|Acquisition Program: ||PMS397 OHIO-Replacement Program ACAT I|
| ||RESTRICTION ON PERFORMANCE BY FOREIGN NATIONALS: This topic is “ITAR Restricted”. The information and materials provided pursuant to or resulting from this topic are restricted under the International Traffic in Arms Regulations (ITAR), 22 CFR Parts 120-130, which control the export of defense-related material and services, including the export of sensitive technical data. Foreign nationals may perform work under an award resulting from this topic only if they hold the “Permanent Resident Card”, or are designated as “Protected Individuals” as defined by 8 U.S.C. 1324b(a)(3). If a proposal for this topic contains participation by a foreign national who is not in one of the above two categories, the proposal may be rejected.|| Objective: ||The objective of this task would be to research, develop, and demonstrate a strain gage (sensor) capable of measuring axisymmetric hoop strain of a large diameter shell. The sensor must be accurate over a broad frequency range, down to and including zero Hz.
|| Description: ||Large structures acted upon by external forces exhibit complex vibrations. A complex vibration pattern consists of a superposition of many simpler vibration patterns (modes), each of which has a characteristic wavelength. Often the impact on the overall health of a structure depends more heavily on the vibration of modes with long wavelengths. For cylindrical shells, hoop strain is a key indicator of structural health. The current method of estimating the hoop strain is to place a large number of point sensors (accelerometers) around the circumference of the structure and apply geometrical weighting to the sensor signals. Many sensors are needed to filter out the shorter wavelength vibration content that can dominate the signals of the individual sensors. Each point sensor requires a cable to provide electric power and to transmit the measured signal to a centralized signal processing unit. Thus, to extract the single measure of axisymmetric hoop strain can require a significant amount of hardware and signal processing.
This proposal is looking for an innovative and cost reducing method that can measure the required strain over a long line or large area. Such a sensor must operate with minimal signal processing and low electrical power. It must operate over a range of environmental conditions (temperature, humidity, noise, and vibration). It must be capable of installation, operation, and maintenance by trained personnel. It should provide a significant signal-to-noise (SNR) improvement over current technology. Such a technology could intrinsically eliminate the need for spatial filtering, thereby radically reducing the signal processing and cabling demands and greatly reducing installation, operational, and servicing cost.
|| ||PHASE I: Develop concepts for a field of distributed sensor technology and propose a candidate set of technologies to test and evaluate in Phase II. The contractor will develop distributed hoop strain sensor concepts to address the requirements mentioned above. Criteria for assessing the technology will include accuracy, latency, linearity, ease of calibration, durability, fragility, electrical power/voltage/current requirements, and electro-magnetic interference.
|| ||PHASE II: The contractor will expand upon the Phase I work to develop a representative prototype of selected sensor concepts. The prototypes will then be demonstrated and tested under a number of operating conditions (temperature, humidity, noise and vibration level) that the government will specify.
|| ||PHASE III: The contractor will support the government in field testing the distributed strain sensor. The contractor will acquire the capability to manufacture the distributed strain sensor and the capability to provide technical support to the government in installation, operation, and maintenance of the strain sensors.
PRIVATE SECTOR COMMERCIAL POTENTIAL/|| ||DUAL-USE APPLICATIONS: The benefits from a distributed strain sensor in reducing installation and servicing cost can be adopted by the private sector for commercial purposes. Such applications include installation of distributed strain sensors on pressure tanks and in the aircraft industry.
|| References: ||
1. Fibre optic sensors in civil engineering structures, R. C. Tennyson, T. Coroy, G. Duck, G. Manuelpillai, P. Mulvihill, David JF Cooper, PW E. Smith, A. A. Mufti, and S. J. Jalali, Can. J. Civ. Eng. 27(5): 880–889 (2000)
2. Localized long gage fiber optic strain sensors, N Y Fan et al 1998 Smart Mater. Struct. 7 257
3. Structural Health Monitoring: Current Status and Perspectives, Fu-Kuo Chang (editor), CRC Press, 1998, ISBN 1566766052, 9781566766050|
|Keywords: ||Strain; prognostic; health; monitoring; sensors; gage|
Questions and Answers:
Q: The new sensor will provide SNR improvements over current technology. Can you provide a reference for the current technology for review, or provide the SNR requirement value and how SNR is defined in this case? Thanks.
A: The current technology is not relevant to the question of SNR in this case because the noise is not primarily electrical. The cylinder vibration can be decomposed into circumferential wavenumbers 0,1,2,3, etc. Circumferential wavenumber 0 is what we call the axisymmetric component. The function that the hoop strain sensor is to perform is to measure this axisymmetric component alone. So in our situation, the signal is axisymmetric component and the noise is the sum of all the other components. At any point along the circumference, the noise can be 20-30 dB higher than the signal, i.e. SNR ~ -30 dB. By using lots of point sensors distributed over the circumference and performing spatial averaging, the SNR can be raised to approximately 10 dB. We are seeking SNR at or above 20 dB.
Thank you for your interest in this SBIR.
Q: 1. How fast should the measurement to be performed? Is it a continue measurement or just one measurement?
2. By "Line-Distributed", is it along the cylindrical shell? How long is the shell and how many data points are expected?
3. What is the estimated amplitude range of the strain to be measured? what is the Young's modulus of the shell? what is the expected frequency range?
4. Is there any limitations in accessing the shell, e.g. if it is under ground, etc.?
5. Do you have any references about the status of current point sensors approach?
A: 1. Measurement latency is one of several criteria listed by which we will compare the proposed solutions. We are not specifying its value. If all else were equal between two solutions, the one with the lowest latency would be deemed superior.
2. The focus is measuring axisymmetric hoop strain, which is a circumferential measure as opposed to a longitudinal measure. Conceivably there will be several locations along the longitudinal axis of the cylinder at which simultaneous measurements of this hoop strain are desired but the focus of this SBIR is the capability to measure at a single location.
3. The maximum static strain is 5% (compression). The maximum dynamic strain is 3%. You can assume steel properties for the shell, though a solution that is constrained to work only for a small range of Young's Modulus would not be considered superior. The length of the shell is not considered relevant but if you can identify a reason why that assumption should be reconsidered, please do so. It is unclear what you mean by 'how many data points are expected?'. If you mean how many points around the circumference then that implies a spatial sampling method which we have found unsatisfactory in our current practice. We are not ruling out sampling however. If your method can preclude the aliasing of higher circumferential wavenumber motion into the estimate of the axisymmetric (zeroth circumferential wavenumber) hoop strain, with lower cabling and signal processing requirements than other solutions, then that would be viewed favorably.
4. We are interested both in in-air applications as well as in-seawater applications but not in in-ground applications. For the in-seawater application, direct access would be provided for installation but the solution must be capable of operating reliably without maintenance via direct access for a year or more.
5. Data for the current point sensors approach is not releasable but the problems with it are primarily not with the sensor technology itself but with the high number of point sensors required to reject high circumferential wavenumber motion aliasing into the axisymmetric measure, and all the cabling and signal processing channels associated with that application.
Q: The solicitation states "The sensor must be accurate over a broad frequency range, down to and including zero Hz." The Question pertains to what specifically is of interest at zero Hz (DC). Consider the following.
1. The most rigorous interpretation would mean you are looking for an "ABSOLUTE" sensor. This would mean that the readout of the sensor (for a cylinder) would be a precise measure of the circumference in addition to all dynamic variations. A rough guess for instantaneous dynamic range here would be >= 120 dB (million to 1).
2. A second interpretation would be that of a "RELATIVE" sensor which is DC coupled. Here the sensor output would not contain the circumference, but rather all variations from the circumference.
QUESTION: Which interpretation is applicable?
A: At zero Hz (DC), we are interested in measuring the change of circumference from some previous stored value. In some test situations, the cylinder will be pre-stressed and then be loaded under increasing external (or decreasing internal) hydrostatic pressure so the previous stored value would be the static strain corresponding to the initial state of pre-stress. We are not requiring an estimate of the absolute circumference as a separate output signal.
Q: In regards to installation of the sensor, can we assume it can be installed either on the outside or inside surface of the cylinder? If either is applicable, which is preferred?
A: We are not mandating which of the two surfaces should host the sensor. We will be weighing ease of installation and maintainability of each solution against those of other solutions.
Q: 1. What is the diameter of the shell (or range of diameters)?
2. Is there information on materials used for the shell? If metallic, what alloys are used? If composite, what are materials and method of fabrication?
A: 1. The shell diameter range is from 1 to 20 meters
2. For shell material you can assume steel. However, if two competing solutions are equal in all other respects but one can be applied only to steel while the other can be applied to steel and other materials, then the second solution would be considered superior.
Q: Please provide the operating conditions/specifications under which the device should work – the range in temperature, relative humidity, pressure. And the conductivity of the medium in which the device will operate, including its corrosiveness.
A: Operating environment: Ranges from in-air to in-seawater
Temperature range: 0+ to 80 degrees Celsius
Static pressure: Ranges from 100 kPa to 100 MPa
Conductivity properties: Of air or seawater (should be capable of performing in both environments)
Corrosiveness: The surface to which the device would be attached would generally be clean but may have light amounts of rust.
Q: I understand that the maximum static and dynamic strain to be measured is 5%/3%. What is the sensitivity requirement, that is, what is the minimum strain change that is required to be measured?
A: We are not specifying the minimum strain as a requirement. Minimum strain sensitivity will be one (of several) criteria for comparing alternative solutiolns.
Q: In the answer to a previous question #2 for this topic, it is stated that the maximum static strain is 5% and maximum dynamic is 3%. For most engineering materials such as the aforementioned steel, the elastic limit is significantly less than 1% strain whether tensile or compressive. Might this be a typo or is the requirement genuinely for 5% strain capability? Thank you.
A: It is not a typographical error.
Q: How often is the measurement required to be made?
A: The Navy has a range of applications with different needs for health monitoring. Some applications are solely temporary for testing. Other applications call for permanent installation and continuous operation. The proposed solution should be capable of both temporary installation (and therefore de-installation after a time) and permanent installation. The proposed solution should be capable of providing continuous monitoring.
Q: 1. I understand that the monitoring should be continuous. What is the required bandwidth in time domain?
2. Is the target signal slow moving, or does it have fast moving component (in time domain) that needs to be captured? How fast?
A: 1. Bandwidth: 0-1000 Hz
2. Latency: We are not specifying the requirement for latency. Latency is one of several criteria by which alternative proposed solutions will be compared.
Q: What is the required frequency range?
A: Frequency range: 0-1000 Hz
Note: Please read carefully the existing Q&A. You will find this information there already, along with other information you might find helpful.