Artificial Intelligence! What Is It?AI involves a computer turning raw data into useful knowledge. For example, IBM has a system called Watson which inputs patient medical records and outputs a list of the patient’s afflictions and recommended treatment plans (to be reviewed by a physician). Normally it takes hours for a physician to review a patient’s medical records, diagnose patient’s afflictions and arrive at treatment plans. Computers can quickly handle much larger amounts of data than a person. Therefore a medical practitioner with an AI assistant doesn’t have to work as hard to see each patient. In other data-based professions like medicine, law, engineering, computer science, and teaching, AI can handle routine tasks like data management and free up time for practitioners to focus on their patients.
The following paragraphs cover several aspects of AI, and you can examine the ones where your interests lie.
- Artificial Intelligence Model Weakness
- Artificial Intelligence Evaluations
- Artificial Intelligence Potential Customers
- Artificial Intelligence Interesting Projects
- Programming Languages
- Python LIbraries
- Neural Network Frameworks
- Integrated Development Environments
- Hardware Technology
Artificial Intelligence as applied to Software/Energy Efficiency
Save Software costs and delays by using the Barkley Aproach.
Save energy and cut costs in buildings by employing the Barkley approach to energy savings.
The Barkley Building Energy Efficiency Program is systematic and automated. Our approach improves energy savings accuracy and reduces retrofit costs. During the energy audit process, we install data collection systems which give us a more accurate measurement of energy usage before, during and after the retrofit.
Rebates and financing are available. In addition, we can cut costs by training a client’s staff to perform data collection and monitoring of building energy usage for the Barkley Group during and after the audit and retrofit process is completed.
Our technical process is tailored to fit each client’s needs, whether it is for one or hundreds of buildings. We use the best instrumentation and available technology to insure satisfactory results.
We use computer models of buildings to calculate the expected energy usage for heating and cooling, as loads vary with changing outside weather – temperature, sun, and wind conditions.
There are many excellent software packages out there for analyzing energy usage in buildings. A widely accepted package is DOE-2. Another system, which is layered on top of DOE-2, is Green Building Studio by Autodesk Green Building Studio and also, Revit Energy Analysis System
The expensive part of analyzing the energy usage of a large number of buildings is getting the correct information about the buildings into the computer. This information includes but may not be limited to; data about building envelope area, window type and area, heating and cooling system types, manufacturers, size and seasonal energy efficiency, wall and roof thermal resistance numbers or U values, building ventilation fans.
Small errors in data can result in large errors in predicted building energy usage.
Years of experience enables us to look at the energy modeling program outputs, and quickly identify results which are unrealistic. In this way, we can quickly trace the unreasonable outputs to the input data and make appropriate corrections. Yet, there are some characteristics of the energy usage of a building which are very difficult to determine during just a walk through. To insure that we have a good understanding of how each building uses energy, we install real time sensors at key points in each building and send the data back to our servers over the internet.
We use this data to do multiple, self-verifying calculations of how each building uses energy, fine tune the analysis, eliminate ambiguities, and minimize unknowns. For example, it may not be possible to determine all the pathways for air infiltration into a building during a building walk through. Using multiple sources of information from a building to perform duplicate calculations improves accuracy and analysis repeatability.
Wireless temperature, humidity, CO, and CO2 and Global Positioning System enabled sensor packages are placed at selected points in each building and the information is collected and sent over the internet back to our severs. This information is then displayed on a 3 dimensional image of each building so that a building owner or manager can see what the environmental conditions are in their buildings, and how much energy each building is using.
1) A Group of buildings captured in 3 dimensions in a LiDar Point Cloud from an Aircraft, with one temperature sensor displayed for one of the buildings at the approximate location of the sensor in the building.
3D rendering of the LiDar point cloud from one side of the buildings
2) Street level fly-through with a thermographic camera. The blues are the colder building surfaces, and the yellows are the warmer building surfaces. Warmer building surfaces lose heat most quickly. We are developing technology so we can collect building envelope and building heat loss data much more quickly and completely from an aircraft flying over the buildings. We currently have a provisional patent on this technology
Integration of Terrestrial LiDar and Thermography. The reds, yellows and orange represent heat losses from the roof and walls of the buildings. The thermography gives the building surface temperatures, and the LiDar gives the building envelope dimensions.