More efficient building energy management leads to lower energy consumption and cost, higher occupancy comfort and less detrimental effects on the environment. Improving building energy management with advanced integrated building control provides a tool to coordinate and optimize control of multiple indoor parameters by considering their interconnected effects on building energy consumption and comfort.
A building integrated optimization requires an approach to calculate building energy consumption, operate in real time, optimize building control parameters, and be able to modify systems operations or schedules in response to environmental or demand response signals inputs. The integrated optimization has significant effects on reductions in energy use and energy costs, reductions in peak load, and improvement of indoor environment quality without replacing the existing equipment. Most of previous research in integrated building control just focused on optimization of specific zone or some of the possible parameters. They also applied their optimization for the current hour without considering its effect on future-hours.
The main goal of this research is to develop an advanced building operation optimization tool for integrated control of lighting, shade, ventilation and heating and cooling systems for whole buildings to reduce building energy consumption, operation cost, and peak load while satisfying occupancy comfort. Also, this optimization tool is capable of coordinating integrated control and demand response by real-time modification of time-of-use prices that are received from utilities. In addition, it applies multi-hour optimization by optimizing several hours simultaneously and considering effects of current hour control parameters on future hour energy consumption. 
As a first step, integrated optimization is investigated based on a developed and validated RC-network model of a typical small office building. Nonlinear optimization is applied to the RC-network model that is created in MATLAB. The optimization results show energy savings up to 35% more than the scheduled control. In addition, multi-hour optimization saved up to 4% of energy cost compare to optimization based on the current hour.
For more accurate building energy and cost calculation, using building simulation software is essential. In this research DOE-2 is chosen as an open source building energy use analysis tool and modified based on integrated optimization requirements by adding functions to DOE-2 source code. DOE-2 requires modifications to accept the control parameters’ online and hourly bases. Accomplished modification is validated by simulating nighttime ventilation strategy. Also, the daylighting and window energy calculation algorithm is modified to operate based on shade position instead of just open or closed shade.
A building-integrated optimization tool is developed by integrating the genetic algorithm optimization method in MATLAB with building energy and cost calculation software (DOE-2). This integrated optimization tool simulates and optimizes building control parameters such as indoor temperatures, shade position, artificial light power, and outdoor air ventilation rates for an entire building. This optimization tool can be easy applied to any type of building and system when their models are available in DOE-2. Moreover, different strategies are proposed for increasing speed of optimization. First, a rule-based decision-making tool is used before integrated optimization that modifies the control parameters optimization domain. Decision-making rules are developed based on sample integrated optimization results. Second, the neural network is trained for energy consumption prediction of building based on energy consumption results from DOE-2 for random control parameters. This trained neural network is connected to a genetic algorithm and replaces DOE-2 for the energy consumption calculation.  Finally, a local optimization method is used after the genetic algorithm to search around genetic algorithm results of control parameters for new control parameters with lower building energy consumption.
The integrated MATLAB and DOE-2 optimization tool is initially evaluated by investigating nighttime ventilation and shade position optimization. The results for nighttime ventilation optimization show total energy savings up to 8% and cooling energy consumption reduction up to 23%. Higher savings occurred on days with high diurnal temperature range and average outdoor temperature near 17 ˚C. The results for shade position optimization indicate that in hot days shades stay nearly closed since the effect of solar heat gain, which increases cooling energy consumption in addition to the detrimental effect of conduction heat transfer, is more effective and important than lighting energy reduction from daylighting. Also, in transient seasons when the building is in heating mode, shades mostly stay open since heat gain and illuminance transmission from windows reduce both heating and lighting energy consumption. In addition, using thick shades and a lower illuminance set-point give optimization more flexibility for energy savings. 
Finally the integrated MATLAB and DOE-2 optimization tool for whole building energy optimization is applied to a typical office building in Montreal. The results show energy savings between 10% and 30%; also higher energy savings potential could be expected during transient seasons compared to very hot or very cold seasons. The results also show peak load savings up to 40%. 
Keywords: building model, energy consumption, integrated control, optimization, DOE-2