Recipient Organization
UNIVERSITY OF ARIZONA
888 N EUCLID AVE
TUCSON,AZ 85719-4824
Performing Department
Biosystems Engineering
Non Technical Summary
Growth models are useful for planning, managing, and evaluating experiments. In this research, the integration of an algal growth model with cloud-connected sensors will enable use of the growth model in real-time culture management. During the planning stage, models can evaluate the effect of bioreactor types and configuration on productivity and the expected relationship between experimental reactors of full scale commercial reactors. Although models have not been directly connected to management in the past, the use of a sensor and feedback data incorporated in real-time with an algal growth model can lead to real time decision-making for amendments, fertilization, pest control, and harvesting. The Huesemann Algae Biomass Growth (HABG) model is unique in that it can accurately predict algal growth rate based on laboratory measured growth coefficients, measured reactor configurations, environmental parameters, and culture conditions. The development of the HABG model for the processes in this proposal will be an important contribution to algae cultivation. The HABG model will explain reduction in growth due to various stresses on the culture when modeled productivity falls behind observed productivity. It will also reveal the researchers when unknown stresses are impacting the culture. This research requires a polyculture model that accounts for growth of several species in a culture. A polyculture version of the HABG model has been programmed in Visual Basic for Applications. The model accounts for light absorption by several species in the culture, but it was not validated. This research will measure light attenuation as a function of depth in cultures in order to validate this component of the polyculture model. This research will connect the HABG model to an online database and develop web pages to show culture conditions and model results in real time. We have extensive experience with the programming, development, and use of the HABG model [1] and we have proven that the HABG model with relevant stress factors accurately calculates algal growth in outdoor raceway monocultures based on diurnal changes in light and temperature and laboratory calibrated growth coefficients. Salinity, shading, and nitrogen stress factors were added to the HABG model, which resulted in better agreement between observed and calculated growth rates in the ARID (Algae Raceway Integrated Design) raceway experiments during the RAFT (Regional Algal Feedstock Testbed) project [2]. Simulation of the effect of shading and nutrient stress factors on the types of raceways used in this research resulted in high correlation between biomass accumulations calculated with the HABG model and observed biomass accumulation in RAFT experiments [3]. One of the difficulties with experiments in small experimental raceways is that temperature and average light intensity are different from large commercial raceways; thus, a temperature model was developed that enables comparison of experimental and commercial raceways [4]. Connecting field sensors through inexpensive microcontrollers is a challenging and time-consuming task; however, in an agricultural experiment conducted by the Waller laboratory, flowmeter, soil, and infrared sensors have been connected to solar power and inexpensive microcontrollers in the field, which are directly connected to the cloud and the Ogden server [5]. In addition, our agricultural growth model, which is similar to the HABG model in several ways, was connected to the online database on the server. We will use the same techniques to connect the HABG model to the server. Through working with experts in IT at the University of Arizona, our agricultural database will have the capability to secure individual growers accounts and protect their data from outside parties and other account holders. The research effort will connect the HABG model and sensors to a server and database and thus enable algae culture management and evaluation with a model and feedback system. It will enable individual users to develop accounts and manage their own systems. For our research, a multi-wavelength based optical density sensor for autonomous monitoring of microalgae [6], temperature sensor, salinity sensor, and dissolved oxygen sensor will be connected to the cloud and a University of Arizona MariaDB database. Our Python version of the HABG model will also be connected to a cloud connected database. Data from field measurements and modeling will then be publicly available through Python Flask website utility. This will enable the use of what we consider as the best algae growth model to be used in real time by algae experimentalists and growers. The HABG model will be adapted to the complex polyculture proposed in this research, which has not been accomplished previously. Observed growth rate coefficients will be fed back to the laboratory team for reevaluation under a range of laboratory-controlled conditions in order to improve previously calibrated coefficients. ?
Animal Health Component
50%
Research Effort Categories
Basic
(N/A)
Applied
50%
Developmental
50%
Goals / Objectives
An algae biomass growth model and real-time sensors will be linked to an online database and a website for real time management. A polyculture growth model will be developed for co-cultures of Nannochloropsis salina, Picochlorum soloecismus, and cyanobacteria in centrate. Nutrient stress factors for N, P, and CO2 will be added to the model. The experimental reactor geometric and thermal characteristics will be used to calibrate a thermal, shading, and evaporation model. The growth model will predict growth rates for expected weather conditions and enable managers to simulate and establish optimal management and harvesting scenarios. The model will be linked to the biomass sensor to compare instantaneous growth rates to model growth rates and to neural network growth rates to detect growth abnormalities. The growth model sensors, and culture health detection system will be incorporated into an online website management system. The objectives are broken into three parts.
Project Methods
1. Algae culture sensors, biomass sensor, and polyculture HABG growth model will be connected to online database. (M1-12)The polyculture model will be reprogrammed in Python and set to run on the server. A MariaDB database will be set up on the server with access to the Python algae growth model and allowing incoming sensor data from the cloud. The database will be formatted to hold and allow access to the sensor data by the model. Algae culture and biomass sensors will be connected to inexpensive microcontrollers that will control the sensors and also have cloud or internet communication cards for cloud or internet connection.2. Calibration of polyculture HABG growth model to APID and/or other systems (M1-24)If the APID proposal is funded, the polyculture growth model will be calibrated for co-cultures of Nannochloropsis salina, Picochlorum soloecismus, and cyanobacteria in centrate. Growth coefficients for the HABG model are already available for N. salina, and P. soloecismus. Growth rate coefficients will need to be developed for cyanobacteria species. Light attenuation curves are already available for N. salina, and P. soloecismus but will need to be developed for cyanobacteria and other species in the culture. Light attenuation in polycultures will be monitored and used to calibrate light attenuation in outdoor raceways. The temperature, shading, and evaporation models will be calibrated to observed temperature, shading, and evaporation for any raceways used in this study. Nutrient stress factors for N, P, and CO2 will need to be calibrated in collaborator's laboratories and added to the model. If this is not possible, then previous nutrient growth kinetic curves that were developed for other species will be adjusted for the species in the APID system. Predator prey and other competition models will be added to the polyculture model in order to simulate interactions between species. The polyculture model will be evaluated and calibrated against observed growth rates in outdoor reactors. This will require the isolation of individual species in the co-cultures on a regular basis and daily observation of ash free dry weight biomass. 3. Development of online management tool (M25-48)Functions will be added to the polyculture HABG model that will predict optimal management and harvesting scenarios and detection of stress and less than expected growth rates in the culture. Optimization will include nutrient management through centrate addition, CO2 or other carbon sources, and harvesting. Harvesting will be based on achieving maximum biomass productivity and optimization of wastewater treatment. In order to detect less than expected growth rate, the model will be linked to the biomass sensor through the online database. Observed instantaneous growth rates will be compared to modeled growth rates and to neural network predicted growth rates to detect growth reduction below expected growth rate. The model comparisons will incorporate the known stresses such as nutrient stress and determine whether unknown stresses caused decreased growth rates. An online website management system will be constructed with the Python Flask library. Web pages will include environmental parameters, temperature and evaporation, growth model simulations and predictions, and observed culture conditions and growth rate. A simplified page will be designed with critical management information such as recommended harvest timing and centrate addition.