This paper investigates the use of hybrid machine learning (HML) for the detection of anomalous multivariate time-series (MVTS). Focusing on a specific industrial use case from geotechnical engineering, where 100's of MVTS need to be analyzed and classified, has permitted extensive testing of the proposed methods with real measurement data. The novel hybrid anomaly detector combines two means for detection, creating redundancy and reducing the risk of missing defective elements in a safety relevant application. The two parts are: (1) anomaly detection based on approximately fifty physics-motivated key performance indicators (KPI) and ( 2) an unsupervised variational autoencoder with long short-term memory layers. The KPI capture expert knowledge on properties of the data that infer the quality of produced elements; these are used as a type of autolabeling. The goal of the extension using machine learning (ML) is to detect anomalies that the experts may not have foreseen. In contrast to anomaly detection in streaming data, where the goal is to locate an anomaly, each MVTS is complete in itself at the time of evaluation and is categorized as anomalous or non-anomalous.The paper compares the performance of different variationalautoencoder architectures (e.g., LSTM-VAE and BiLSTM-VAE). The results of using a genetic algorithm to optimize the hyperparameters of the different architectures are also presented. It is shown that modeling the industrial process as an assemblage of sub-processes yields a better discriminating power and permits the identification of inter-dependencies between the sub-processes. Interestingly, different autoencoder architectures may be optimal for different sub-processes; here two different architectures are combined to achieve superior performance. Extensive results are presented based on a very large set of real-time measurement data.

This article presents a new approach of quality control to vibro ground improvement techniques based on hybrid machine learning (ML), i.e., a combination of classical analysis and ML techniques. The process is monitored with an instrumented rig equipped with multiple sensors. Key performance indicators (KPIs) are used to identify anomalous foundation columns. As the foundation columns are sub‐surface, there is no direct access to ground truth; consequently, unsupervised ML is applied to the recorded time‐series data. The risk of not detecting defective elements is reduced by the combination of two independent methods for anomaly detection, KPI‐ and ML‐based classification. The ML is used to gain a deeper process understanding and to detect anomalies which were not considered in the design phase of the KPI. New pre‐processing techniques were derived from the insights gained from the ML classifier; this led to a more robust classifier. It is shown how unsupervised ML, using a multi‐channel variational autoencoder (VAE) with long short‐term memory (LSTM) layers, can be utilized in a knowledge discovery process (KDP).