This paper presents a comprehensive survey of state-of-the-art intelligent fault detection and diagnosis in district heating systems. Maintaining an efficient system is crucial, as faults can lead to increased heat loss, customer discomfort, operational cost. Intelligent help identify diagnose faulty behavior automatically by utilizing artificial intelligence or machine learning. In our survey, we review discuss 57 papers published the last 12 years, highlight recent trends, current research...

This study introduces a novel systematic approach to address the challenge of labeled data scarcity for fault detection and diagnosis (FDD) in District Heating (DH) systems. To replicate real-world DH scenarios, we have created controlled laboratory emulation generic substation integrated with climate chamber. Furthermore, present an FDD pipeline using isolation forest one-class support vector machine alongside random diagnosis. Our research analyzed impact sampling frequencies on models,...

Gene expression microarray experiments frequently generate datasets with multiple values missing. However, most of the analysis, mining, and classification methods for gene data require a complete matrix array values. Therefore, accurate estimation missing in such has been recognized as an important issue, several imputation algorithms have already proposed to biological community. Most these approaches, however, are not particularly suitable time series profiles. In view this, we propose...

Presently, with the increasing number and complexity of available gene expression datasets, combination data from multiple microarray studies addressing a similar biological question is gaining importance. The analysis integration datasets are expected to yield more reliable robust results since they based on larger samples effects individual study-specific biases diminished. This supported by recent suggesting that important signals often preserved or enhanced experiments. An approach...

Training of machine learning models in a Datacen-ter, with data originated from edge nodes, incurs high communication overheads and violates user's privacy. These challenges may be tackled by employing Federated Learning (FL) technique to train model across multiple decentralized devices (workers) using local data. In this paper, we explore an approach that identifies the most representative updates made workers those are only uploaded central server for reducing network costs. Based on...

Federated Learning (FL) provides a promising solution for preserving privacy in learning shared models on distributed devices without sharing local data central server. However, most existing work shows that FL incurs high communication costs. To address this challenge, we propose clustering-based federated solution, entitled via Clustering Optimization (FedCO), which optimizes model aggregation and reduces In order to reduce the costs, first divide participating workers into groups based...

Abstract Data has become an integral part of our society in the past years, arriving faster and larger quantities than before. Traditional clustering algorithms rely on availability entire datasets to model them correctly efficiently. Such requirements are not possible data stream scenario, where arrives needs be analyzed continuously. This paper proposes a novel evolutionary algorithm, entitled EvolveCluster, capable modeling evolving streams. We compare EvolveCluster against two other...

Abstract Many machine learning models deployed on smart or edge devices experience a phase where there is drop in their performance due to the arrival of data from new domains. This paper proposes novel unsupervised domain adaptation algorithm called DIBCA++ deal with such situations. The uses only clusters’ mean, standard deviation, and size, which makes proposed modest terms required storage computation. study also presents explainability aspect algorithm. compared its predecessor, DIBCA,...

A novel integration approach targeting the combination of multi-experiment time series expression data is proposed. recursive hybrid aggregation algorithm initially employed to extract a set genes, which are eventually interest for biological phenomenon under study. Next, hierarchical merge procedure specifically developed purpose fusing together multiple-experiment pro.les selected genes. This employs dynamic warping alignment techniques in order account adequately potential phase shift...

Machine learning software accounts for a significant amount of energy consumed in data centers. These algorithms are usually optimized towards predictive performance, i.e. accuracy, and scalability. This is the case stream mining algorithms. Although these adaptive to incoming data, they have fixed parameters from beginning execution. We observed that having lead unnecessary computations, thus making algorithm inefficient. In this paper we present nmin adaptation method Hoeffding trees....

Abstract Recently machine learning researchers are designing algorithms that can run in embedded and mobile devices, which introduces additional constraints compared to traditional algorithm design approaches. One of these is energy consumption, directly translates battery capacity for devices. Streaming algorithms, such as the Very Fast Decision Tree (VFDT), designed devices due their high velocity low memory requirements. However, they have not been with an efficiency focus. This paper...

We propose a higher order mining (HOM) approach for modelling, monitoring and analyzing district heating (DH) substations' operational behaviour performance. HOM is concerned with over patterns rather than primary or raw data. The proposed uses combination of different data analysis techniques such as sequential pattern mining, clustering analysis, consensus minimum spanning tree (MST). Initially, substation's modeled by extracting weekly performing analysis. performance monitored assessing...