Energy Modeling: Predicting and Optimizing Consumption Course

Lesson 1 • Key Metrics and Performance Indicators

Defines energy intensity, load factor, and demand profiles used throughout modeling. Connects raw consumption data to actionable performance benchmarks.

Lesson 2 • Introduction to Modeling Approaches

Contrasts physics-based, statistical, and hybrid modeling paradigms. Guides selection of the appropriate approach for a given project scope.

Lesson 3 • Data Sources and Measurement Methods

Surveys utility bills, sub-metering, and sensor networks as primary data inputs. Establishes data quality standards required for reliable model inputs.

Lesson 1 • Internal and Solar Heat Gains

Quantifies occupant, lighting, and equipment heat gains alongside solar irradiance. Demonstrates how internal loads shift peak demand timing and magnitude.

Lesson 2 • Thermal Mass and Dynamic Response

Examines how mass, capacitance, and time constants affect load profiles over time. Enables accurate dynamic simulation rather than static load estimates.

Lesson 3 • Building Envelope Performance

Analyzes insulation, glazing, and air barriers as determinants of thermal resistance. Connects envelope properties to heating and cooling load magnitudes.

Lesson 4 • HVAC System Load Interactions

Links envelope and internal gains to HVAC sizing and part-load performance curves. Prepares students to model system-level energy consumption accurately.

Lesson 1 • Regression Fundamentals for Energy Data

Introduces ordinary least squares and its assumptions applied to energy datasets. Establishes the statistical foundation for all regression-based energy models.

Lesson 2 • Time-Series Methods for Energy

Introduces autocorrelation, ARIMA, and seasonal decomposition for interval energy data. Extends regression skills to temporal patterns and short-term forecasting.

Lesson 3 • Change-Point and Piecewise Models

Applies ASHRAE Guideline 14-style change-point regression to capture nonlinear weather response. Enables accurate savings measurement across heating and cooling regimes.

Lesson 4 • Variable Selection and Feature Engineering

Covers weather variables, occupancy proxies, and calendar features as model predictors. Teaches systematic feature construction to improve model accuracy.

Lesson 1 • Simulation Engine Architecture

Explains the calculation sequence, time-step logic, and zone network of major simulation engines. Provides the conceptual map needed to use tools effectively.

Lesson 2 • Building Geometry and Zoning

Covers geometry creation, thermal zone assignment, and adjacency conditions in simulation tools. Accurate geometry is the prerequisite for valid load calculations.

Lesson 3 • Simulation Output Analysis

Extracts, aggregates, and visualizes hourly outputs to diagnose model behavior. Connects raw simulation results to project decisions and reporting needs.

Lesson 4 • Weather Files and Climate Inputs

Explains TMY, AMY, and future climate weather files and their impact on results. Correct weather selection is critical for location-specific energy predictions.

Lesson 1 • Calibration Techniques and Workflows

Applies manual adjustment, sensitivity analysis, and automated optimization to close model-to-meter gaps. Provides a repeatable workflow applicable across project types.

Lesson 2 • Interval Data Calibration

Uses 15-minute and hourly metered data to calibrate load shape, not just monthly totals. Interval calibration reveals operational issues invisible in monthly billing data.

Lesson 3 • Uncertainty Sources in Models

Identifies input uncertainty in schedules, equipment, and weather as primary calibration drivers. Prioritizing high-impact uncertainties accelerates convergence to a calibrated model.

Lesson 4 • Calibration Concepts and Standards

Defines calibration objectives, tolerance thresholds, and applicable industry protocols. Frames calibration as an iterative, evidence-based process rather than curve fitting.

Lesson 1 • Parametric and Sensitivity Analysis

Runs parametric sweeps across ECM variables to map the energy-cost response surface. Reveals which parameters drive the largest savings before detailed analysis.

Lesson 2 • Energy Conservation Measure Identification

Systematically identifies ECMs across envelope, systems, controls, and operations. Structured identification prevents overlooking high-value opportunities.

Lesson 3 • Optimization Algorithms for Energy

Introduces genetic algorithms, particle swarm, and gradient-based methods for multi-variable optimization. Enables automated search for optimal ECM combinations beyond manual analysis.

Lesson 4 • Economic Evaluation of Measures

Applies simple payback, NPV, and IRR to rank ECMs by financial attractiveness. Translates energy savings into economic terms decision-makers can act on.

Lesson 1 • Tree-Based and Ensemble Methods

Applies random forests and gradient boosting to short-term energy load forecasting. Tree-based models handle nonlinearity and mixed feature types common in energy data.

Lesson 2 • Anomaly Detection and Fault Diagnosis

Uses isolation forests, autoencoders, and control charts to flag abnormal consumption. Enables proactive fault detection before waste accumulates.

Lesson 3 • ML Fundamentals for Energy Applications

Covers bias-variance tradeoff, train-test splits, and cross-validation in an energy context. Establishes rigorous ML workflow habits before introducing specific algorithms.

Lesson 4 • Model Deployment and Monitoring

Covers model serialization, API integration, and drift detection for production energy models. Ensures ML models remain accurate as building operations evolve over time.

Lesson 1 • Portfolio-Scale Energy Analysis

Scales single-building methods to multi-site portfolios using normalization and clustering. Identifies highest-priority sites for intervention across large asset bases.

Lesson 2 • Quality Assurance and Peer Review

Establishes QA checklists, independent review processes, and version control for model deliverables. Ensures professional-grade reliability before results influence capital decisions.

Lesson 3 • Decarbonization Pathway Modeling

Models electrification, fuel switching, and renewable integration scenarios against carbon targets. Translates energy model outputs into greenhouse gas reduction roadmaps.

Lesson 4 • Communicating Results to Stakeholders

Translates technical model outputs into executive summaries, dashboards, and investment cases. Bridges the gap between modeling rigor and decision-maker comprehension.