Source: INTERNATIONAL ELECTRONIC MACHINES CORPORATION submitted to
COMPLETE AUTONOMOUS VENTILATION ENVIRONMENT SYSTEM (CAVES) FOR SMALL FARMS
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
COMPLETE
Funding Source
Reporting Frequency
Annual
Accession No.
1025831
Grant No.
2021-33530-34699
Cumulative Award Amt.
$99,834.00
Proposal No.
2021-00914
Multistate No.
(N/A)
Project Start Date
Jul 1, 2021
Project End Date
Feb 28, 2022
Grant Year
2021
Program Code
[8.12]- Small and Mid-Size Farms
Project Director
Metacarpa, D.
Recipient Organization
INTERNATIONAL ELECTRONIC MACHINES CORPORATION
850 RIVER ST
TROY,NY 121801239
Performing Department
(N/A)
Non Technical Summary
International Electronic Machines Corp., a leader in the field of smart sensors and controls for health, maintenance, and other fields, will develop the Complete Autonomous Ventilation Environment System (CAVES), an intelligent, customizable, scalable, adaptable control and ventilation system that will directly monitor all the key parameters of barn environment - temperature, humidity, fan speed/wind velocity, vapors/gases such as ammonia, hydrogen sulfide, CO, and CO2, -- and combine them with knowledge of the barn size and configuration and the number and type of livestock to reliably ensure proper ventilation at minimal power usage by controlling both fans and inlet configuration. CAVES will be able to be retrofitted to existing ventilation fans and may be powered via line power or by a solar system with appropriate energy storage as the dramatic reduction in energy consumption would now allow for such a renewable energy system for ventilation to be affordable; one aspect of the project will be evaluating designs for extremely lightweight solar installations that can go on very low-weight-tolerance roofs, or that can be installed on-ground conveniently and with minimal space requirements. Overall, CAVES should reduce power demand by between 40 and 80 percent over standard ventilation operations while maintaining a completely safe and comfortable environment for the target livestock, offering a very quick ROI even on a small farm.
Animal Health Component
100%
Research Effort Categories
Basic
(N/A)
Applied
100%
Developmental
(N/A)
Classification

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
40453302020100%
Goals / Objectives
International Electronic Machines Corp., a leader in the field of smart sensors and controls for health, maintenance, and other fields, will develop the Complete Autonomous Ventilation Environment System (CAVES), an intelligent, customizable, scalable, adaptable control and ventilation system that will directly monitor all the key parameters of barn environment - temperature, humidity, fan speed/wind velocity, vapors/gases such as ammonia, hydrogen sulfide, CO, and CO2, -- and combine them with knowledge of the barn size and configuration and the number and type of livestock to reliably ensure proper ventilation at minimal power usage by controlling both fans and inlet configuration. CAVES will be able to be retrofitted to existing ventilation fans and may be powered via line power or by a solar system with appropriate energy storage as the dramatic reduction in energy consumption would now allow for such a renewable energy system for ventilation to be affordable; one aspect of the project will be evaluating designs for extremely lightweight solar installations that can go on very low-weight-tolerance roofs, or that can be installed on-ground conveniently and with minimal space requirements. Overall, CAVES should reduce power demand by between 40 and 80 percent over standard ventilation operations while maintaining a completely safe and comfortable environment for the target livestock, offering a very quick ROI even on a small farm.Reduce ventilation energy costs by between forty and eighty percent (translating to a total energy savings of between 8-16% based on an average of 20% of a farm's energy used for ventilation); this produces a ROI of between 6 months and 2 years based on calculated cost of CAVES.This will be completed by incorporating gas sensors with variable speed fan control to maintain ambient conditions (temperature and gas concentrations) combined with processing to understand a particular barns exhaust attributes.Determine the ability to use renawable energy systems, specifically photovoltaic, based upon the reduced energy consumptionMaintain livestock comfort throughout the year. This is determined by maintaining temperature and background gas levels at a to be determined value through the use of the control of the exhaustReduce or eliminate discomfort and health impacts of irritant gases on both livestock and farm workersAs determined by comparing the pre and post background gas ambient concentrations and time durations
Project Methods
Work Plan1.1Task 1: Requirements and Specifications Finalization1.1.1Task DescriptionIEM will first assemble a preliminary requirements and specifications document,(with more details on proposed components, as described in the following Tasks) and a set of questions addressing issues that are currently unclear. This will include, though not be limited to, typical fan installations found in smaller farms (number, size, type), with considerations including proportion of farms with line power to barns, loading limits for various types of barns, and so on.1.2Task 2: Research and Select Sensors and Processing System1.2.1Task DescriptionIn this task, IEM will select the sensors for use in Phase I.The major sensor types expected for Phase I of CAVES include temperature, humidity, wind velocity (inlet air speed), and one or more gas sensors (ammonia, CO2, H2S, etc.). IEM has considerable experience in the selection of sensors for particular applications. Some of these, such as temperature and humidity, are so widely available that there are numerous selections of roughly equal quality. There are two general types of wind sensor: the venerable anemometerand the ultrasonic wind sensor, which present their own advantages and disadvantages. The primary advantage of the anemometer is that it is very inexpensive, while the ultrasonic sensors are much more costly but have no moving parts and tend to be much more reliable.1.3Task 3: Research and Design Retrofit Fan Control1.3.1Task DescriptionIn this task, IEM will research the necessary design and electronic elements necessary to retrofit existing types of barn fans to make them controllable variable speed fans.Fans as a class are some of the simplest of electronic devices: they consist of a motor (copper windings, magnets, and ferrous cores), a driving shaft spun by the motor, and blades attached to the shaft. The speed at which it turns depends on the specific design of the motor and the frequency and/or voltage of electricity supplied. Both of the latter may be easily changed by appropriate electronics. The key for CAVES is to ensure that any retrofit for existing fans is easily installed, inexpensive, reliable, safe, and efficient.1.4Task 4: Determine Key Parameters for Ventilation1.4.1Task DescriptionIn this task, IEM will determine all the key parameters to be used to direct and control the ventilation operations of CAVES.This task will proceed in parallel to the sensor selection tasks, as the determination of key parameters will of necessity influence the sensor selection. Some ventilation parameters are very basic to the industry, and are even covered by various industry directives (such as typical air volume replacements per hour), but CAVES will be performing intelligent demand and condition controlled ventilation and must evaluate the entire ventilation situation properly.Some parameters already known to be of interest include the following:Temperature and humidity.Type of livestock.Ammonia and H2SCO2/COInlet airspeedBarn configurationEnergy use1.5Task 5: Select/Design Power Subsystems1.5.1Task DescriptionIn this task, IEM will examine and select the power source/approach to be used for CAVES.Most barns are supplied with line power, making it simple to power the CAVES units. For this, IEM will design a line power supply converter to provide all the components of CAVES with appropriate types and voltages of power. IEM has performed similar work many times for multiple other projects, and would anticipate no particular difficulties in doing so for CAVES.CAVES' energy reduction approach makes solar power a potentially feasible source for full-barn ventilation, something which would be much more problematic with the more common and energy-intensive methods. Therefore, IEM will examine the feasibility of a solar-powered version of CAVES for use in areas where line power is unavailable, scarce, or unreliable.1.6Task 6: Create Phase I Ventilation Control and Communication Software1.6.1Task DescriptionIn this task, IEM will develop the Phase I version of CAVES' ventilation control and communication software.Currently, IEM assumes the communication between the CAVES nodes will be wireless, although with power connections comes the possibility of data transfer over power lines, which would eliminate the possible complications of wireless communication. However, for purposes of CAVES it will be assumed that wireless is required for data transfer and control, as this is the more challenging scenario and does not prevent CAVES from working in the wired situation.The more challenging software task is the designing of the actual ventilation decisionmaking code, as this will include both simple objective standards (maintain livestock comfort zone of X) and far more interesting learning and decisionmaking operations. For example, while it may be relatively straightforward (though perhaps not easy) to simulate airflow through an idealized barn such as that illustrated in Figure 2, airflow through a legacy barn with connected additionsis extremely complex. IEM would not expect, in fact, to model such airflow, but instead to learn it empirically - by having the connected CAVES nodes observe, record, and extrapolate the changes in temperature, humidity, and gas concentration as ventilation proceeds. This novel feature is what in particular will enable the largest cost savings across a wide variety of barn structure designs enabling the wide impact to the small farm community.?1.7Task 7: Assemble, Test, and Demonstrate Phase I CAVES1.7.1Task Description?The results of this task will be: initial prototypes of all CAVES components, test results of the individual subsystems, a list of Action Items from the test results, revised subsystem designs, final subsystem prototypes, a fully assembled Phase I CAVES Prototype, test results of the Prototype, demonstrations of the Phase I capabilities, and assembled data of performance results from the tests and demonstrations thereof.1.8Task 8: Project Management and ReportingThis task will be ongoing throughout the project, as it involves the coordination of all operations within the project and all interactions with outside parties, including arranging kickoff and other meetings with the USDA, preparing interim and final reports, monitoring progress of each task and solving potential scheduling conflicts, contacting vendors and consultants, and so on. A schedule of all tasks, assuming an 8-month duration and start date of 7/1/2021

Progress 07/01/21 to 02/28/22

Outputs
Target Audience:During this project, IEM visited several local farms to increase our contact with the eventual target clientele, and to learn more about the actual operating conditions of real-life small farms. These included numerous farmsat various points in and around the Capital District of New York. In particular the farms were focused on those that are small and medium sized dairy farms with mostly older (>50 year) barns. The focus on the older barns was to include a wide variety of barn configurations that commonly have been modified over the years. In addition these older "legacy" barns do not commonly have a configuration with a focus on exhaust ventilation efficiency. This allowed IEM to learn key points about the use of ventilation, access to electrical power, and other operational elements of farms relevant to CAVES, and to familiarize some of the local target audience with IEM and our goals. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? Nothing Reported How have the results been disseminated to communities of interest?As described under Target Audience, IEM has visited a number of local farms and apprised them of IEM's interest and the nature of our CAVES project. IEM has developeda one-page summary brochure to be used to educate other farmers and farming groups of this emerging technology. Further contact of note, including approaching one or more for field testing opportunities, is expected if and when CAVES proceeds to Phase II. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? Executive Summary: IEM has achieved the Phase I goals of the Complete Autonomous Ventilation Environment System (CAVES) project. Testing shows that CAVES can accurately detect and predict CO2 concentrations exceeding comfort or health limits and proactively trigger the fans to prevent that event; the fans are operated only when needed, showing a significant reduction in energy usage over that described by actual farm owner-operators. The best combination of algorithms for fan control showed both very good maintenance of CO2 concentrations and a drastic reduction in fan usage; under CAVES' control the fan system used <20% of its normal energy. IEM also began work on evaluating modeling software which may be needed during Phase II. The economics of small farms are highly constrained, and saving energy can have a significant impact on their ability to perform well. In addition, it is known that livestock are better producers when kept at optimal comfort, as CAVES will allow, and reduction of energy usage not only reduces overall pollution and demand on the power grid, but also opens up opportunities for the use of solar or other renewable sources in areas that were previously impractical. As described in more detail below, IEM has: Researched, tested, and selected key gas sensors for Phase I Designed and built a first prototype of the CAVES system, including: Independent wireless sensor nodes Fan control nodes (which may also include sensors) Central Gateway First-cut smart control software Designed and created a Phase I Testbed with: Multiple controllable fans of different types and sizes Controllable temperature Controllable CO2 levels Performed initial experiments Evaluated the results of the initial experiments Refined the control algorithms Performed final Phase I experiments Evaluated the final results, showing All three algorithms both controlled CO2 concentrations and reduced energy usage The third (combining both thresholding and rate of change) demonstrated both very fine control of CO2 levels and an energy reduction of 89% Showed a likely ROI for the system would be less than 2 years Began work on using airflow and contaminant modelingsoftware for modeling barn gas behaviors Work/Accomplishments Following an internal kickoff meeting to set expectations and priorities for the Phase I work, IEM commenced the technical efforts. Sensors were one early focus of the work, as the selected sensors would drive much of the operation and capabilities of CAVES. IEM examined multiple vendors for their offerings in various gas areas (most importantly ammonia, hydrogen sulfide, and CO2). After research, IEM selected a particular CO2 sensor for the Phase I work; CO2 is easily dispersed, relatively safe and nonirritating in use, and available from local suppliers. Mechanically, if the system functions well for CO2 it should function well for any of the gases of interest, and there was thus no good reason to demand two or three separate gas sensors and gas sources for Phase I; Phase II work would be expected to expand on the sensing capabilities. IEM examined multiple wireless communications strategies, including a custom low-powerdesign. After testing and examination of the application space, IEM selected a particular method that suited the demands of CAVES. IEM also selected the core elements for the wireless sensor node, including onboard processor, power source and distribution, and so on. During this process, IEM constructed a network design and concept of operations which included a gateway for central communication and control, sensor nodes, and fan control nodes. Research into fan designs showed several possible types in use. The simplest were on-off designs that could not be easily modified for speed control, and thus represented the largest challenge for CAVES' planned benefits (since CAVES could then only save energy through a smart duty-cycle rather than through reduction in fan speed). IEM chose to take this approach and designed our main fan control nodes with a single on-off control relay, relying on smart software to achieve energy savings. Interviews with target users (farmers) indicated that when fans were needed, they would be turned on in the morning and left on essentially throughout the work day; on hot days they might remain on constantly to prevent overheating. IEM did design nodes capable of controlling fans that did have variable-speed capability, demonstrating that they would function as well as the simpler nodes. For testing, IEM adapted a large environmental chamber. The chamber is roughly the size of a small (20') cargo container, provided with environmental controls, and can be sealed against weather; it is used for testing IEM's larger sensor assemblies for our wayside railroad systems. Based on the volume of the test chamber, IEM examined a wide range of fans and selected a large, medium, and several smaller "muffin" fans which would be able to exchange the entire volume of the test chamber in periods ranging from seconds to many minutes. During construction of the insert, IEM conducted simple tests of the CAVES system - sensors, actuator circuits, and gateway - in the chamber with a cylinder of CO2 as a source for detection; the door would be opened whenever the system indicated the fans should be activated (as the fans were not yet in place). This allowed the development of the first-cut version of the CAVES software, first with a simple thresholding algorithm, and next with a learning and anticipation algorithm to detect and predict the increases of CO2 and activate fans before CO2 concentration exceeded limits. IEM conducted two sets of major testing, with an evaluation period following the first that allowed IEM to refine the algorithms and testing methodologies. Ultimately, IEM was able to show that even with relatively simple algorithmic control, CAVES could reliably maintain comfortable CO2 levels and do so by running the fans only 11% of the time (as compared to the established situation in which fans are simply left running). This reduction in power would allow the projected final CAVES system to have an ROI of less than two years, and may encourage the use of renewable solar power for such applications. IEM also evaluated theutility of modeling software for the later Phase II development by acquiring and performing testing of airflow and gaseous contamination modeling in order to determine the algorithms starting points for fan control based upon minimized power usage.

Publications