also show that more data do not always lead to a better model due to system uncertainty, such as shifts in climate or seasonal wastewater patterns. Furthermore, the individual components of the model (wet- and dry-weather) often require different volumes of historical observations for optimal forecasting performance, thus highlighting the need for a flexible re-calibration toolchain rather than a one-size-fits-all approach.

Smarter Stormwater Systems.

Existing stormwater systems require significant investments to meet challenges imposed by climate change, rapid urbanization, and evolving regulations. There is an unprecedented opportunity to improve urban water quality by equipping stormwater systems with low-cost sensors and controllers. This will transform their operation from static to adaptive, permitting them to be instantly "redesigned" to respond to individual storms and evolving land uses.

TDMA-based dual-mode communication for mobile wireless sensor networks.

Small highly mobile robots, and in particular micro air vehicles (MAVs), are well suited to the task of exploring unknown indoor environments such as buildings and caves. Such a task imposes a number of requirements on the underlying communication infrastructure, with differing goals during various stages of the mission. This work addresses those requirements with a hybrid communications infrastructure consisting of a stationary mesh network along with the mobile nodes. The combined network operates in two independent modes, coupling a highly efficient, low duty cycle, low throughput mode for routing and persistent sensing with a burst mode for high data rate communication. By strategically distributing available frequency channels between the mobile agents and the stationary nodes, the overall network provides reliable long-term communication paths while maximizing data throughput when needed.