Occupant Behavior Simulation for Accurate Energy Demand Prediction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing building performance simulation tools rely on fixed schedules for occupant behavior, which fail to provide realistic energy consumption predictions due to their lack of detail and realism.

Innovation Solution

A system and method for generating occupant activities based on recorded schedules, using a computer simulation that constructs histograms, normalizes feature values, and generates attributes for simulated occupant behavior, allowing for more detailed and realistic occupancy models.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If fixed schedules are used for occupant behavior simulation, then the simulation process is simple, but the energy consumption predictions are not realistic

Engineering Contradiction:

Improvesimplicity of simulation processVSAvoidaccuracy of energy consumption predictions

Core Design Contradiction:

Ease of operationVSMeasurement precision

Solution Approach 1:

The patent transforms fixed static schedules into dynamic stochastic simulations that generate variable occupant behaviors. The system uses probability distributions and random sampling to create diverse activity patterns while maintaining statistical consistency with observed behavior, enabling realistic energy predictions without excessive complexity

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes the parameters of occupant behavior from fixed values to probability distributions. By defining activity types, durations, and timings as stochastic parameters with specific distributions, the simulation generates varied realistic scenarios while maintaining control through distribution parameters derived from observed data

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If detailed occupant behavior models are created, then energy demand predictions become accurate, but the model complexity increases

Engineering Contradiction:

Improveaccuracy of energy demand predictionsVSAvoidcomplexity of occupancy model

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments occupant behavior into discrete activity types (work, meetings, breaks, etc.) with specific attributes. Each activity type is modeled separately with its own probability distributions and rules, allowing detailed representation of complex behaviors through composition of simpler segmented components

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system creates synthetic copies of observed occupant behaviors through stochastic simulation. By sampling from probability distributions fitted to observed data, the model generates virtual occupant schedules that replicate real behavior patterns without requiring direct measurement of every individual, reducing complexity while maintaining accuracy

Inventive Principle:

Principle #26Copying

Data Source

PatentUS10586181B2Generation of occupant activities based on recorded occupant behavior

Publication Date: 2020.03.10 AUTODESK INC

AI summary

A method and apparatus for simulating occupant behavior in buildings may be used to predict the energy use of a building structure. The activities of actual building occupants are recorded and provided as an input to the occupant behavior simulation. The occupant behavior simulation generates simulated occupant schedules with similar behavioral patterns. An arbitrary set of factors can be used to select plausible activity types, durations, and numbers of participants during an occupant behavior simulation. The simulated occupant schedules may then be incorporated into a building performance simulation to help architects predict the energy demand associated with different building design options.