Classroom Cooling System with Neural Networks Based on Model Predictive Control (original) (raw)
Related papers
Neural networks based predictive control for thermal comfort and energy savings in public buildings
Energy and Buildings, 2012
This paper presents a neural network based predictive control for TRT (Top Gas Pressure Recovery Turbine). TRT is a nonlinear system with time-delay. To keep the top gas pressure stable, predictive control is developed and a RBF neural network which is trained using the input-output data of the practical process as predictive modeL RBFNN (radial basis function neural network) can approach any nonlinear function in theory. Simulation results show that the neural network based predictive controlier can obtain satisfactory performance in top gas pressure control.
Control of Air Conditioning Systems Using Neural Network
International Journal of Electronics and Electrical Engineering, 2015
Nowadays air conditioning system is the necessity part of human life. The different controllers used for controlling air conditioning system like on-off controllers or PID controllers. But these controllers cannot give the sufficient response and consume high power. This paper aims to control the air conditioning system such that the output temperature of air conditioning systems are getting as required by the operator with fast response and low consuming power. For this purpose neural network controllers are designed which is feedback to the air conditioning system. We designed such neural network control system that speculates its own control law. The advantage of using neural networks are this controller is self-learning system and give the faster and better response with comparison to another controllers and give the zero overshoot output.
International Journal on Advanced Science, Engineering and Information Technology, 2019
Predicating the required building energy when it is in the design stage and before being constructed considers a crucial step for in charge people. Hence, the main aim of this research is to accurately forecast the needed building cooling energy per hour for educational buildings at University of Technology in Iraq. For this purpose, the feed forward artificial neural network (ANN) has been selected as an efficient technique to develop such a predication system. Firstly, the main building parameters have been investigated and then only the most important ones were chosen to be used as inputs to the ANN model. However, due to the long time period that is required to collect actual consumed building energy in order to be employed for ANN model training, the hourly analysis program (HAP), which is a building simulation software, has been utilized to produce a database covering the summer months in Iraq. Different training algorithms and range of learning rate values have been investigated, and the Bayesian regularization backpropagation training algorithm and 0.05 learning rate were found very suitable for precise cooling energy prediction. To evaluate the performance of the optimized ANN model, mean square error (MSE) and correlation coefficient (R) have been adopted. The MSE and R indices for the predication results proved that the optimized ANN model is having a high predication accuracy with 5.99*10-6 and 0.9994, respectively.
Energy and Buildings
Most existing commercial building energy management systems (BEMS) are reactive rule-based. This means that an action is produced when an event occurs. In consequence, these systems cannot predict future scenarios and anticipate events to optimize building operation. This paper presents the procedure of implementing a predictive control strategy in a commercial BEMS for boilers in buildings, and describes the results achieved. The proposed control is based on a neural network that turns on the boiler each day at the optimum time, according to the surrounding environment, to achieve thermal comfort levels at the beginning of the working day. The control strategy presented in this paper is compared with the current control strategy implemented in BEMS that is based on scheduled on/off control. The control strategy was tested during one heating season and a set of key performance indicators were used to assess the benefits of the proposed control strategy. The results showed that the implementation of predictive control in a BEMS for building boilers can reduce the energy required to heat the building by around 20% without compromising the user's comfort.
A Model-Based Predictive Control Method for Thermal Environment in Low-Energy Buildings
Due to the increasing demands of energy conservation and emission reduction, the efficient control of indoor thermal environment aims at realizing thermal comfort with the least energy consumption. Taking an ultra-low energy public building as the case study, a physics-based model was established based on the measured datasets in order to provide databases for training data-driven modelling, which is a model-based predictive control (MPC) method established via neural network. After calibration, predictions of thermal comfort and loads under different temperature settings were obtained in summer and winter, based on input parameters of time, outdoor temperature and humidity, solar radiation and indoor humidity. Afterwards, hourly optimal temperature settings were recommended to take good use of the thermal inertia, and thus provide optimal references for the intelligent operation and control.
Neurobat, a Predictive and Adaptive Heating Control System Using Artificial Neural Networks
International Journal of Solar Energy, 2001
The paper describes a predictive and adaptive heating controller, using artificial neural networks to allow the adaptation of the control model to the real conditions (climate, building characteristics, user's behaviour). The controller algorithm has been developed and tested as a collaborative project between the CSEM (Centre Suisse d'Electronique et de Microtechnique, Neuchâtel, Switzerland, project leader), and the LESO-PB (Solar Energy and Building Physics Laboratory, EPFL, Lausanne, Switzerland). A significant support has been provided by leading Swiss industries in control systems. The project itself has been funded by the Swiss Federal Office of Energy (SFOE).
Influence of control logic on variation of indoor thermal environment for residential buildings
2016
This study proposes an advanced thermal control method that employs artificial neural network (ANN) models for predictive and adaptive thermal control. Two predictive and adaptive control logic approaches were proposed to simultaneously control indoor temperature and humidity as well as predicted mean vote (PMV) in a residential building. Their thermal performance was analysed and compared with that of non-ANN-based counterparts to evaluate architectural variables such as envelope insulation and building orientation. A numerical computer simulation method was used for the tests after demonstration of its validity based on comparison with results of field measurement. Analysis results revealed that the proposed predictive and adaptive control methods conditioned the indoor temperature, humidity and PMV effectively. The periods during which each thermal factor was in a comfortable range increased, and overshoots and undershoots out of the targeted comfortable ranges were reduced when using the ANN model. The results demonstrate the functionality of the proposed method for variation in architectural variables and that the ANN model has the potential to be successfully applied to building thermal controls.
Energies, 2014
This study aims at developing an indoor temperature control method that could provide comfortable thermal conditions by integrating heating system control and the opening conditions of building envelopes. Artificial neural network (ANN)-based temperature control logic was developed for the control of heating systems and openings at the building envelopes in a predictive and adaptive manner. Numerical comparative performance tests for the ANN-based temperature control logic and conventional non-ANN-based counterpart were conducted for single skin enveloped and double skin enveloped buildings after the simulation program was validated by comparing the simulation and the field measurement results. Analysis results revealed that the ANN-based control logic improved the indoor temperature environment with an increased comfortable temperature period and decreased overshoot and undershoot of temperatures outside of the operating range. The proposed logic did not show significant superiority in energy efficiency over the conventional logic. The ANN-based temperature control logic was able to maintain the indoor temperature more comfortably and with more stability within the operating range due to the predictive and adaptive features of ANN models.
Sustainability, 2020
For the sustainable use of building spaces, various methods have been studied to satisfy specific conditions required by the characteristics of space types and the energy use in operation. However, several effective control approaches adopting the latest statistical tools may have problems such as higher control precision increases energy consumption, or lower energy consumption decreases their control precision. This study proposes an optimized model to reach the indoor set-point temperature by controlling the amount of heating supply air and its temperature and investigates the efficiency of an adaptive controller to maintain indoor thermal comfort within setting ranges. In the consistency of the comfort level, the fuzzy logic controller was found to be 1.76% and the artificial neural network controller to be 17.83%, respectively, more efficient than the conventional thermostat. In addition, for energy use efficiency, both of the controllers were confirmed to be over 3.0% more efficient. Consequently, the network-based controller with the adaptive controller checking comfort levels effectively works to improve both energy efficiency and thermal comfort. This improvement can be significant in places such as commercial high-rises, large hospitals, and data centers where many spaces are intensively woven with appropriate thermal environments to maintain users' workability.
IFAC Proceedings Volumes (IFAC-PapersOnline), 2012
The problem of controlling a heating ventilating and air conditioning system in a single zone of a building is addressed. Its formulation is done in order to maintain acceptable thermal comfort for the occupants and to spend the least possible energy to achieve that. In most operating conditions these are conflicting goals, which require some sort of optimisation method to find appropriate solutions over time. In this work a model based predictive control methodology is proposed. It consists of three major components: the predictive models, implemented by radial basis function neural networks identified by means of a multi-objective genetic algorithm; the cost function that will be optimised to minimise energy consumption and provide adequate thermal comfort; and finally the optimisation method, in this case a discrete branch and bound approach. Each component will be described, and experimental results obtained within a classroom will be presented, demonstrating the feasibility and performance of the approach.