Search NASA⌕ Search

SEARCH · Search NASA

Results for “component testing”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,153 records · Page 64

Finite element analysis of compliant cladding and base metal systems

This report describes a roadmap towards a design method for high temperature nuclear cladded components that does not require long term testing of clad materials. Molten Salt Reactors (MSRs) will operate using corrosive fuel or coolant salts. The existing Class A metallic materials qualified for ASME Boiler and Pressure Vessel Code Section III, Division 5 design are not optimal for corrosion resistance when exposed to MSR salts. Corrosion could then severely limit the service life of MSRs. One option to alleviate this could be to Code qualify new corrosion resistant materials. However, this is a lengthy and expensive process and MSR developers have a near term need for a design solution that enables reactor designs with long service lives. An alternative would be to use cladded components – overlay some existing Class A material with a thin layer of some corrosion resistant material. This report describes the outlines of a design strategy that would enable this approach. The strategy relies on approximate design analysis methods for certain types of clad/base systems that avoids requiring long-term mechanical properties for the clad materials. This would allow the use of cladded components in the near term, without requiring long testing programs. However, it will limit the types of clad materials that can be used. This report describes the design framework, demonstrates the effectiveness of the approximate analysis methods, and confirms the approach using full scale finite element analysis of prototypical cladded high temperature components. It also describes a roadmap towards a complete design method and surveys available clad materials and bonding methods for suitably, considering only mechanical and not environmental effects.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Radiation Hardened Electronics Destined For Severe Nuclear Reactor Environments

Post nuclear accident conditions represent a harsh environment for electronics. The full station blackout experience at Fukushima shows the necessity for emergency sensing capabilities in a radiation-enhanced environment. This NEET (Nuclear Energy Enabling Technologies) research project developed radiation hardened by design (RHBD) electronics using commercially available technology that employs commercial off-the-shelf (COTS) devices and present generation circuit fabrication techniques to improve the total ionizing dose (TID) hardness of electronics. Such technology not only has applicability to severe accident conditions but also to facilities throughout the nuclear fuel cycle in which radiation tolerance is required. For example, with TID tolerance to megarads of dose, electronics could be deployed for long-term monitoring, inspection and decontamination missions. The present work has taken a two-pronged approach, specifically, development of both board and application-specific integrated circuit (ASIC) level RHBD techniques. The former path has focused on TID testing of representative microcontroller ICs with embedded flash (eFlash) memory, as well as standalone flash devices that utilize the same fabrication technologies. The standalone flash devices are less complicated, allowing better understanding of the TID response of the crucial circuits. Our TID experiments utilize biased components that are in-situ tested, and in full operation during irradiation. A potential pitfall in the qualification of memory circuits is the lack of rigorous testing of the possible memory states. For this reason, we employ test patterns that include all ones, all zeros, a checkerboard of zeros and ones, an inverse checkerboard, and random data. With experimental evidence of improved radiation response for unbiased versus biased conditions, a demonstration-level board using the COTS devices was constructed. Through a combination of redundancy and power gating, the demonstration board exhibits radiation resilience to over 200 krad. Furthermore, our ASIC microprocessor using RHBD techniques was shown to be fully functional after an exposure of 2.5 Mrad whereas the COTS microcontroller units failed catastrophically at <100 krad. The methods developed in this work can facilitate the long-term viability of radiation-hard robotic systems, thereby avoiding obsolescence issues. As a case in point, the nuclear industry with its low purchasing power does not drive the semiconductor industry strategic plans, and the rapid advancements in electronics technology can leave legacy systems stranded.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Development of a Robust Reference Electrode in Aggressive Chemical and Radiation Environments in the Hanford Waste Tanks

The Hanford site stores more than 200 million liters of radioactive and chemically hazardous wastes from the production of weapons materials. The wastes are stored in 177 underground carbon-steel storage tanks, separated between 149 single shell tanks (SSTs) and 28 double shell tanks (DSTs). The DSTs provide critical retrieval and interim storage before the waste is vitrified in the Waste Treatment and Isolation Plant (WTP). The tanks will need to remain in-service far beyond the initial 40-year design life, and effective corrosion control practices must remain in force to extend the tanks’ lifespans. This effort includes direct measurements of corrosion rate (e.g., ultrasonic measurements and corrosion coupons) and electrochemical processes (e.g., linear polarization measurements and open circuit potential measurements). The Hanford site began monitoring the corrosion potential in select DSTs in 2008. Of the 45 reference electrodes that have been installed, 29 have failed and 6 others provided unreliable results. DOE-EM is supporting a 3-year program to develop a chemical and radiation resistant reference electrode for application in the Hanford tanks. The first year of the program focused on understanding the failure mechanism for the reference electrodes and identification of candidate construction materials that would mitigate degradation of the electrodes in the waste environment. During the second year of the program, the objectives were to: 1) test candidate materials under simulated waste conditions, 2) design components that will extend the service life of the electrode, 3) fabricate materials for prototype reference electrodes, and 4) assemble prototype reference electrodes for accelerated testing. The reference electrode is constructed of four principal parts: 1) junction, 2) casing, 3) inner chamber backfill materials, and 4) the sensing wire. Principally, improvements of the junction, casing, and inner chamber backfill materials are being pursued. The junction material at the interface between the waste and the inner chamber of the reference electrode was identified as a critical component in the failure of the reference electrodes. Nine candidate replacement junction materials were tested under simulated waste conditions to evaluate permeation rate. These materials included a variety of polymeric and ceramic materials, some of which were 3-D printed. Thus far, porous polyvinylidene fluoride materials have performed satisfactorily and are being considered for prototype development. The commercial electrode casing materials in general have performed well. Additionally, 3-D printing of chemically and mechanically stable materials is being investigated as a means for further improvement in fabrication consistency. SRNL has also investigated altering the reference electrode design to extend the service life. The new design of the interior of the reference electrode casing creates a longer, more tortuous path between the junction material and the electrode sensing wire. A finite element model was used to optimize the design without adversely impacting the circuit resistance of the electrode during the measurements, thus preserving the measurement accuracy while enhancing the service life. The inner chamber back fill materials are also critical to the performance of the reference electrode. Materials that are resistant to intruding tank waste and provide a conductive path to the sensing wire were investigated. Gel and powder materials that are interspersed with a conductive chloride bearing material were tested for their influence on diffusion and electrode resistance. All the investigated materials and components will be assembled, with collaboration from commercial vendors, to fabricate the initial prototypes. Accelerated testing of the prototypes will be initiated in Year 2 of the program and will be completed in Year 3. A recommendation on the materials of construction and the design of the new robust reference electrode will be presented to the Hanford tank farm facility.

Sykes, Kiana [Savannah River National Laboratory (↗

Combined-Accelerated Stress Testing of Photovoltaic Materials

By applying multiple environmental stresses in fieldrepresentative combinations and sequences, combined-accelerated stress testing (CAST) identifies degradation modes and failure mechanisms of photovoltaic (PV) modules and components that are missed by single stress factor accelerated testing.

14 SOLAR ENERGY↗

Strain gauge for testing microreactor hexagonal core blocks in the Single Primary Heat Extraction and Removal Emulator

To support the development and deployment of microreactor technologies, experiments that help verify and validate reactor systems and components are performed at non-nuclear test facilities. The Single Primary Heat Extraction Removal Emulator at Idaho National Laboratory is one of these test facilities and was used to monitor a test article that has a prototypic geometry of a heat-pipe cooled microreactor core block. To collect crucial temperature and strain data during testing, temperature and strain sensing fiber optics were embedded to the surface of the test article using an ultrasonic additive manufacturing technique. To support and provide benchmark strain data for the embedded sensors, a commercial resistive strain gauge was attached. This report will discuss the results from the deployment of the commercial strain gauge which includes the setup/attachment strategies, data acquisition, and analysis of the strain data.

36 MATERIALS SCIENCE↗

Status Update on the Development of Transducers and Bonding Techniques for Enabling Acoustic Measurements of Damage in Microreactor Components

This report provides an overview of potential sensors and sensor-bonding techniques to enable the online acoustic interrogation of microreactor components and enhance structural health monitoring capabilities. The report focuses primarily on optical fiber–based acoustic sensors and describes initial experimental progress toward the deployment of these sensors for microreactor applications. The general approach is to monitor the resonant frequencies of microreactor components and search for evidence of structural defects that could indicate imminent failure. If properly identified, then the components could be repaired during the next reactor outage to prevent costly unplanned shutdowns. The ability to monitor the structural health of components could also reduce the need for time-consuming visual inspections and reduce staffing to improve microreactor economic viability. Increased sensor density is also one of the first steps to moving toward eventual semiautonomous operation. The expected microreactor conditions in which acoustic sensors must survive are characterized, including temperatures, neutron fluences, thermomechanical strains, and vibrational frequencies. Optical fiber–based acoustic sensors are identified as an attractive candidate for acoustic monitoring because of their high accuracy, immunity to electromagnetic interference, and resiliency in high-temperature, high-radiation environments. Optical fiber–based intrinsic sensors, such as type-II fiber-Bragg gratings and Fabry-Pérot Cavities (FPCs), are particularly attractive for a microreactor environment because of their high temperature stability, and FPCs also enable higher frequency interrogation with a lower sensitivity to radiation-induced drift. This report describes multiple interrogation systems, but the best interrogation system for a given situation will depend on the specific microreactor application, including the desired acoustic vibrational amplitudes, modes, and resonant frequencies. Initial experiments included fabricating three FPCs, tack-welding these FPCs to stainless-steel pipes or rods and performing room-temperature acoustic sensing tests to capture the vibrational frequency content. Peaks were identified in the measured frequency spectra and compared with the theoretical fundamental frequencies obtained from Euler-Bernoulli beam theory. Two of the three FPCs measured vibrational frequencies that generally matched those obtained from theory. Future work will include similar testing on pipes or other microreactor components with intentional flaws to evaluate the ability to determine changes in resonant frequencies. Finally, these tests will be repeated at high temperatures, potentially with an applied thermomechanical stress, to include environmental conditions similar to those for a microreactor application.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

First Results of AUP $Nb_{3}Sn$ Quadrupole Horizontal Tests

The Large Hadron Collider will soon undergo an upgrade to increase its luminosity by a factor of ~10. A crucial part of this upgrade will be replacement of the NbTi final focus magnets with Nb3Sn magnets that achieve a ~50% increase in the field strength. This will be the first ever large-scale implementation of Nb3Sn magnets in a particle accelerator. This talk will present the program to fabricate these components and first results from horizontal tests of fully assembled cryo assemblies.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Portable self-contained pressure testing manifold and associated methods

A portable, case-enclosed pressure test manifold, configured to support both pneumatic and hydrostatic pressure testing, and comprising a downstream fluid path (e.g., a system test gauge, a ball valve, a system test port, a system relief valve, and other components/piping) and an upstream fluid path (e.g., a low-pressure gauge, a high-pressure gauge, and other components/piping). A regulator and/or the ball valve selectively operate the downstream fluid path, defining a hydrostatic pressure testing path, to receive a liquid; and the upstream fluid path in combination with the downstream fluid path, defining a pneumatic pressure testing path, to receive a gas. Manifold components are pressure rated based on use and are ASME Code B31.3 compliant. A flexible hose connects with either of the pneumatic pressure testing path and the hydrostatic pressure testing path of the manifold. The case includes parking receptables and a carrying handle configured to stow the hose during transport.

Milholland, Roger A.↗

Nuclear code case development of printed-circuit heat exchangers with thermal and mechanical performance testing

The U.S. Department of Energy has agreed to fund a three-year integrated research project to close technical gaps involved with compact heat exchangers to be used in nuclear applications. This paper introduces the goals of the project, the research institutions, and industrial partners working in collaboration to develop a draft Boiler and Pressure Vessel Code Case for this technology. Heat exchanger testing, as well as non-destructive and destructive evaluation, will be performed by researchers across the country to understand the performance of compact heat exchangers. Testing will be performed using coolants and conditions proposed for Gen IV Reactor designs. Preliminary observations of the mechanical failure mechanisms of the heat exchangers using destructive and non-destructive methods is presented. Unit-cell finite element models assembled to help predict the mechanical behavior of these high-temperature components are discussed as well. Performance testing methodology is laid out in this paper along with preliminary modeling results, an introduction to x-ray and neutron inspection techniques, and results from a recent pressurization test of a printed-circuit heat exchanger. The operational and quality assurance knowledge gained from these models and validation tests will be useful to developers of supercritical CO 2 systems, which commonly employ printed-circuit heat exchangers.

42 ENGINEERING↗

Evaluation of Fabric Membranes for Use in Saltstone Drain Water System

Saltstone Disposal Unit 2 contains a sheet drain fabric intended to separate solids from drain water to be returned to the Salt Feed Tank. A similar system installed in Vault 4 appears to be ineffective in keeping solids out of the drain water return lines. Waste Solidification Engineering is considering installation of an additional fabric membrane to supplement the existing sheet drain in SDU 2. Amerdrain 200 is the product currently installed in SDU 2. This product is no longer available, so Sitedrain 94 was used as the replacement product in this testing. Fabrics with apparent opening sizes of 10, 25, 50 and 100 microns were evaluated. These fabrics were evaluated under three separate test conditions, a water flow test, a solids retention test and a grout pour test. A flow test with water showed that installation of an additional filter layer will predictably reduce the theoretical flux through the sheet drain. The manufacturer reports the flux for Sitedrain 94 as 150 gpm/ft 2 by ASTM D-4491. This compares reasonably well with the 117 gpm/ft 2 obtained in this testing. A combination of the 10 micron fabric with Sitedrain 94 could be expected to decrease flux by about 10 times as compared to Sitedrain 94 alone. The different media were used to filter a slag and fly ash mixture from water. Slag historically has the smallest nominal particle size of the premix components. Cement was omitted from the test because of its reactivity with water would prohibit accurately particle size measurements of the filtered samples. All four media sizes were able to remove greater than 95% of particles larger than 100 microns from the slurry. The smaller opening sizes were increasingly effective in removing more particles. The 10 micron filter captured 15% of the total amount of solids used in the test. This result implies that some insoluble particles may still be able to enter the drain water collection system, although the overall solids rejection is significantly improved over the current design. Test boxes were filled with grout to evaluate the performance of the sheet drain and fabrics in a simulated vault environment. All of the tests produced a similar amount of drain water, between 8-11% of the amount of water in the mix, which is expected with the targeted formulation. All of the collected drain waters contained some amount of solids, although the 10 micron filter did not appear to allow any premix materials to pass through. The solids collected from this box are believed to consist of calcium carbonate based on one ICP-AES measurement. Any of the four candidate fabrics would be an improvement over the sheet drain alone relative to solids removal. The 10 micron fabric is the only candidate that stopped all premix material from passing. The 10 micron fabric will also cause the largest decrease in flux. This decrease in flux was not enough to inhibit the total amount of drain water removed, but may lead to increased time to remove standing water prior to subsequent pours in the facility. The acceptability of reduced liquid flux through the 10 micron fabric will depend on the amount of excess water to be removed, the time available for water removal and the total area of fabric installed at the disposal cell.

42 ENGINEERING↗

In-Reactor Oxidation of Zircaloy-4 Under Low Water Vapor Pressures

Complementary in- and ex-reactor oxidation tests have been performed to evaluate the oxidation and hydrogen absorption performance of Zircaloy-4 (Zr-4) under relatively low partial pressures (300 and 1000 Pa) of water vapor at specified test temperatures (330° and 370°C). Data from these tests will be used to support fabrication of components intended for isotope-producing targets and provide information regarding the temperature and pressure dependence of oxidation and hydrogen absorption of Zr-4 over the specified range of test conditions. Comparisons between in- and ex- reactor test results were performed to evaluate the influence of irradiation.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

HYTEST Phase I Facility Commissioning and Modeling

The purpose of this document is to report the first year accomplishments of two coordinated Laboratory Directed Research and Development (LDRD) projects that utilize a hybrid energy testing laboratory that couples various reactors to investigate system reactance behavior. This work is the first phase of a series of hybrid energy research and testing stations - referred to hereafter as HYTEST facilities – that are planned for construction and operation at the Idaho National Laboratory (INL). A HYTEST Phase I facility was set up and commissioned in Bay 9 of the Bonneville County Technology Center (BCTC). The purpose of this facility is to utilize the hydrogen and oxygen that is produced by the High Temperature Steam Electrolysis test reactors operating in Bay 9 to support the investigation of kinetic phenomena and transient response of integrated reactor components. This facility provides a convenient scale for conducting scoping tests of new reaction concepts, materials performance, new instruments, and real-time data collection and manipulation for advance process controls. An enclosed reactor module was assembled and connected to a new ventilation system equipped with a variable-speed exhaust blower to mitigate hazardous gas exposures, as well as contract with hot surfaces. The module was equipped with a hydrogen gas pump and receiver tank to supply high quality hydrogen to chemical reactors located in the hood.

08 HYDROGEN↗

CRADA: Improved H2/Br2 Flow Batteries

This CRADA is established to jointly develop H2/Br2 flow batteries with improved materials sets and understanding between TVN Systems and LBNL. The focus of the activities is the testing and optimization of flow battery components to meet efficiency and power requirements. The proposed work is essentially all experimental where TVN Systems will use their knowledge and expertise in membrane-electrode-assembly (MEA) design and fabrication using novel catalyst and membrane materials and LBNL will use their existing knowledge and expertise in flow-battery cell assembly and testing for performance and durability.

25 ENERGY STORAGE↗

Structural integrity of additively manufactured stainless steel with cold sprayed barrier coating under combined cyclic loading

Integration of metal additive manufacturing (AM) and cold spray (CS) technologies provide an unprecedented opportunity to manufacture coated material systems with complex geometrical features. The application of these material systems in functionally critical components requires adequate structural integrity, particularly in the presence of cyclic loading. This article researches the multiaxial fatigue (axial-torsional cyclic loading) behavior of a coated material system consisting of 15Cr-5Ni precipitation-hardening stainless steel (15-5 PH SS) substrate additively manufactured by direct metal laser sintering with a layer of chromium carbide nickel (CrC-Ni) barrier coating deposited by CS. The influence of AM and CS-induced residual stresses on fatigue performance of test specimens was thoroughly studied. Additionally, the effect of surface roughness and processes induced defects were considered to explain the crack growth mechanism. Stresses assessed by synchrotron X-ray diffraction indicated a substantial accumulation of residual stresses, particularly in the outer surface of the as-fabricated 155 PH SS specimens. The state of residual stress was changed notably following the deposition of CrC-Ni coating in the axial, hoop, and radial directions of the fatigue test specimen. Also, CS deposition of CrC-Ni coating caused significant improvement in the surface quality of the additively manufactured components. Fatigue test results indicated, CS deposition of CrC-Ni substantially enhances the fatigue life of the AM-produced 15-5 PH SS substrate in all loading conditions, particularly in the high cycle fatigue regime. The improvement in the fatigue life of the specimens with coating was associated with a reduction in equivalent residual stress at the substrate surface and improvement in the specimens' surface condition (i.e., reduced surface roughness). The fractographic analysis of the specimen indicated the cracks tend to initiate in the surface of both as-fabricated and cold-sprayed specimens. However, the mechanism of crack growth changed notably following the deposition of CrC-Ni coating. The cracks tended to propagate in the planes parallel or with a small deviation from the build layers of the AM-produced specimens. On the other hand, deposition of CrC-Ni coating increased the deviation of crack growth plane from the build layers of the substrate.

36 MATERIALS SCIENCE↗

Data Summary Report Small Scale Melter Testing of HLW Algorithm Glasses: Matrix 1 Tests

Eight tests using different HLW feeds were conducted on the DM100-BL to determine the effect of variations in glass properties and feed composition on processing rates and melter conditions (off-gas characteristics, glass processing, foaming, cold cap, etc.) at constant bubbling rate. In over seven hundred hours of testing, the property extremes of glass viscosity, electrical conductivity, and T 1% , as well as minimum and maximum concentrations of several major and minor glass components were evaluated using glass compositions that have been tested previously at the crucible scale. Other parameters evaluated with respect to glass processing properties were +/-15% batching errors in the addition of glass forming chemicals (GFCs) to the feed, and variation in the sources of boron and sodium used in the GFCs. Tests evaluating batching errors and GFC source employed variations on the HLW98-86 formulation (a glass composition formulated for HLW C-106/AY-102 waste and processed in several previous melter tests) in order to best isolate the effect of each test variable. These tests are outlined in a Test Plan that was prepared in response to the Test Specification for this work. The present report provides summary level data for all of the tests in the first test matrix (Matrix 1) in the Test Plan. Summary results from the remaining tests, investigating minimum and maximum concentrations of major and minor glass components employing variations on the HLW98-86 formulation and glasses generated by the HLW glass formulation algorithm, will be reported separately after those tests are completed.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Development of New Antifoam for High Level Waste Processing in DWPF

During processing in the Defense Waste Processing Facility (DWPF), the evaporation of sludge leads to large amounts of foaming. This foaming can cause a foam-over, which contaminates other parts of the facility, most importantly the condensate. Currently, the antifoam being used is not sufficient at more extreme pH values and relatively high concentrations are required to minimize foaming. This means that a foam-over is more likely in parts of the process where the pH moves away from the neutral point and antifoam is needed in large quantities to prevent such occurrences. Adding this antifoam can also double processing time. The current antifoam also degrades into flammable byproducts, which creates a hazard in the processing. There are two main ideas proposed in order to solve the problems with the current antifoam: A new antifoam; A mechanical solution (sprays, agitator, etc). The focus of this summer was to develop a new antifoam that works across a broader range of pH values, can be added in lower concentrations, and does not degrade into flammable byproducts. This solution was pursued by obtaining different superspreaders and antifoams from companies across the country in order to perform a series of tests on them; these tests include superspreader testing, Foamscan testing, boiling testing, and Sludge Receipt and Adjustment Tank (SRAT) and Slurry Mix Evaporator (SME) process testing. To begin testing, four different superspreaders were tested on their ability to spread. The four superspreaders include: Y-17112, Y-17309, Antifoam 747 (the old antifoam), and Silwet L77 (the main component in Antifoam 747). These superspreaders were tested across a range of pH values and concentrations from 100 ppm to 5000 ppm. After completion of the tests, the results indicated that Y-17112 spread across a wider pH range than the others, even at lower concentrations. This is ideal for developing a new antifoam, as the pH throughout processing in DWPF varies and the lower the necessary concentration, the less antifoam that must be added per cycle. In order to test whether or not the superspreaders work as an antifoam, they have to be tested on real foam. In order to test this, a foamy simulant similar to the foaminess of the sludge in DWPF needed to be developed. This required testing many different sludges and designing our own simulant with different concentrations of insoluble solids. Eventually, a simulant similar enough to what is seen in DWPF was identified. Each superspreader will be tested on the simulant at different pH values and at different concentrations. The simulant will be allowed to foam and small amounts of the superspreader will be added while the Foamscan software tracks the height of the foam column; this allows us to know if the superspreader decreased the foam or not. Next, boiling testing in the Foamscan will be completed. This is testing identical to the Foamscan testing at room temperature, only at boiling conditions similar to DWPF. After boiling testing, the superspreader that is deemed most successful will be tested in a real, scaled-down SRAT/SME cycle. This will ensure it works with all the real chemicals that are added in DWPF and that it will remain successful in all the conditions that it may face. After the SRAT/SME testing, a decision will be made on whether or not the superspreader is successful and it may or may not be chosen as the new antifoam for DWPF.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Task allocation in geo-distributed cyber-physical systems - 256

This paper studies the task allocation algorithm for a distributed test facility (DTF), which aims to assemble geo-distributed cyber (software) and physical (hardware in the loop) components into a prototype cyber-physical system (CPS). This allows low cost testing on an early conceptual prototype (ECP) of the ultimate CPS (UCPS) to be developed. The DTF provides an instrumentation interface for carrying out reliability experiments remotely such as fault propagation analysis and in-situ testing of hardware and software components in a simulated environment. Unfortunately, the geo-distribution introduces an overhead that is not inherent to the UCPS, i.e. a significant time delay in communication that threatens the stability of the ECP and is not an appropriate representation of the behavior of the UCPS. This can be mitigated by implementing a task allocation algorithm to find a suitable configuration and assign the software components to appropriate computational locations, dynamically. This would allow the ECP to operate more efficiently with less probability of being unstable due to the delays introduced by geo-distribution. The task allocation algorithm proposed in this work uses a Monte Carlo approach along with Dynamic Programming to identify the optimal network configuration to keep the time delays to a minimum. (authors)

97 MATHEMATICS AND COMPUTING↗

SU-E-J-256: Predicting Metastasis-Free Survival of Rectal Cancer Patients Treated with Neoadjuvant Chemo-Radiotherapy by Data-Mining of CT Texture Features of Primary Lesions

Purpose: The purpose of this study is to investigate the relationship between computed tomographic (CT) texture features of primary lesions and metastasis-free survival for rectal cancer patients; and to develop a datamining prediction model using texture features. Methods: A total of 220 rectal cancer patients treated with neoadjuvant chemo-radiotherapy (CRT) were enrolled in this study. All patients underwent CT scans before CRT. The primary lesions on the CT images were delineated by two experienced oncologists. The CT images were filtered by Laplacian of Gaussian (LoG) filters with different filter values (1.0–2.5: from fine to coarse). Both filtered and unfiltered images were analyzed using Gray-level Co-occurrence Matrix (GLCM) texture analysis with different directions (transversal, sagittal, and coronal). Totally, 270 texture features with different species, directions and filter values were extracted. Texture features were examined with Student’s t-test for selecting predictive features. Principal Component Analysis (PCA) was performed upon the selected features to reduce the feature collinearity. Artificial neural network (ANN) and logistic regression were applied to establish metastasis prediction models. Results: Forty-six of 220 patients developed metastasis with a follow-up time of more than 2 years. Sixtyseven texture features were significantly different in t-test (p<0.05) between patients with and without metastasis, and 12 of them were extremely significant (p<0.001). The Area-under-the-curve (AUC) of ANN was 0.72, and the concordance index (CI) of logistic regression was 0.71. The predictability of ANN was slightly better than logistic regression. Conclusion: CT texture features of primary lesions are related to metastasisfree survival of rectal cancer patients. Both ANN and logistic regression based models can be developed for prediction.

60 APPLIED LIFE SCIENCES↗