Building Energy Control Research Articles

Installation and Testing of a Two-Level Model Predictive Control Building Energy Management System

Installation and Testing of a Two-Level Model Predictive Control Building Energy Management System

Machine Learning Applications for Smart Building Energy Utilization: A Survey

The United Nations launched sustainable development goals in 2015 that include goals for sustainable energy. From global energy consumption, households consume 20–30% of energy in Europe, North America and Asia; furthermore, the overall global energy consumption has steadily increased in the recent decades. Consequently, to meet the increased energy demand and to promote efficient energy consumption, there is a persistent need to develop applications enhancing utilization of energy in buildings. However, despite the potential significance of AI in this area, few surveys have systematically categorized these applications. Therefore, this paper presents a systematic review of the literature, and then creates a novel taxonomy for applications of smart building energy utilization. The contributions of this paper are (a) a systematic review of applications and machine learning methods for smart building energy utilization, (b) a novel taxonomy for the applications, (c) detailed analysis of these solutions and techniques used for the applications (electric grid, smart building energy management and control, maintenance and security, and personalization), and, finally, (d) a discussion on open issues and developments in the field.

Controllable thermal rectification design for buildings based on phase change composites

Phase-change material (PCM) is widely used in thermal management due to their unique thermal behavior. However, related research in thermal rectifier is mainly focused on exploring the principles at the fundamental device level, which results in a gap to real applications. Here, we propose a controllable thermal rectification design towards building applications through the direct adhesion of composite thermal rectification material (TRM) based on PCM and reduced graphene oxide (rGO) aerogel to ordinary concrete walls (CWs). The design is evaluated in detail by combining experiments and finite element analysis. It is found that, TRM can regulate the temperature difference on both sides of the TRM/CWs system by thermal rectification. The difference in two directions reaches to 13.8 K at the heat flow of 80 W/m2. In addition, the larger the change of thermal conductivity before and after phase change of TRM is, the more effective it is for regulating temperature difference in two directions. The stated technology has a wide range of applications for the thermal energy control in buildings with specific temperature requirements.

Passive sustainability strategies in traditional Gaziantep residences: a critical report on historical development

Abstract In recent years, the increase in the need for energy and the rapid depletion of energy resources have revealed that energy control in buildings is of vital importance. This situation requires energy efficiency not only in new buildings but also in existing building stocks. For this reason, passive strategic measures affecting thermal comfort and energy efficiency in historical buildings built in Gaziantep in the late 18th and early 19th centuries have been evaluated in this study. First, the cultural, technological and natural images of the region that are effective in the development of these strategies were identified. The reflection of these images on architectural design and the parameters affecting energy use as a result were revealed. In order to examine the effect of passive strategy measures in the spaces, a field study was carried out on a selected Gaziantep traditional residential building. For the field study, measurements were taken from the selected building and the street every hour during the day. These measurements have been made under the following conditions: sunny inside the courtyard, sunny outside the courtyard, shade inside the courtyard and shade outside the courtyard. Afterwards, all measurements have been compared. As a result, it is concluded that the courtyard systems that form the cultural, technological and natural image of the region have a great effect on thermal comfort.

Seasonal threshold to reduce false positives for prediction-based outlier detection in building energy data

Recent studies have shown that artificial intelligence-based model predictive control systems have the potential to optimize building operations for energy and cost. But for these systems to function as designed, they require high-quality datasets, which can be challenging when working with actual sensor data that may contain missing or outlier values. Using contaminated data in these carefully tuned machine learning algorithms can cause errors in the prediction results, leading to unexpected building operations. There is a need to clean incoming data in real-time by detecting the outliers in incoming sensor data and correcting them to an appropriate value before incorporating them into the prediction algorithm. This research proposes using a seasonal threshold for prediction-based point outlier detection methods to accurately detect the outliers while maintaining the reliability of the data by reducing false positive detections. The approach combining the global and seasonal threshold showed improvement to the detection accuracy by 83 % and reduced the falsely positives by up to 78 % for the best improvement, making the method a more trustworthy way to detect outliers in building energy data. This method can utilize an already existing model to detect outliers is a promising approach, especially for buildings that have already been implemented or are in the process of implementing an artificial intelligence-based control systems. This will allow the system to predict and control building energy without interruption and with reduced risk of unexpected operation due to corrupted data.

Spawn: coupling Modelica Buildings Library and EnergyPlus to enable new energy system and control applications

Spawn is DOE's next-generation tool chain for whole building energy control simulation. Spawn couples traditional imperative load-based envelope modelling with new equation-based modelling of HVAC and controls. Spawn uses EnergyPlus for the former and the Modelica Buildings Library for the latter. Because it leverages the Modelica Buildings Library, Spawn can evaluate advanced energy systems at the building and district scale, including new architectures and controls for heat pump systems with storage, and the coupling of such systems to electrical distribution networks. Spawn's Modelica integration likewise enables it to simulate realistic control sequences and therefore to bridge energy simulation and control implementation workflows. From EnergyPlus, Spawn inherits efficient envelope simulation and the ability to use existing envelope model authoring tools. This paper describes the architecture and implementation of Spawn, which automatically couples Modelica and EnergyPlus for run-time data exchange. This paper closes with examples that illustrate Spawn's modelling and simulation processes.

Modular responsive facade proposals based on semi-regular and demi-regular tessellation: daylighting and visual comfort

Responsive facades can reduce building energy consumption and control daylight and natural ventilation to improve user comfort. This study aims to develop alternative responsive facade systems based on semi-regular and demi-regular tessellations. For this purpose, first, the tessellation method used to generate responsive facades is introduced. Then, the geometric and parametric design principles and the movement capabilities of the proposed facade systems are presented. Finally, a set of analyses are performed to test and compare the performances of the facade systems based on daylight metrics and indoor glare comfort. This study contributes to the literature with the proposed facade systems that can adapt to changing environmental conditions, provide flexibility in shape control and simplicity in mechanism design, and improve building performance. The analysis results show that all the proposed facade systems provide the desired visual comfort and daylight levels at different configurations.

Utilization of Artificial Intelligence for Building Energy Control

Utilization of Artificial Intelligence for Building Energy Control

Action-based personalized dynamic thermal demand prediction with video cameras

Real-time and accurate prediction of indoor thermal demands is of great significance for improving building energy control and human satisfaction. This study proposes a system for dynamically predicting personal thermal demand with ordinary cameras, which integrates action acquisition, action feature extraction, action classification and thermal demand prediction. Three deep learning models were developed and integrated with multiobject tracking algorithms to build a personal thermal demand model. The proposed system and models were validated through a field measurement. The results show that the defined thermal adaptive behaviors can characterize human thermal sensation well. The action feature extraction model constructed based on the apparent feature extraction network combined with the proposed action classification model can determine whether the occupants have taken a thermal adaptive behavior with 92% accuracy and classify the behavior into specific categories with 86% accuracy. The thermal demand prediction model can achieve 95% accuracy, and the overall accuracy of the personal thermal demand model is 91%. This study provides technical and theoretical support for designing intelligent building decision-making schemes based on noninvasive thermal demand prediction.

Building flexibility revenue in modeled future bulk power systems with varying levels of renewable energy

The buildings sector can provide important demand-side flexibility for the integration of renewable energy. With the changing power system and rapid advancement in building energy control technology, technology providers and demand response aggregators need to know what potential revenues could be obtained by providing grid services. We dispatch a normalized and parameterized model of building load shifting against marginal service prices from grid investment and operational models to produce a database of the capacity, energy, and ancillary service revenues (gross value of providing bulk power system services) for a marginal kilowatt-hour (kWh) of shiftable load. The database is geographically disaggregated, hourly, and parameterized so that flexibility value can be estimated for a wide range of building technologies. The database covers the contiguous United States under three 2030 grid scenarios. Given the high degree of uncertainty in such grid projections, the results are perhaps best interpreted in terms of regional climate and grid mixes and are thus potentially applicable in non-U.S. contexts. Upon analyzing the results, we find the monthly mean gross value of building load shifting is 0–38 cents/kWh-day and the daily gross value of shifting the highest-value hour each day can be up to 620 cents/kWh-day. The different revenues obtained by aggregating results in different ways, as well as observed regional and seasonal differences, suggest different building technologies and grid environments might call for demand response program designs and business models.

CntrlDA: A building energy management control system with real-time adjustments. Application to indoor temperature

Rule-Based Control (RBC) and Model Predictive Control (MPC) have been traditionally used to control building heating, ventilation and air conditioning (HVAC) systems. They, however, present shortcomings when faced with efficiently controlling these systems at a larger level. Reinforcement Learning (RL) has recently emerged as a viable alternative, showing promising results compared to previous methods, but still having some difficulties with untrained situations or sudden changes. CntrlDA is our proposal on improving the RL formulation by coupling it with data assimilation (DA), a technique commonly used in numerical weather prediction. Our battery of experiments, in a building simulation environment, shows that training a RL control agent with DA and external data, leads to better performance than training the agent using only the simulation data. The RL control agent with DA maintains the temperature range 15.6% more often than the RL control agent without DA. It is also shown that by including a DA stage in the control process, the agent better deals with unexpected events (which are common in real-life systems and particularly in building energy control scenarios). We show that it maintains the range 15.4% more often than the system without DA with no significant added cost of resources.

Quantifying demand flexibility of building energy systems under uncertainty

Quantification of demand flexibility is of key importance to decision makers at the planning level, when designing demand response programs and determining capacity requirements for new battery storage installations, and, at the operation level, when assessing the demand-side potential for providing grid services. While different frameworks have been proposed for demand flexibility quantification, the literature still lacks a framework that considers uncertainty sources, affecting building energy control problems, when directly quantifying the demand flexibility potential. We propose a novel framework based on stochastic model predictive control for direct demand flexibility quantification in a bottom-up manner that can account for uncertainty. The proposed framework is utilized to quantify the flexibility potential of a building under uncertainty arising from weather forecasts and the available reduced-order model. The results are compared with those obtained using an existing deterministic approach. We demonstrate that the deterministic approach tends to overestimate the flexibility potential on account of ignoring uncertainty, and the occupants’ thermal comfort may be jeopardized if the grid asks for the flexibility estimated with the deterministic approach. The study accentuates the importance of developing an accurate model for model-based demand flexibility quantification owing to the significant effect of the modeling uncertainty.

Hardware-Based Emulator with Deep Learning Model for Building Energy Control and Prediction Based on Occupancy Sensors’ Data

Heating, ventilation, and air conditioning (HVAC) is the largest source of residential energy consumption. Occupancy sensors’ data can be used for HVAC control since it indicates the number of people in the building. HVAC and sensors form a typical cyber-physical system (CPS). In this paper, we aim to build a hardware-based emulation platform to study the occupancy data’s features, which can be further extracted by using machine learning models. In particular, we propose two hardware-based emulators to investigate the use of wired/wireless communication interfaces for occupancy sensor-based building CPS control, and the use of deep learning to predict the building energy consumption with the sensor data. We hypothesize is that the building energy consumption may be predicted by using the occupancy data collected by the sensors, and question what type of prediction model should be used to accurately predict the energy load. Another hypothesis is that an in-lab hardware/software platform could be built to emulate the occupancy sensing process. The machine learning algorithms can then be used to analyze the energy load based on the sensing data. To test the emulator, the occupancy data from the sensors is used to predict energy consumption. The synchronization scheme between sensors and the HVAC server will be discussed. We have built two hardware/software emulation platforms to investigate the sensor/HVAC integration strategies, and used an enhanced deep learning model—which has sequence-to-sequence long short-term memory (Seq2Seq LSTM)—with an attention model to predict the building energy consumption with the preservation of the intrinsic patterns. Because the long-range temporal dependencies are captured, the Seq2Seq models may provide a higher accuracy by using LSTM architectures with encoder and decoder. Meanwhile, LSTMs can capture the temporal and spatial patterns of time series data. The attention model can highlight the most relevant input information in the energy prediction by allocating the attention weights. The communication overhead between the sensors and the HVAC control server can also be alleviated via the attention mechanism, which can automatically ignore the irrelevant information and amplify the relevant information during CNN training. Our experiments and performance analysis show that, compared with the traditional LSTM neural network, the performance of the proposed method has a 30% higher prediction accuracy.

Research on Energy Control System of Office Building Based on the BIM and BP Neural Network Technology

As the proportion of building energy consumption in the total energy consumption continues to increase, more and more countries pay attention to the prediction analysis and operation control of building energy consumption. In this paper, an office building energy management and control system based on the building information modeling (BIM) and back-propagation neural network (BPNN) technology is proposed. Firstly, the BPNN algorithm is used to predict the energy consumption of the building, and then the model predictive control (MPC) method is employed to search the optimal control parameters to realize the energy saving of air-conditioning system and other equipment. In addition, the office building environment model is established through the BIM technology, which can real-time display the operating conditions of equipment and more visually control the equipment in the office building. The experimental results show that the system can satisfy the comfort level and significantly reduce the energy consumption.

Reexamine the relationship between new environmental paradigm and low-carbon consumption behavior

Carbon emission is an important cause of ecological imbalance and environmental pollution. Environmental issues such as energy consumption and carbon emissions caused by rapid economic development are important content that must be controlled at present. Reducing energy consumption in daily life is one of the most important energy-saving measures in densely populated cities. This study analyzed the environmental attitude of Shanghai citizens and the life energy-saving strategies promoted by the local government and found that the fulfillment of energy-saving behavior in daily life depends on people’s awareness of ecological balance and the degree of energy crisis. At the same time, the study found that energy-saving through building energy conservation, public resources and control of living consumption to promote the balanced development of ecology. The implementation of building energy conservation, electricity conservation, control of living consumption and other energy conservation behavior are also important ways to save energy.

A Powerful Tool for Optimal Control of Energy Systems in Sustainable Buildings: Distortion Power Bivector

In the field of building constructions, there is undeniably a growing need to optimize the energy systems which are a key target in new modern constructions and industrial buildings. In this sense, energy systems are being traced for the development of energy distribution networks that are increasingly smart, efficient, and sustainable. Modern generation and distribution energy systems, such as microgrids control systems, are being affected by the presence of linear and nonlinear loads, resulting a distorted voltage and current waveforms. Thus, it is stated that industrial and residential building heating and cooling loads behave essentially like sources of harmonics. This paper presents a new framework based on geometric algebra (GA) to the definition of a multivectorial distortion power concept, which is represented by a bivector that is geometrically interpreted to distinguish the rotated distortion and distortion power bivectors in these kinds of loads. Both bivectors, and their relations to the phase angles of distorted voltage are the main subject of this paper to interpret an optimal control of building energy. Numerical examples are used to illustrate of the suggested distortion power concept, as well as the information it provides for energy control in new buildings in a more sustainable way.

A data-driven electric water heater scheduling and control system

Domestic hot water (DHW) heating accounts for up to 30% of average household energy use. Compared to gas fired water heaters, electric water heaters (EWH) can be powered by renewable generation resources, thus making it a potential renewable heating option. Furthermore, with the growing need for energy storage, incorporation of renewable resources, and initiatives worldwide, the electrification of DHW heating is expected to continue the rapid growth. However, many commercial EWH products with monitoring and alerting functionalities lack the intelligence to optimize and perform predictive control with data; on the other hand, research studies with refined models and simulations come short in incorporating real-time data and providing robust optimal controls under uncertainties in real-world settings. This paper presents a EWH Smart Scheduling and Control System using data-driven disturbance forecasts in a robust Model Predictive Control (MPC) to accomplish various demand side management objectives. Testing with a real-world EWH dataset and a two-state EWH model, robust MPC simulations are conducted on a central EWH supplying DHW for a multi-unit apartment building with quantified prediction uncertainty. Results show that the proposed system is capable of anticipating DHW demand with an uncertainty interval covering up to 97% of the actual demand during the test days and reducing electricity cost up to 33.2% as well as maintaining a desired DHW temperature without affecting user comfort. Further, the flexibility of the system to alter load profiles under different Demand Response (DR) programs is demonstrated. Reductions in both power and water consumption can be accomplished. The proposed system can create an implementable solution of forecasting DHW usage and optimizing controls as a part of a robust and reliable building energy management and control system in real-world settings.

Distributed real-time temperature and energy control of energy efficient buildings via geothermal heat pumps

Geothermal Heat Pumps (GHPs) are a type of Heating Ventilation and Air Conditioning (HVAC) systems that use low-temperature resources from soil and groundwater for heating/cooling. In recent years, there is an increasing interest in GHP systems due to their high energy efficiency and abundant geothermal resources. Thus, the optimization and control design of the GHP system has become a hot topic. On the other hand, as the GHP system is an ideal responsive load, the mechanism design for the GHP system to realize demand response (DR) in a Virtual Power Plant (VPP) without affecting user comfort is particularly essential. In this paper, we propose a distributed real-time temperature and energy management method via the GHP systems for multi-buildings, where both floor and radiator heating/cooling distribution subsystems in multiple thermal zones are considered. We design an energy demand response mechanism for a single GHP to track the given energy consumption command for participating in the VPP aggregation/disaggregation. Besides, a coordination mechanism for multiple GHPs is designed for the community-level operator in joining the VPP aggregation/disaggregation. Both designed schemes are scalable and do not need to measure or predict any exogenous disturbances such as the outdoor temperature and the heating disturbances from external sources, e.g., user activity and device operation. Finally, four numerical examples for the simulation of two different scenarios demonstrate the effectiveness of the proposed methods.

Simulation Based Energy Control and Comfort Management in Buildings Using Multi-Objective Optimization Routine

Building energy management systems with high-level of sophistication have to control and manage a large set of actuators and other equipment and evaluate performance of each and every-subsystem on periodic basis. In the present study, a control algorithm has been developed as an engineered solution for intelligent energy control and comfort management in buildings. A hybrid genetic algorithm - particle swarm optimization based multi-objective optimization routine is developed to compute the optimal set-point level of heating, ventilation, and air conditioning and lighting systems with a view to balancing energy consumption and occupants' comfort. Occupants' comfort is evaluated for indoor air quality as CO2 concentration, thermal and visual comfort. Case studies with a different set of optimal parameters have been worked out to calculate the amount of energy consumed as well as comfort level achieved. Overall occupants' comfort was improved by 17% and daily, weekly and monthly building energy consumption was reduced by 2.5%, 7.7%, and 17.9%, respectively. The developed intelligent control strategy can be integrated with building automation systems to achieve finely tuned real-time optimized comfort management

Social Network Analysis of Building Energy and Carbon Policy Networks in Developing Countries

More than two thirds of countries worldwide, the majority being developing countries still have not implemented mandatory regulations to control building energy usage and carbon emissions despite the rapidly increasing building stock in these regions. Thus, it is vital to develop effective energy and carbon policies for buildings in developing countries. Developing such effective policies in turn, needs well-structured policy networks consisting of efficiently connected key actors in policy implementation. A database search revealed significantly less progress in implementing the building energy and carbon policies in developing countries, compared with developed countries. Previous literature suggested Policy Network Analysis (PNA) as a powerful tool to assess current policy level actor coalitions, determine structural holes in stakeholder integration and decide on future policy improvements. Subsequently, a database search was carried out through the “Web of Science” publications repository to identify the previous PNA approaches and to identify a potential way forward. Among the available methods, Social Network Analysis (SNA) was identified as the most suitable and frequently used method which provides extensive analytical capabilities for policy network visualization and analysis. The preliminary findings lead to the novel SNA model that is proposed to map and assess required building energy and carbon policy networks which integrate both policy actors and constraints. This proposed model can be effectively utilized to identify the significance, centrality and relational structures of policy actors and policy constraints. This paper will benefit future researchers and relevant policy makers by providing an effective platform on which to further develop more effective energy and carbon policies for the building sector by strengthening stakeholder collaboration and integration.