Contact Us

For inquiries about air conditioning equipment optimization in buildings (to be developed), please contact us here.

Challenges in managing the operation of heat source equipment in buildings

Heat source equipment in buildings
Operation Management Issues

When operated by the operation supervisor

problem_1

Too much human cost
Too much

problem_2

Skilled
Know-how required

For automatic operation based on pre-setting

problem_1

Changes in conditions in the building
Inability to make precise operational changes in response to changes in building conditions
Inability to make precise operational changes in response to changing conditions in the building

problem_2

Energy
Wasteful in usage and not good for the environment
Not good for the environment

solution_1

AI-BASED AIR CONDITIONING CONTROL SOLVES THIS PROBLEM.

AI-based air conditioning control solves

Predicting Air Conditioning Demand and Optimal Operation Control
Significant cost reduction and contribution to SDGs through the use of AI

Prediction of air conditioning demand
and optimal operation control
Significant cost reduction and contribution to SDGs by utilizing AI
and contribution to the SDGs

WHILE UTILIZING THE EXISTING AIR CONDITIONING SYSTEM, AN AI CONTROLLER IS RETROFITTED TO THE CONTROL EQUIPMENT. ARAYATHE GOAL IS TO ACHIEVE SIGNIFICANT COST REDUCTIONS (20% OR MORE) BY USING THE AUTONOMOUS AI TECHNOLOGY OF THE AI CONTROLLER TO PREDICT THE REQUIRED AIR CONDITIONING CAPACITY (DEMAND) AND TO MANUFACTURE HEAT IN ACCORDANCE WITH DEMAND AND THE ELECTRICITY RATE STRUCTURE.

DEMAND FORECASTING AI

and

AI FOR ENERGY-EFFICIENT HEAT PRODUCTION

Realized in

DEMAND FORECASTING AI

and

energy-saving

FOR DISTRICT HEATING AND COOLING (DHC) AND LARGE HEAT SOURCES

AI FOR THERMAL MANUFACTURINGRealized in

FOR DISTRICT HEATING AND COOLING (DHC) AND LARGE HEAT SOURCES

Air Conditioning Optimization Solutions
Air conditioning optimization solutions
logo-card-1
1-1

99.8

%

note (supplementary information) symbol

With sophisticated demand forecasting of

Reduce waste and cut costs

Based on historical data and weather data
Predicts future heat demand
Visualization of forecast results

*PREDICTION AFTER 1 HOUR WITH AI CHIREI®.
1-2

HEAT SOURCE CONTROL BY AUTONOMOUS AI

about

20

%

Cost reduction of

Based on demand forecasts
Optimal operation and control of heat source equipment
Cost reductions are realized.

1-3

With a wealth of experience in AI development
, we offer the most suitable proposal for each individual.

Based on our extensive experience in AI development
Based on our extensive experience in AI development
the best solution for your situation.

THREE ADVANTAGES OF USING AI FOR AIR CONDITIONING CONTROL

card-2_1

01

Highly accurate forecasting and control by utilizing a variety of input data
control and cost reduction by utilizing a variety of input data

Highly accurate forecasting and control by utilizing a variety of input data
Highly accurate prediction and control can be achieved and costs can be reduced.
and cost reduction

In addition to historical data on power used in the facility, a variety of information sources can be used for forecasting and control, including external temperature, humidity, and weather forecasts, as well as human flow data from IoT devices and surveillance cameras. Based on this information, AI predictive models are used to estimate the facility's future heat load, heat production by the chiller, and power consumption, and to calculate the best control settings to implement forecasts.

Examples of data that can be used
card-2_1-1

Temperature/Humidity

card-2_1-2

weather forecast

card-2_1-3

Camera information
(number of people counted)

card-2_2

02

Data-driven settings that do not rely on rules of thumb
Comfortable environment at all times

Data-driven without relying on rules of thumb
Data-driven settings that do not rely on rules of thumb
Ensures a comfortable environment at all times

Multiple parameters, such as temperature setting and operating time, can be predicted based on past, present, future, and other data, rather than rules of thumb, for detailed and optimal settings. Flexible processing can be changed according to the situation, reducing waste and ensuring that users of a building or facility are comfortable at all times.

LABOR COSTS CAN ALSO BE REDUCED BECAUSE AI MAKES SETTINGS BASED ON DATA AND OPERATES AUTOMATICALLY.
card-2_2-1
card-2_3

03

Ensure environmental sustainability and contribute to the SDGs

Ensure environmental sustainability and contribute to the SDGs

Air conditioning accounts for about 50% of a building's energy use, and its optimization is an important step toward achieving the Sustainable Development Goals (SDGs). This will contribute to reducing carbon dioxide emissions and mitigating climate change.

Power ratio by use in an average office building
card-2_3-1

Electricity is sometimes perceived as a clean energy source, but generating electricity emits a lot of carbon dioxide. Reducing electricity use will help reduce carbon dioxide emissions and mitigate climate change.

Source: Agency for Natural Resources and Energy estimates

Expected cost reductions and energy savings

Under complex constraints that cannot be handled by rule-based systems, optimal control is implemented in detail according to the situation.
Control is performed by taking into account not only the current situation but also predicted future conditions, making it possible to achieve optimal (energy-saving) control when viewed as a whole, rather than temporarily.
(energy saving) by considering not only the current situation but also the predicted future state.

card-3_1
card-3_2

Download a detailed introduction to our air conditioning optimization solutions.

ArayaFeatures and Examples of Demonstrations of Air Conditioning Optimization Solution of

ArayaAir Conditioning Optimization of Solution Features and Examples of Demonstration Experiments

feature_1

99.8% accurate demand forecasting to reduce waste and cut costs

AI district heating and cooling (a.k.a. AI Chirei®), which uses AI to reduce CO2 emissions and costs, has been developed and tested. AI Chirei®)," which reduces CO2 emissions and costs through AI, was developed and a demonstration experiment was launched to achieve highly accurate demand
Demand forecasting with a high degree of accuracy was achieved. (Joint verification experiment by Nikken Sekkei Ltd.
Ltd. and Araya)

AI CHIREIⓇ IS A REGISTERED TRADEMARK OF NIKKEN SEKKEI LTD.

DEMONSTRATION EXPERIMENT 1: HIGHLY ACCURATE DEMAND FORECASTING IN DISTRICT HEATING AND COOLING (DHC)

Facilities where the demonstration tests were carried out

DEMONSTRATION IS UNDERWAY FOR THE FOLLOWING DHCS.

Heat producing machines: about 5 units, heat storage tanks: available
Customers: Commercial facilities and office buildings (total area: approx. 70,000 m2)
Average daily demand for chilled water: Approx. 300 GJ in summer and 30 GJ in winter
Annual heat production cost: 50-100 million yen

Scope of the demonstration

Only forecasting of demand * Forecasts are displayed on the screen, and heat production instructions are rule-based (manual) by the operator.
Forecasts are made within the production capacity of one heat production machine (approx. 2000 MJ/30 min).

Progress of the Demonstration Experiment

On the simulation, we were able to predict the amount of demand with very high accuracy at a practical level. This enables us to save energy by reducing wasteful startup of heat source equipment.
This enables energy savings by reducing wasteful startup of heat source equipment.
Currently, the system is being installed at an actual site and is being tested by on-site operators to verify its energy-saving effects.

feature_1-1
feature_1

HEAT SOURCE CONTROL BY AUTONOMOUS AI REDUCES COSTS BY ABOUT 20%.

AUTONOMOUS AI IS A METHOD IN WHICH THE AI ITSELF LEARNS BY TRIAL AND ERROR WHAT ACTION (AIR CONDITIONING CONTROL) IS BEST (COST REDUCTION), AND WHEN THE AI TAKES AN ACTION (SUCH AS TURNING ON/OFF THE NUMBER OF HEAT PRODUCING MACHINES IN OPERATION), THE CONDITION OF THE AIR CONDITIONING ENVIRONMENT (SUCH AS CHANGES IN HEAT STORAGE AND COLD SUPPLY) CHANGES, AND WHEN GOOD RESULTS ARE OBTAINED (SUCH AS HEAT PRODUCTION COSTS BEING REDUCED), THE AI IS HIGHLY REWARDED. AI CAN LEARN BY TRIAL AND ERROR SO THAT IT RECEIVES MORE REWARDS.

feature_2-1

Demonstration Experiment 2: Annual heat production costs reduced by approximately 20% at a large heat source

Facilities where the demonstration tests were carried out

Demonstrations were conducted with the following large heat sources

Heat producing machines: about 5 units, heat storage tanks: Yes
Commercial facility (approx. 60,000 m2)
Cold water demand (daily average): approx. 200 GJ in summer, approx. 30 GJ in winter
Discounts are available if usage is reduced during peak hours.
Annual heat production cost: 20-40 million yen

Scope of the demonstration

Simulation of heat production only (demand forecasting uses rule-based algorithms in the existing system)
Verification of heat production cost savings on simulation

Results of the demonstration

Annual heat production costs could be reduced by about 20% (on simulation). The cost savings are particularly high in the summer months by avoiding heat production during more expensive times of the year.

feature_2-2

[To be developed] Approach to optimize air conditioning equipment in a building

CURRENTLY, WE ARE PREPARING TO DEVELOP A SYSTEM TO CONTROL MULTIPLE AIR CONDITIONING UNITS IN A BUILDING TO OPTIMIZE POWER CONSUMPTION WHILE SATISFYING ROOM TEMPERATURE, COMFORT, ETC., BASED ON OUR EXPERIENCE IN AI DEVELOPMENT FOR DISTRICT HEATING AND COOLING AND LARGE HEAT SOURCES. IF ANYONE IS INTERESTED IN MORE DETAILS, PLEASE FEEL FREE TO CONTACT US.

plan_1 plan_2

CASE STUDY OF AIR CONDITIONING OPTIMIZATION BY AI

seminarImage

DEVELOPED "AI DISTRICT HEATING AND COOLING (A.K.A. AI CHIREI®)" AND STARTED DEMONSTRATION TESTS.

Industry: Construction companies, energy supply companies

Issue: Approximately 90% of the existing facilities were built 20 years ago, and their energy-saving performance was declining.

seminarImage

Significantly reduced power consumption through the use of deep reinforcement learning

Industry: Construction companies, office buildings

In facility replacement planning, it is difficult to determine when and which components should be changed during renovation to save energy.

PAGE TOP