Grantee Research Project Results
Final Report: A Smart-Sensor Approach to Automating and Optimizing Agricultural Water Reuse
EPA Contract Number: 68HERC22C0005Title: A Smart-Sensor Approach to Automating and Optimizing Agricultural Water Reuse
Investigators: Mather, William
Small Business: Quantitative BioSciences, Inc
EPA Contact: Richards, April
Phase: I
Project Period: December 1, 2021 through May 31, 2022
Project Amount: $100,000
RFA: Small Business Innovation Research (SBIR) Phase I (2022) RFA Text | Recipients Lists
Research Category: Small Business Innovation Research (SBIR) , SBIR - Water
Description:
The goal of this SBIR Phase I project was to develop a smart- sensor approach for a water reuse system. Access to clean, reliable water supplies is critical to our quality of life and our economy, yet across the country millions of Americans do not have access to groundwater that meets drinking water standards. The contaminants that plague drinking water range from common water toxins, such as arsenic and cadmium, to excess nutrients (nitrogen and phosphorus), which are particularly problematic in agricultural regions. Over the next thirty years, the world's population is expected to reach at least 10 billion people, which will create an increase in demand for food and other products that cannot be met with traditional agricultural approaches. In fact, agricultural resources are almost already completely exploited, with very little arable land left to farm, and climate change and population sprawl pose impending threats that will need to be met with rapid innovation. Ultimately, drastic changes in agrotechnology are needed to sustainably source critical food supplies and other essential products.
Recent advances in sensing and machine learning technologies are opening up the possibility for "smart sensor" approaches to automation and optimization of agricultural resources that could have a significant impact on our ability to meet the demands of a growing population while addressing the need for sustainable farming. While the emergence of these technologies is exciting, innovative projects are needed to integrate state-of-the-art approaches into simple, user-friendly systems that lend themselves to rapid and widespread adoption. Our project aims to meet this challenge by combining a state-of-the-art real-time nutrient sensor with an artificial intelligence platform that will incorporate multiple streams of data to provide a "drop in" technology that can integrate with almost any type of low-input water reuse technology to enable simple, affordable, and sustainable water reclamation for a broad range of agricultural applications.
Our project work plan involves three Objectives tailored toward demonstrating technical feasibility of our technology:
Objective 1: Video and audio integration for the purpose of water irrigation management: Novel monitoring functionality will be added to our CLFarm platform to provide awareness of irrigation processes and their byproducts. (Task 1) An infrared spectrum-enabled "phenocam" camera and low-cost networked cameras will be leveraged to provide readouts of crop health, lagoon state, and algae pond state. (Task 2) We will use a collection of microphones (contact, directional, and omnidirectional) to provide audio waveforms that will be digested into quantitative and qualitative content, including fluid flow state (on vs. off) and mechanical operation state (normal vs. deviated states). For both tasks, traditional signal processing and machine learning-based processing will be used.
Objective 2: Qube monitoring of lagoon water and an algae-based advanced water treatment system: We will use our proprietary Qube sensor to monitor both lagoon water and an existing algae raceway pond fed by lagoon water to demonstrate the utility of the Qube's continuous nutrient monitoring for nutrient management, including as a compliance-checking tool before reuse of nutrient-rich water for irrigation. (Task 1) We will enable fluid flow from a collection lagoon full of wastewater (from cattle barn flush runoff and digester waste) into a test raceway pond with cultured algae. (Task 2) A Qube will monitor incoming lagoon water and water from the algae pond alongside YSI sensors (including those for temperature, pH, ORP, salinity, and DO) to demonstrate performance of these sensors in establishing nutrient awareness in a non-potable water reuse system.
Objective 3: CLFarm webpage and mobile app creation for a non-potable water reuse system: A proto-type capable of integrating the products of Objectives 1 and 2 will be created to provide both qualitative and quantitative content based on historical and real-time measurements. Here, we will heavily leverage our existing work on frontend webpages and backend processes for our CloudLab platform to accelerate progress. (Task 1) We will develop webpage interfaces and necessary backend integration for monitoring and analysis based on video, audio, Qube, and other measurements. These webpages will be sufficient to monitor all activities measured by our non-potable water reuse system. (Task 2) We will develop a mobile phone app for CLFarm to improve convenience and enable mobile phone technology, including proper mobile phone notifications to assist in automated equipment monitoring and awareness of other events.
In the next section, we will discuss our success in achieving these Objectives and demonstrating feasibility of a novel non-potable water reuse system suitable for commercial development.
Summary/Accomplishments (Outputs/Outcomes):
By the end of this 6-month Phase I SBIR project, we completed all major Objectives and established feasibility of a novel water reuse monitoring and algae treatment system. Objective 1 focused on novel and low-cost audio and video monitoring solutions that provided new insight into the water reuse operation at Fiscalini Farms. Our most novel results here concerned the development of real time acoustic monitoring of water flows using custom hardware based on microphones coupled to system-on-chip, microcontroller, and wireless (WiFi and LoRa) technologies. Integration of audio and video technology allowed us to gather a real time understanding of water flows, fluid level heights, and crop health among other measurements. Objective 2 focused on development and monitoring of an advanced algae treatment pond that, after scaling up, could help to capture lagoon nutrients in lieu of depositing nutrients back into fields or waste streams. Our approach depended not only on developing measurement technology but also on developing algae remediation technologies. We developed effective lagoon water clarification strategies and robust algae cultures that could process high levels of ammonia from lagoon water. This resulting pond culture provided us with time-dependent ammonia concentration chemical signal that we sensed using our Qube biosensor, and the results matched well when compared to standard Hach tests for ammonia. Objective 3 provided web-enabled interfaces for the data generated by Objectives 1 and 2. Results housed on a central server were accessed by a Quasar web/mobile app of our design that could request live or historical data from our data API and display results to a user. Data presented to the user included current and historical images, acoustic spectra, crop health, and notification ``icons'' that summarize system state using mixed graphical and text information. This web/mobile app interface is easy to modify and customize, such that future versions of the interface can be tailored to specific customers.
Conclusions:
Overall, these results support high technical feasibility of the water reuse monitoring platform proposed, and a Phase II project that increases the scale and polish of this platform would likely produce a viable commercial product. Objective 1 results established proof of principle for cost-effective acoustic measurement devices that could stream information concerning the ``soundscape'' of a water reuse operation or similar operation. Though we only explored acoustic signal for water flows and machine operation, the acoustic contribution from human, livestock, and potentially even pest sources could also be monitored. Furthermore, machine learning edge computing layers that go beyond our simpler proof of principle processing would allow us to extract richer features to create a "smart ear" distributed network. Similarly, a dense network of video sensors would provide additional information that could be enhanced by machine learning via edge computing. This system could be powered by solar power as needed (no significant failures in our engineered solar system were noticed), with long distance power-over-ethernet providing power to nearby devices without the need for protective conduit. Objective 2 results established feasibility concerning the creation, maintenance, and monitoring of an algae treatment pond capable of processing raw lagoon water that is heavily laden with manure products and very high nitrogen (ammonia) content. Clarification of lagoon water for measurement and algae pond-based nutrient capture was itself a major step forward that will likely impact the design of a potential Phase II project, since the cost-effective chitosan clarification agent we selected will allow for significant volumes of lagoon water to be clarified before algae pond processing. Development of ammonia-tolerant and environmentally-robust algae culture was also useful as a key step towards a larger operation. Demonstration that our Qube technology could measure the ammonia content of the pond as ammonia was captured provided the final necessary component we would desire for a managed algae pond in a future Phase II project. Objective 3 results demonstrated that we could integrate and present all of these results to an end user via modern web/mobile app technology. Use of Quasar as a web framework was key in this regard, since Quasar provided most all standard features desired for a web/mobile app. After creation of middleware to manage backend processes to provide the data to this webpage, a mixture of existing and new frontend Quasar components provided an example of how a more fully-featured app would present our monitoring information to an operator or manager in a future Phase II project. Collectively across all Objectives, our results demonstrated that QBI is extremely well-poised for future development of a water reuse operation using the technologies presented.
Our commercialization efforts for this project involved a combination of work with our Technological and Business Assistance (TABA) provider, Foresight, and customer outreach performed by our own leadership team. Foresight was instrumental in producing a major commercial deliverable, the Commercialization Readiness Assessment Report, a document that provided valuable project-specific commercialization analysis related to our technology. In particular, Foresight determined the following: "Based on the research, patent, and competition, the proposed technology appears to be favorably positioned to bring a new, novel, and unique product to the marketplace. No instance of a direct competitor has been found in our research. There are some cases of work that is like the proposed research, but none of the research found a technology that utilizes both a biosensor and a continuous audio/video monitor for water systems. Successful development and testing of the proposed technology, along with the identification of markets that are receptive to the adoption of this technology, could place Quantitative Biosciences in the lead to capture this market niche." This and other information in the report will be invaluable for our company as we plan our commercial direction and formulate our work plan for a potential Phase II of this project.
Our leadership team also put a significant effort into customer outreach. We used targeted internet searches and LinkedIn to discover potential customers and end-users for our technology. We focused both on agricultural end users as well as customers in other industries that have a need for improved "smart" sensing, such as water treatment. Many of these contacts were enthusiastic and spoke about a need for a more flexible and comprehensive approach to process control and optimization and provided advice and suggestions for features that would facilitate easy end user adoption.
As a result of this work, we have also enrolled in a technology acceleration platform with Isle Utilities. Isle's TAG program accelerates the market uptake of new water technologies by engaging the industry during the pre-commercial stages of development and by leveraging external investment from venture capital investors. They operate TAG forums for water utilities in different regions around the world, and after introducing our technology to them, they offered to give us a place in several of their upcoming forums. We will likely participate in their West Coast program in the fall in order to take advantage of potential partnerships in our region.
We also met with Rabia Chaudhry, the TPOC for this project. During that meeting, we were made aware of several opportunities that may allow us to better integrate into water safety monitoring. In particular, California's leadership in the direct potable water reuse space was identified as a possible good pairing with our small molecule and pathogen monitoring technologies in addition to technologies developed in the current project. We were also made aware of REUSExplorer, which will further help identify where our technologies intersect with water regulations. Lastly, the National Blue Ribbon Commission was identified as a resource to identify monitoring needs.
Our final month of commercialization effort involved continued outreach to potential end users as well as the development of a commercial partnership for Phase II. We had a meeting with a local algae agriculture firm, which is currently planning a large-scale expansion of their agricultural operation and has a pressing need for improved sensing to enable optimized water recycling and nutrient dosing. While we are still developing a finalized work plan, our Phase II project will likely involve an expanded operation with our long-time partner, the Fiscalini Farmstead, as well as a newly developed partnership with this second firm, in order to demonstrate the diversity of customers for whom we can deliver solutions that both improve their operations and address environmental concerns related to agricultural water use.
The perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.