Energy Demand Forecasting
Energy demand is the result of a large number of factors. Therefore, forecasting requires advanced AI technology as well as in-depth knowledge of energy.
We are able to provide highly accurate forecasts by leveraging our advanced AI technology and over 15 years of experience in energy measurement and management for more than 5,000 facilities.
We are able to provide highly accurate forecasts by leveraging our advanced AI technology and over 15 years of experience in energy measurement and management for more than 5,000 facilities.
Prediction target
Electricity Demand
It is possible to forecast electricity demand from high-voltage facilities of all types, including factories, office buildings, schools, hospitals, etc., as well as general households.
Heat value (heat load)
It is possible to forecast the heat load of boilers and hot and cold water generators. It can also be used to forecast district heat supply.
Other
Please consult us about other energy needs (gas, water, etc.) as well.Features
AAutomatic learning to improve accuracy
This system automatically learns the differences between forecasts and actual results on a continuous basis. This makes it possible to make facility-specific forecasts.
BVisualization and analysis of demand factors
It is important to analyze the factors of energy demand. We will conduct multifaceted analysis with available data such as correlation with meteorological data, regression analysis of historical data, and seasonal and weekly periodicity.
CTwo types of forecasts
Our energy demand forecast outputs two types of forecasts, one for planning and one for adjustment.For planning
Forecast the energy demand of a future day for 24 hours at a time. This can be used for power procurement planning and advance planning of facility operation.
For adjustment
Adjusts and re-outputs forecasts up to several hours in the future, reflecting the weather data and actual power demand of the immediately preceding day. This can be used to adjust facility operations in response to sudden changes in weather conditions.
Example of use
Energy supply companies
Urban planning (smart city)
Building automation
Air conditioning control
Power generation private consumption system
BEMS
HEMS
Solar Power Generation Prediction
We have achieved highly accurate solar power generation prediction by learning theoretical power generation values calculated from solar radiation prediction data and actual power generation data.
Forecast target
Solar power generation facilities in generalFrom general households to large scale (up to mega solar).
Example of prediction results
High prediction accuracy even in cloudy weather
Learning the characteristics of the facility
Easy to detect power generation efficiency decline and failure
Heat quantity prediction
By learning meteorological data such as temperature, which has the greatest impact on air conditioning load, and past performance, it is possible to directly predict the heat quantity (heat load) of chilled water generators and boilers.
Various data affecting heat load can be added according to the environment, so please consult us.
Various data affecting heat load can be added according to the environment, so please consult us.
