#Regression_analysis is a #statistical method that examines the connection between #energy_consumption and #factors like #building_characteristics, #weather_data, #occupancy_patterns, #equipment_efficiency, #production, and other variables. By fitting regression models to the data, typically using techniques like #multiple_linear or #nonlinear regression, analysts can estimate the #influence of each factor on energy usage and #predict future consumption levels under various scenarios.
As energy auditors, having access to data like monthly #production_quantities and monthly #electrical_consumption provides us with valuable insights and opportunities. This data can be analyzed as follows:
📍 Constructing a graph with production on the x-axis and electrical consumption on the y-axis reveals a scatter of points. If these points form a #linear pattern, it indicates a #positive correlation—meaning as production increases, electrical consumption also increases, affirming the correctness of the process.
📍 Examining the data points helps us determine if there's any recurring #pattern in production over time, if there are short-term #trends, or if there are #unusual_instances where low production coincides with high electrical consumption.
📍 Creating a linearization for data points involves generating a #line that represents the points, accompanied by an #equation that describes the relationship between variables.
📈 The equation facilitates #forecasting of electrical consumption for a known quantity of production in the future.
📈 The slope of the #line represent the : ( )kWh/ton.
📈 The point where the line intersects the y-axis denotes the #base_load of the facility, representing the energy consumed when production is halted. This includes energy usage from HVAC, lighting, CCTV, etc. A high base load serves as an indicator that something may be amiss in the process.
📎 The line now represents my production expectations. Some points fall below it, while others are above it. We scrutinize the points below the line to determine why production quantity was achieved with lower consumption. We'll continue this analysis, ensuring the decrease in consumption isn't due to external factors, before implementing any actions.
