Ondrej Holub

Knowledge-Based Fault Propagation in Building Automation Systems

This paper describes a knowledge-based approach that can reason about effects of faults and causes of abnormal situations in building automation systems (BAS). Combining an ontology-based building information model (BIM), which models a BAS formally and semantically, with rules encoding expert knowledge, the fault propagation approach can automatically determine causalities in BAS and propagate faults along the causalities in both forward and backward direction. This enables an immediate assessment of potential consequences of faults respectively an analysis of the root cause(s). The fault propagation approach can enhance fault detection and diagnosis by considering BAS as a whole, being aware of the potentially far reaching consequences of faults, instead of just focusing on single pieces of equipment or zones. This provides a better understanding of BAS and improves the decision making and prioritization of the right emergency and maintenance actions.

HVAC simulation model for advanced diagnostics

This article presents an HVAC simulator designed for validation of advanced methods for diagnostics and predictive maintenance of building automation systems. A formal framework for the diagnostics and maintenance methods is briefly introduced. The design and implementation of the simulator are described in more detail and concluded with a typical example.

Platform for coordination of energy generation and consumption in residential neighborhoods

This article presents a platform, which aims at energy savings in residential neighborhoods via coordination of energy generation and consumption. The platform includes hardware infrastructure and service oriented information system. While an overview of the platform is given, the energy management services are described in more detail. Assessment of platform robustness, flexibility and scalability is provided and demonstrated on examples.

End user models for residential demand response

This paper presents a platform for Automated Demand Response (ADR) services in residential neighborhoods. While the platform is briefly introduced, main focus is on user behavior. A model of residential end users behavior is defined. Next, a procedure to extract model parameters from available data is presented. Value of the behavior model is demonstrated on two use cases.

Automatic setup of fault detection algorithms in building and home automation

The complexity and diversity of building automation systems (BAS) and the various faults that may happen in buildings make it very challenging to set up fault detection and diagnostics (FDD). This knowledge-intensive, expensive tasks urgently needs to be automated due to the limited time and budget available in this domain. A novel approach for computer-aided, automatic setup of FDD for BAS is described in this paper. Using the Web Ontology Language (OWL), the characteristics, requirements and necessary configuration of the FDD algorithms are formally specified in a machine-interpretable way. Also buildings and their BAS are described by OWL as building information model (BIM). Given these descriptions, the approach can match the specific characteristics and requirements of the FDD algorithms with the specific circumstances in the BAS in a formal evaluation process. All applicable FDD algorithms can be determined and subsequently configured to the individual characteristics of the BAS.

Ontology-based automatic setup of virtual sensors in building automation systems

Effective monitoring, fault detection and diagnostics (FDD) of building automation systems require a large number of datapoints. But the set of available sensors in a building is typically limited. Virtual sensors (VS) can be applied as cost saving alternative to real sensors. Implementing and setting up VS however can be a complex tasks that urgently needs to be automated. A novel approach for computer-aided, automatic setup of VS is presented. It is based on reusing generic, pre-implemented VS algorithms. Using OWL, the characteristics, requirements and configuration of the VS algorithms are specified in a machine-interpretable way. By that, the approach can match the specific characteristics and requirements of the VS algorithms in a formal evaluation process with the specific circumstances in the BAS, which are described in an ontology-based building information model (BIM). All applicable VS algorithms can be determined and subsequently configured to the individual characteristics of the BAS.

Efficient HVAC controls: A symbolic approach

Heating, ventilation, and air conditioning (HVAC) systems contribute significantly to energy consumption in buildings, and are the target of an ongoing push towards energy efficiency, which reflects on new requirements for the configuration and operation of HVAC units. On-off actuators are being replaced by staged or modulated ones, and additional components are being added to the HVAC units, which have to be properly managed by new, advanced control architectures. As such, it is increasingly more difficult for traditional control approaches to cope with new devices and meet new objectives. This leads to an opportunity for alternative, novel approaches to be adopted by the industry: promising are recently developed formal methods for controller synthesis. In this article, we employ formal, symbolic techniques to synthesize controllers for a model of a roof-top unit conditioning a low-rise commercial building. The synthesized controller is validated on a simulation with practically relevant operating conditions.

Adaptive quantile estimation in performance monitoring of building automation systems

Heating, ventilation and air conditioning (HVAC) systems in buildings are subject to increasing expectations on performance. As such, HVAC systems need to be continuously monitored in order to early detect any anomalies which would degrade their performance. The building automation domain is subject to constraints which translate into the need for computationally efficient self-learning techniques. This paper provides an industrial R&D perspective on the topic with a more detailed focus on a modified algorithm for adaptive quantile estimation, which can be applied in various analytical workflows. This is shown on the examples of control performance monitoring and equipment fault detection. Real world results for control performance monitoring are presented and analyzed in more detail.