Collaborating Authors


Hybrid Attention Networks for Flow and Pressure Forecasting in Water Distribution Systems Machine Learning

Multivariate geo-sensory time series prediction is challenging because of the complex spatial and temporal correlation. In urban water distribution systems (WDS), numerous spatial-correlated sensors have been deployed to continuously collect hydraulic data. Forecasts of monitored flow and pressure time series are of vital importance for operational decision making, alerts and anomaly detection. To address this issue, we proposed a hybrid dual-stage spatial-temporal attention-based recurrent neural networks (hDS-RNN). Our model consists of two stages: a spatial attention-based encoder and a temporal attention-based decoder. Specifically, a hybrid spatial attention mechanism that employs inputs along temporal and spatial axes is proposed. Experiments on a real-world dataset are conducted and demonstrate that our model outperformed 9 baseline models in flow and pressure series prediction in WDS.

myDevices and éolane Join Forces to Help Wastewater Treatment Facilities Around the World Prevent Sewage Backups Caused by the COVID-19 Crisis


LOS ANGELES--(BUSINESS WIRE)--Sewage systems are impacted as consumers flush disinfectant wipes, paper towels, and napkins, an unintended consequence of the COVID-19 pandemic. Wipes get caught on misaligned pipe joints choking sewer lines and wrapping around pump motors, causing clogs, excessive strain, and infrastructure damage. These clogs can result in overflows of raw sewage into local rivers and lakes, and creating backups into people's homes. Significant blockages often require municipal staff to clear them, at a time when efforts and tax dollars need to be focused on critical services. "Wastewater treatment facilities around the state already are reporting issues with their sewer management collection systems," the California State Water Board said in a statement.

Why next year could be a turning point for project management and AI


Artificial Intelligence hasn't quite arrived in the project management sphere yet, but it's on its way. Gartner forecasts that 80 per cent of project management roles will be eliminated by 2030 as AI takes on traditional project management functions such as data collection, tracking and reporting. The same report highlights that programme and portfolio management (PPM) software is behind the times, and AI-enabled PPM is only just beginning to surface in the market. However, while some tasks will inevitably be automated, it opens up other opportunities for project managers. It's important to know the difference between how AI-enabled automation can change project management and how AI-enabled insights from massive databases can make a difference.

Microrobots made from pollen help remove toxic mercury from wastewater

New Scientist

Tiny robots made using pollen could one day be used to clean contaminated water. Waste water from some factories contains mercury, a metal that can cause illness if consumed. There are techniques to remove mercury in water treatment plants, but they are time consuming and expensive. Martin Pumera at the University of Chemistry and Technology, Prague, in the Czech Republic, and his colleagues are working on a low-cost alternative.

Artificial intelligence sustains critical infrastructure during COVID-19


The adoption of artificial intelligence and machine learning technologies has never been more critical. Due to COVID-19, many organizations need to find a new way of working. Ensuring production rates are reliable, if not increased, while limiting the number of personnel - in some cases down to 50%. Many asset heavy industries, such as water, transportation & energy are considered critical infrastructure. Every effort needs to be made to maintain these.

Can't Boil This Frog: Robustness of Online-Trained Autoencoder-Based Anomaly Detectors to Adversarial Poisoning Attacks Machine Learning

In recent years, a variety of effective neural network-based methods for anomaly and cyber attack detection in industrial control systems (ICSs) have been demonstrated in the literature. Given their successful implementation and widespread use, there is a need to study adversarial attacks on such detection methods to better protect the systems that depend upon them. The extensive research performed on adversarial attacks on image and malware classification has little relevance to the physical system state prediction domain, which most of the ICS attack detection systems belong to. Moreover, such detection systems are typically retrained using new data collected from the monitored system, thus the threat of adversarial data poisoning is significant, however this threat has not yet been addressed by the research community. In this paper, we present the first study focused on poisoning attacks on online-trained autoencoder-based attack detectors. We propose two algorithms for generating poison samples, an interpolation-based algorithm and a back-gradient optimization-based algorithm, which we evaluate on both synthetic and real-world ICS data. We demonstrate that the proposed algorithms can generate poison samples that cause the target attack to go undetected by the autoencoder detector, however the ability to poison the detector is limited to a small set of attack types and magnitudes. When the poison-generating algorithms are applied to the popular SWaT dataset, we show that the autoencoder detector trained on the physical system state data is resilient to poisoning in the face of all ten of the relevant attacks in the dataset. This finding suggests that neural network-based attack detectors used in the cyber-physical domain are more robust to poisoning than in other problem domains, such as malware detection and image processing.

Big data and Artificial Intelligence to Control Algal Blooms


Toxic algal blooms are a problem that is globally increasing due to nutrients pollution and climate change. Although the use of chemicals may provide temporary relief to the problem, it does not offer a solution. Now an alternative method for chemical algae control is available. Based on the acquisition of big data, artificial intelligence and ultrasound, this novel method can control algal blooms in large water surfaces without disrupting the ecosystem. Toxic blooms of algae are increasing globally in our waterways, causing a variety of health-related issues and environmental degradation.

Application of artificial intelligence to wastewater treatment: A bibliometric analysis and systematic review of technology, economy, management, and wastewater reuse


Bibliometric analysis and systematic review of AI applied to wastewater treatment. Wastewater treatment technology, economy, management, and reuse were discussed. Prediction accuracy of AI technologies on pollutant removal ranged 0.64–1.00. Application of AI technology could reduce operational costs by up to 30 %. Combined AI methods could provide higher accuracy and lower error. Wastewater treatment is an important step for pollutant reduction and the promotion of water environment quality.

How can Big Data enable smart collection systems and protect Wastewater Treatment Plants


Digitization is essential for delivering these centralized water collection services and supporting efficient urbanization. It allows networks to benefit from online connectivity and management platforms that feed on information (Big Data) and can handle data far more effectively than can human operators. Commonly known as Artificial Intelligence (AI) systems, these revolutionary processes have advanced the way wastewater networks can be managed, helping to protect Wastewater Treatment Plants from damage, maximizing process efficiencies and enabling expanded water reuse projects. Digitizing a city's wastewater networks through AI starts with good data. To understand how such a system is able to provide operational wastewater intelligence, it is important to understand Big Data, AI, Machine Learning and their applications.

Intrusion Detection for Industrial Control Systems: Evaluation Analysis and Adversarial Attacks Machine Learning

--Neural networks are increasingly used in security applications for intrusion detection on industrial control systems. In this work we examine two areas that must be considered for their effective use. Firstly, is their vulnerability to adversarial attacks when used in a time series setting. Secondly, is potential overestimation of performance arising from data leakage artefacts. T o investigate these areas we implement a long short-term memory (LSTM) based intrusion detection system (IDS) which effectively detects cyber-physical attacks on a water treatment testbed representing a strong baseline IDS. The first attacker is able to manipulate sensor readings on a subset of the Secure Water Treatment (SWaT) system. By creating a stream of adversarial data the attacker is able to hide the cyber-physical attacks from the IDS. For the cyber-physical attacks which are detected by the IDS, the attacker required on average 2.48 out of 12 total sensors to be compromised for the cyber-physical attacks to be hidden from the IDS. The second attacker model we explore is an L bounded attacker who can send fake readings to the IDS, but to remain imperceptible, limits their perturbations to the smallest L value needed. Additionally, we examine data leakage problems arising from tuning for F 1 score on the whole SWaT attack set and propose a method to tune detection parameters that does not utilise any attack data. If attack aftereffects are accounted for then our new parameter tuning method achieved an F 1 score of 0.811 0.0103. I NTRODUCTION Deep learning systems are known to be vulnerable to adversarial attacks at test time. By applying small changes to an input an attacker can cause a machine learning system to mis-classify with a high degree of success. There has been much work on both developing more powerful attacks [1] as well as defences [2]. However, the majority of adversarial machine learning research is focused on the image domain, with consideration of the different challenges that arise within other fields needed [3]. This phenomenon of adversarial examples becomes particularly pertinent when aiming to defend machine learn-Pre-print.