Several faults affect heating, ventilation, and air conditioning (HVAC) chiller systems, leading to energy wastage, discomfort for the users, shorter equipment life, and system unreliability. Early detection of anomalies can prevent further deterioration of the chiller and energy wastage. In this work, a data-driven approach is used in order to detect faults that usually plague chiller systems. In particular, the proposed approach employs a kernel principal component analysis (KPCA) in order to capture the normal operative conditions of the system; the learning method turns out to be effective in handling nonlinear phenomena through the use of the Gaussian kernel, which, by means of a self-tuning procedure, ensures good accuracy properties while maintaining enough generalization characteristics. The effectiveness of the proposed fault detection method is evaluated by means of tests on emulated and real chiller data. The KPCA approach is first proved to exhibit better detection performances than linear PCA, and then, it is corroborated through the comparison with local outlier factor, one-class support vector machine, and isolation forest.

Efficient management of energy resources is crucial in smart buildings. In this work, model predictive control (MPC) is used to minimize the economic costs of prosumers equipped with production units, energy storage systems, and electric vehicles. To this purpose, the predictive control manages the available energy resources by exploiting future information about energy prices, absorption and production power profiles, and electric vehicle (EV) usage, such as times of departure and arrival and predicted energy consumption. The predictive control is compared with a rule-based technique, herein referred to as a heuristic approach, that acts in an instant-by-instant fashion without considering any future information. The reported results show that the studied predictive approach allows one to achieve charging profiles that adapt to variable operating conditions, aiming at optimal performances in terms of economic cost minimization in time-varying price scenarios, reduction of rms current stresses, and recharging capability of EV batteries. Specifically, unlike the heuristic method, the MPC approach is proven to be capable of efficiently managing the available energy resources to ensure a full recharge of the EV battery during nighttime while always respecting all system constraints. In addition, the proposed control is shown to be capable of keeping the peak power absorption from the grid constrained within set limits, which is a valuable feature in scenarios with widespread adoption of EVs in order to limit the stress on the electrical system.

The diffusion of distributed energy resources in distribution networks requires new approaches to exploit the users' capabilities of providing ancillary services. Of particular interest will be the coordination of microgrids operating as an aggregate of demand and supply units. This work reports a model predictive control (MPC) application in microgrids for the efficient energy management of energy storage systems and photovoltaic units. The MPC minimizes the economic cost of aggregate prosumers into a prediction horizon by forecasting generation and absorption profiles. The MPC is compared in realistic conditions with a heuristic strategy that acts in a instant manner, without taking into account signals prediction. The work aims at investigating the effect that different types of energy tariffs have in enhancing the end-users' flexibility, based on three examples of currently applied tariffs, comparing the two storage control modes. The MPC always achieves a better solution than the heuristic approach in all considered scenarios from the cost minimization point of view, with an improvement that is amplified by increasing the energy price variations between peak and off-peak periods. Furthermore, the MPC approach provides a cost saving when compared to the case considering a microgrid endowed with only photovoltaic units, in which no storage is installed. Findings in this work confirm that storage units better perform when some knowledge of future demand and supply trends is provided, ensuring an economic cost saving and an important service for the overall community.

Effective and efficient energy management is crucial in smart electrical systems like nanogrids. The application of model predictive control (MPC) for the energy management of electrical energy storage in nanogrids for smart buildings is considered in this paper. An MPC strategy is devised in order to improve the nanogrid operation at the points of connection with the utility grid, integrating in the control information about future power generation and load consumption. The strategy is compared with a heuristic approach that exploits the energy storage only on an instantaneous basis, without any future predictions. It is shown that the MPC approach can provide convenient references to the power converters interfacing energy resources with the grid and significant long-term advantages as compared to the heuristic strategy.

This work presents the application of model predictive control (MPC) for the energy management of smart buildings in microgrids. It is shown that by means of the described MPC formulation the power exchange at the point of connection of the building can be made close to a given power reference, typically available in microgrid contexts, and, therefore, more predictable. The MPC has the advantage of using in a smart way the storage, with a limited stress to the electronic power interface of the storage system. In principle, the approach is capable of exploiting all the available information about generation and absorption profiles, while minimizing operation costs. In addition, an economic meaning of the terms of the cost function is given, which allows to better appreciate the economic impact of the set control parameters. The approach is evaluated considering realistic conditions and parameters and compared with a heuristic method that operates in an instantaneous basis, that is, without considering any information on future system evolution.

Water chillers play a crucial role in HVAC systems. Faulty operations of chillers can lead to energy wastage, system unreliability and shorter equipment life. Due to the intrinsic complexity of these systems, which are nonlinear with interrelated parameters, and since data regarding unforeseen phenomena and abnormalities are not usually available for air-conditioning installations, the development of fault detection algorithms is a burdensome task. In this paper, a data-driven approach is used in order to develop a fault detection methodology that makes no use of a priori knowledge about abnormal phenomena. In particular, we exploit a local Principal Component Analysis to handle nonlinear cases and to accent novelties with respect to non-faulty operations variability. The performance of the proposed approach is assessed by using a synthetic dataset which is related to an ai-rcondensed water chiller in both fault-free and faulty conditions.

Faulty operations of Heating, Ventilation and Air Conditioning (HVAC) chiller systems can lead to discomfort for the users, energy wastage, system unreliability and shorter equipment life. Faults need to be early diagnosed to prevent further deterioration of the system behaviour and energy losses. In this paper a model-based approach is used in order to detect important chiller systems faults. First, a linear dynamic black-box model is identified for each of the relevant characteristic features of the system during the normal functioning of the chiller. Then, an on-line correlogram method verifies the whiteness property of the residuals in order to distinguish anomalies from normal operations. A decision table, that matches the influence of anomalies with the characteristic features, allows to identify chiller faults. The proposed fault detection and diagnosis approach is assessed by using real chiller data provided by the ASHRAE research project RP-1043.

Faulty operations of Heating, Ventilation and Air Conditioning (HVAC) chiller systems can lead to discomfort for the users, energy wastage, system unreliability and shorter equipment life. Faults need to be early diagnosed to prevent further deterioration of the system behaviour and energy losses. Since it is not a common practice to collect historical data regarding unforeseen phenomena and abnormal behaviours for HVAC installations, in this paper, a semi-supervised data-driven approach is employed for fault detection and isolation that makes no use of a priori knowledge about abnormal phenomena. The proposed method exploits Principal Component Analysis (PCA) to distinguish anomalies from normal operation variability and a reconstruction-based contribution approach to isolate variables related to faults. The diagnosis task is then tackled by means of a decision table that associates the influence of faults to certain characteristic features. The Fault Detection and Diagnosis (FDD) algorithm performance is assessed by exploiting experimental datasets from two types of water chiller systems.

In this paper we present a model-based approach for designing efficient control strategies with the aim of increasing the performance of Heating, Ventilation and Air-Conditioning (HVAC) systems with ice Cold Thermal Energy Storage (ice CTES). The use of TES systems ensures reduced energy costs and energy consumption, increased flexibility of operation, reduced equipment size and pollutant emissions. A simulation environment based on Matlab/Simulink® is developed, where the thermal behaviour of the plant is analysed by a lumped formulation of the conservation equations. In particular, the ice CTES is modelled as a hybrid system, where the water phase transitions (solid-melting-liquid, liquid-freezing-solid) are described by combining continuous and discrete dynamics, thus considering both latent and sensible heat. Three standard control strategies and a model predictive control approach are developed and compared. Extensive simulations confirm that the MPC provides the best control in terms of energy efficiency and cooling load demand satisfaction with respect to standard control strategies.

Faulty operations of Heating, Ventilation and Air Conditioning (HVAC) chiller systems can lead to discomfort for the occupants, energy wastage, unreliability and shorter equipment life. Such faults need to be detected early to prevent further escalation and energy losses. Commonly, data regarding unforeseen phenomena and abnormalities are rare or are not available at the moment for HVAC installations: for this reason in this paper an unsupervised One-Class SVM classifier employed as a novelty detection system to identify unknown status and possible faults is presented. The approach, that exploits Principal Component Analysis to accent novelties w.r.t. normal operations variability, has been tested on a HVAC literature dataset.

Faulty operations of Heating, Ventilation and Air Conditioning (HVAC) systems can lead to discomfort for the occupants, energy wastage, unreliability and shorter equipment life. Cost-effective Fault Detection and Diagnosis (FDD) methods can therefore ensure an increase in the system uptime, reliability, and overall efficiency. In this paper, a simulation environment based on Matlab/Simulink® is used in order to evaluate the performance of a FDD method using Support Vector Machines (SVMs). In detail, the proposed method is evaluated by performing extensive simulations to allow the investigation of the most common and relevant faults affecting this kind of systems.

In this paper, Artificial Neural Networks (ANNs) are used to achieve cooling load forecasting in HVAC (Heating, Ventilating, and Air Conditioning) systems. Load forecasting is crucial in plant configurations making use of thermal storage technologies, where, during the nighttime, part or most of the energy required during daytime is produced at lower cost by cooling or icing water. Load forecasting is then needed to quantify the energy to be stored for the following daytime and to set up strategies for its release during daytime. Although many algorithms have been presented in the literature for load forecasting, they often need as input a large data set, that is not always available in practical situations. In this paper, we present an algorithm based on ANNs that allows to obtain sufficiently accurate load predictions by exploiting a limited data set, obtained by measuring quantities that are typically available in standard HVAC installations. Furthermore, knowledge of the current thermal load (which is needed to setup the data set for ANN training) can be obtained by using a load estimation algorithm previously proposed by some of the authors, that only need basic knowledge of the system hydronics. Another distinctive feature of the algorithm is the use of the AHU schedule as a means for inferring information on the internal loads, which is in general not available in practice. Simulation results for both CAV and VAV HVAC systems confirm the viability of the approach.