Gregor P. Henze

Experimental Analysis of Model-Based Predictive Optimal Control for Active and Passive Building Thermal Storage Inventory

This paper demonstrates model-based predictive optimal control of active and passive building thermal storage inventory in a test facility in real time using time-of-use differentiated electricity prices without demand charges. A novel supervisory controller successfully executed a three-step procedure consisting of (1) short-term weather prediction, (2) optimization of control strategy over the next planning horizon using a calibrated building model, and (3) post-processing of the optimal strategy to yield a control command for the current time step that can be executed in the test facility. All primary and secondary building mechanical systems were effectively orchestrated by the model-based predictive optimal controller in real time while observing comfort and operational constraints. It was determined that even when the optimal controller is given imperfect weather forecasts and when the building model used for planning control strategies does not match the actual building perfectly, measured utility cost savings relative to conventional building operation can be substantial. Central requirements are a facility that lends itself to passive storage utilization and a building model that includes a realistic plant representation. Savings associated with passive building thermal storage inventory proved to be small in this case because the test facility is not an ideal candidate for the investigated control technology. Moreover, the facilitys central plant revealed the idiosyncratic behavior that the chiller operation in the ice-making mode was more energy efficient than in the chilled-water mode. Field experimentation is now required in a suitable commercial building with sufficient thermal mass, an active TES system, and a climate conducive to passive storage utilization over a longer testing period to support the laboratory findings presented in this study.

Development of a Predictive Optimal Controller for Thermal Energy Storage Systems

This paper describes the development and simulation of a predictive optimal controller for thermal energy storage systems. The “optimal” strategy minimizes the cost of operating the cooling plant over the simulation horizon. The particular case of a popular ice storage system (ice-on-coil with internal melt) was investigated in a simulation environment. Various predictor models were analyzed with respect to their performance in forecasting cooling load data and information on ambient conditions (dry-bulb and wet-bulb temperatures). The predictor model provides load and weather information to the optimal controller in discrete time steps. An optimal storage charging and discharging strategy was planned at every time step over a fixed look-ahead time window utilizing newly available information. The first action of the optimal sequence of actions was executed over the next time step and the planning process was repeated at every following time step. The effect of the length of the planning horizon was inves...

The performance of occupancy-based lighting control systems: A review

This paper reviews the literature on occupancy-based lighting control as a prelude to the application of sensor networks to building management. Many buildings include systems to detect occupancy and control building services. Current systems use single measurement points to detect occupancy, and there can be significant uncertainty associated with the measurement of occupancy. Long time delay and high detector sensitivity settings compensate for this uncertainty, but these diminish the savings that could be achieved with more accurate occupancy measurement. More effective control may be provided by more extensive sensing, using a network of occupancy sensors, and more extensive analysis of sensor data. The literature reviewed in this paper establishes the need for an investigation of the performance of sensor networks when used for lighting control.

A model predictive control optimization environment for real-time commercial building application

A model predictive control (MPC) environment is described. The environment integrates Matlab and EnergyPlus with a modified particle swarm optimizer to predict optimal building control strategies. A supporting framework is described which couples the environment to a building automation system, allowing real-time optimization considering operator overrides and updated weather forecasts. Challenges unique to integration with EnergyPlus for real-time optimization are discussed. Application of the environment is demonstrated in two simulation cases. First, the environment is used to determine hourly cooling set points minimizing daily energy cost for EnergyPluss Benchmark Large Office building. Results suggest 5% cost savings during the study period. Second, the environment is used to determine hourly supply water temperature and circulator availability that minimize daily energy consumption for a small office building having a thermally activated building system (TABS). Compared to the base case, energy savings up to 54% are reported, with often improved occupant comfort.

Impact of Forecasting Accuracy on Predictive Optimal Control of Active and Passive Building Thermal Storage Inventory

This paper evaluates the benefits of combined optimal control of both passive building thermal capacitance and active thermal energy storage systems to minimize total utility cost in the presence of forecasting uncertainty in the required short-term weather forecasts. Selected short-term weather forecasting models are introduced and investigated with respect to their forecasting accuracy as measured by root mean square error, mean bias error, and the coefficient of variation. The most complex model, a seasonal autoregressive integrated moving average (SARIMA), shows the worst performance, followed by a static predictor model that references standard weather archives. The best prediction accuracy is found for bin models that develop a characteristic daily profile from observations collected over the past 30 or 60 days. The model that projects yesterdays patterns one day into the future proved to be a surprisingly poor predictor. We test the predictor models in the context of a predictive optimal control task that optimizes building global temperature setpoints and active thermal energy storage charge/discharge rates in a closed-loop mode. For the four locations investigated in this article—Chicago, IL, Denver, CO, Omaha, NE, and Phoenix, AZ—it was determined that the 30-day and 60-day bin predictor models lead to utility cost savings that are only marginally inferior compared to a hypothetical perfect predictor that perfectly anticipates the weather during the next planning horizon. In summary, the predictive optimal control of active and passive building thermal storage inventory using time-of-use electrical utility rates with significant on-peak to off-peak rate differentials and demand charges is a highly promising control strategy when perfect weather forecasts are available. The primary finding of this paper is that it takes only very simple short-term prediction models to realize almost all of the theoretical potential of this technology.

Guidelines for improved performance of ice storage systems

Abstract This paper describes simulation-based results of an investigation of a commercial cooling plant with an ice storage system. Various ice storage systems, chiller compressors, and building types were analyzed under four different control strategies. Optimal control as the strategy that minimizes the total operating cost (demand and energy charges) served as a benchmark to assess the relative performance of three conventional controls (chiller-priority, constant-proportion, and storage-priority control) and to determine aspects in need of improvement in order to apply these conventional controls better and to enhance the cost saving potential of ice storage systems. Independent of the non-cooling electrical load profile, it was found that good efficiency of the cooling plant in the icemaking mode and rate structures with strong load-shifting incentives are prerequisites for making cool storage successful. Chillers with poor performance at subfreezing evaporator temperatures require significant on- to off-peak differentials in the energy and demand rates to yield substantial savings. The relative performance benefit of optimal control over conventional controls increases when rate-based load-shifting incentives are weak. With cooling-related electrical loads being large compared to non-cooling loads, all conventional controls improve their performance when slowly recharging during off-peak periods to contain off-peak demand. On-peak demand reduction of storage-priority is near-optimal for many cases. Guidelines are presented to improve the load-shifting performance of chiller-priority and constant-proportion control.

Statistical Analysis of Neural Networks as Applied to Building Energy Prediction

It has been shown that a neural network with sufficient hidden units can approximate any continuous function defined on a closed and bounded set. This has inspired the use of neural networks as general nonlinear regression models. As with other nonlinear regression models, tools of conventional statistical analysis can be applied to neural networks to yield a test for the relevance or irrelevance of a free parameter. The test, a version of Walds test, can be extended to yield a test for the relevance or irrelevance of an input variable. This test was applied to the building energy use data of the Energy Prediction Shootout II contest. Input variables were selected by initially constructing a neural network model which had many inputs, then cutting out the inputs which were deemed irrelevant on the basis of the Wald test. Time-lagged values were included for some input variables, with the time lag chosen by inspecting the autocovariance function of the candidate variable. The results of the contest entry are summarized, and the benefits of applying Walds test to this problem are assessed.

Parametric Analysis of Active and Passive Building Thermal Storage Utilization

Cooling of commercial buildings contributes significantly to the peak demand placed on an electrical utility grid. Time-of-use electricity rates encourage shifting of electrical loads to off-peak periods at night and on weekends. Buildings can respond to these pricing signals by shifting cooling-related thermal loads either by precooling the buildings massive structure or by using active thermal energy storage systems such as ice storage. While these two thermal batteries have been engaged separately in the past, this paper investigates the merits of harnessing both storage media concurrently in the context of optimal control for a range of selected parameters. A parametric analysis was conducted utilizing an EnergyPlus-based simulation environment to assess the effects of building mass, electrical utility rates, season and location, economizer operation, central plant size, and thermal comfort. The findings reveal that the cooling-related on-peak electrical demand and utility cost of commercial buildings can be substantially reduced by harnessing both thermal storage inventories using optimal control for a wide range of conditions.

Adaptive Optimal Control of a Grid-Independent Photovoltaic System

This paper investigates adaptive optimal control of a grid-independent photovoltaic system consisting of a collector, storage, and a load. The algorithm is based on Q-Learning, a model-free reinforcement learning algorithm, which optimizes control performance through exploration. Q-Learning is used in a simulation study to find a policy which performs better than a conventional control strategy with respect to a cost function which places more weight on meeting a critical base load than on those non-critical loads exceeding the base load.Copyright

Energy and Cost Minimal Control of Active and Passive Building Thermal Storage Inventory

In contrast to building energy conversion equipment, less improvement has been achieved in thermal energy distribution, storage and control systems in terms of energy efficiency and peak load reduction potential. Cooling of commercial buildings contributes significantly to the peak demand placed on an electrical utility grid and time-of-use electricity rates are designed to encourage shifting of electrical loads to off-peak periods at night and on weekends. Buildings can respond to these pricing signals by shifting cooling-related thermal loads either by precooling the building s massive structure (passive storage) or by using active thermal energy storage systems such as ice storage. Recent theoretical and experimental work showed that the simultaneous utilization of active and passive building thermal storage inventory can save significant amounts of utility costs to the building operator, yet increased electrical energy consumption may result. The article investigates the relationship between cost savings and energy consumption associated with conventional control, minimal cost and minimal energy control, while accounting for variations in fan power consumption, chiller capacity, chiller coefftcient-of-performance, and part-load performance. The model-based predictive building controller is employed to either minimize electricity cost including a target demand charge or electrical energy consumption. This work shows that buildings can be operated in a demand-responsive fashion to substantially reduce utility costs with marginal increases in overall energy consumption. In the case of energy optimal control, the reference control was replicated, i.e., if only energy consumption is of concern, neither active nor passive building thermal storage should be utilized. On the other hand, cost optimal control suggests strongly utilizing both thermal storage inventories.