|
|
 |
|
|
Technical Data Analysis in the Aerospace Industry
Challenges in Aerospace and How the PV-WAVE Family of Products Delivers Solutions
A Business White Paper by Visual Numerics, Inc. - April 2002
Introduction
The aerospace industry includes companies involved in airframe manufacture,
space systems, satellite systems, and contractors or Independent Software Vendors for these industries. The types of work where technical data analysis is required include flight test, avionics, structural dynamics, defect analysis, aerodynamics, ground control or ground science, space systems, and various research and development areas. There are common requirements for technical data analysis tools and some challenges specific to this industry that call for solutions that can address these challenges and deliver on productivity, meet delivery schedules, and provide a positive return on investment. The PV-WAVE family of products offers solutions that are particularly suited to the challenges in aerospace.
Technical Data Analysis Requirements in Aerospace
Aerospace Data
 In the aerospace industry, there are many situations where a tool is needed to read, transform, analyze and chart data, either in an exploratory post-processing mode or a real-time data collection mode. This data is very often time-series data, that is to say multiple channels of data collected at fixed or irregular time steps. Flight test data and satellite telemetry data are two common examples. The data may also be multidimensional X-Y data that does not have a time component. There are many kinds of data that have an independent (X axis) and dependant (Y axis) relationship. Data of this form is often stored in a spreadsheet format.
 Another common data type is image data or other data collected in two dimensions that can be viewed and analyzed as images. Remote sensed data and data used in various kinds of structural analysis can take this form. The sources of the data can come from many different kinds of instrumentation including: data loggers for flight test instrumentation or various materials analysis systems, satellite control systems, and satellite science systems (e.g. remote
sense data).
The data is sometimes collected and analyzed in real time, such as in satellite control systems, but more often saved and analyzed in a post-processing environment. The form of this saved data can be ASCII (textual) data, binary data, data from commercial databases, and data exported from desktop tools like Microsoft Excel™. Image data may be stored in a common image format like TIFF or JPEG, but it is more likely that the format is either company or project specific, or proprietary to the data collection system being used. Data sets are often very large and can range from a few megabytes to a few gigabytes.
Data Manipulation, Filtering and Analysis
In technical data analysis in the aerospace industry, it is common to extract subsets of data from large multidimensional datasets for analysis and visualization. This may mean selecting only certain columns of data or data between certain time stamps. Data may be further transformed to correct for instrument base readings or gain adjustments. For image data geometric
transformations (e.g. rotation, translation) are often required.
In the case of image data, further analysis may take the form of image processing routines such as histogram equalization, image enhancement, classification and segmentation. For time series or X-Y data the analysis may include smoothing, curve fitting, and various advanced filter designs such as finite impulse response (FIR) a nd infinite impulse response (IIR) filters.
Data Visualization
 The most common graphic display in aerospace applications is a simple time history plot with a date/time lower axes and one or more vertical axes representing time dependant data. More general two-dimensional X-Y charts are also used where the X axes is not time dependant. Data which is represented as a two dimensional array is usually viewed as images, contour plots, or 3D surface plots. Various other chart types such as histograms, polar plots, or high-end 3D visualizations are used
on occasion. Data overlaid on cartographic maps is sometimes used in remote sensing applications.
Data visualization refers to the ability to explore data in an interactive fashion to gain insight that is difficult to obtain from the raw numbers or even from simple static charts. Techniques such as interactive zooming, profile plotting of image data, and novel ways of representing large amounts of multidimensional data in a single view are some ways that data visualization can be used to explore data. These techniques are useful in aerospace applications for data screening and
validation.
Challenges in Technical Data Analysis in Aerospace
There are a number of requirements for data analysis software that are common
across many different areas of the aerospace industry. These requirements
usually make it difficult or impossible to use single -purpose off-the-shelf
applications to meet these specific needs. Some of these aerospace specific
requirements include:
- The need for different kinds of tools depending on project requirements
and the intended audience: interactive desktop tools, tools to build
applications for non-programmer end-users, and web based tools.
- The ability to read a large number of often proprietary data formats
- The ability to read, manipulate and chart very large datasets
- Specific numerical analysis functionality especially in signal and image
processing
- The need to access legacy code in C, C++ or FORTRAN
- Hardcopy output and batch report generation
- The need to re-use tools, code and expertise from project to project to
control costs
- The need to rapidly develop custom solutions
- Multi-platform support including Windows 98/NT/2000 and UNIX/Linux
The Solution - the PV-WAVE Family of Products
The PV-WAVE family of products has been used in the aerospace industry for
over a decade. The breadth of functionality and methods of delivering this functionality have changed over this time as new technologies have been introduced, and with a growing demand for off-the-shelf solutions. We want to introduce you to some of the ways that customers use PV-WAVE in aerospace, the functionality it provides, and offer some specific case studies.
The PV-WAVE family includes the PV-WAVE programming language, the TSWAVE application, an off-the-shelf solution for time-series data analysis, and JWAVE, a way to web-enable applications developed using the PV-WAVE language.
While not discussed in this paper, much of the underlying numeric technology in PV-WAVE comes from the IMSL numeric libraries, industry standard numeric libraries that are also available from Visual Numerics as callable libraries in C,
C++, FORTRAN and JAVA. Reasons for using these libraries include a need for creating OEM applications and for applications in high performance computing environments. In this paper we will just consider the IMSL libraries as they are integrated into PV-WAVE.
Architecture for a Solution
The PV-WAVE Family provides a number of tools and ways of using the tools to
provide many different kinds of solutions that all have a common language underneath. This maximizes your productivity by reusing your training and code for different requirements, rather than trying to use a variety of tools requiring more training, greater difficulty in maintaining code, and problems in integrating the applications and data. Here are some common ways of using the PV-WAVE family of products:
An Interactive Desktop Solution
You can use PV-WAVE from the command prompt to explore data and try out new ideas. PV-WAVE can be used interactively for ad hoc analysis of data and to test out small code fragments while developing code. Commands are interpreted as you type them and there is no compilation of code required.
A Development Environment
Create simple to complex applications including user interfaces for your own use or for other end-users. PV-WAVE is a full-featured language well suited to creating applications, and is often used to replace applications developed in C which are much more costly to develop. PV-WAVE applications, even those with user interfaces, have the added advantage that they can be moved between platforms and run without recompiling.
An Off-the-Shelf Solution
TS-WAVE provides an off-the-shelf extensible application for analyzing and plotting many kinds of time series data. It allows you to read, transform and plot data in time history plots, tabular data views, X-Y plots, and contour plots. Through a point and click interface you can customize the plots extensively by modifying many plot attributes and adding annotation. It has built-in report generation features and allows batch processing. If you have data formats not currently supported you can write your own data readers using the PV-WAVE language. You can also use the PV-WAVE language to add new analysis functionality to TS-WAVE.
A Web-Enabled Solution
JWAVE lets you web-enable your applications and leverage the code you have already written. You can web-enable applications without writing any Java code, or create Java applets or applications for more interactivity. JWAVE is useful if you have a requirement that nonprogrammers and interested parties at off-site locations be able to access your data and create reports that can be viewed in web browsers. JWAVE allows you to reuse the code you have already developed and web-enable it for these end-users.
How the PV-WAVE Family Meets the Challenges in Aerospace
The PV-WAVE family addresses all of the technical data analysis requirements
discussed above for applications in aerospace, and also addresses the specific challenges in aerospace. Let's look at how these industry specific functional requirements are met in PV-WAVE:
The ability to read a large number of often proprietary formats
Data can sometimes come in standard image formats such as TIFF, GIF, JPEG, or DICOM, which PV-WAVE can easily read, but more often it is a hardware defined or in-house format. Custom code and be easily developed to read these many different file formats.
 The data may be text based ASCII data in a comma-separated, tab-separated, or fixed format, and may contain header and other imbedded, format specific data. PV-WAVE has high level routines for easily parsing and reading this data. The "Data Connect" routines allow you to read many ASCII files, such as an Excel report file, with a single command.
Binary data can be read with similar ease, with entire multi-dimensional arrays read with a single command. Using the PV-WAVE ODBC or DB Connect options, data can also be read from relational databases like ORACLE and SYBASE using a familiar SQL SELECT statement. Custom data readers can be developed in PV-WAVE code and used in the TSWAVE application.
The ability to read, manipulate and chart very large datasets
PV-WAVE introduces no software limitations on the amount of data that can be read into memory or processed and displayed.  Its 4GL language makes matrix, array and other numeric operations perform as well as native C code, allowing peak performance for applications operating on large datasets up to gigabytes in size.
Specific numeric functionality
PV-WAVE Advantage includes the industry standard IMSL library for mathematical and statistical functionality. This library of routines has been used for over 30 years in a wide variety of areas, and includes all the numeric functionality needed in most aerospace applications.
The need to reuse code and expertise from project to project
Code written in PV-WAVE, thanks to its readable and compact form, is easily re-used. Libraries of PV-WAVE routines can be written and used in different, project-specific applications. Many PV-WAVE routines are actually written in PV-WAVE code and source code is provided for them, making it easy to extend or modify the functionality in routines we provide.
The need to rapidly develop custom solutions
The PV-WAVE 4GL language lets you develop complex functionality, including user interface components and interactive graphics, in a fraction of the code required in languages such as C and C++. The ability to run code from the command prompt in PV-WAVE gets you out of the write, compile and debug method of programming. With PV-WAVE you can debug and recompile code while running an application.  You can develop powerful and concise code using PV-WAVE using its powerful language features such as: data structures like lists and associative arrays, a date/time native data type, data searching and indexing tools, vector, matrix and tensor functions, and the ability to compose and execute PV-WAVE commands at runtime from within an application.
The need to access legacy code
PV-WAVE can call legacy or third-party code in a number of ways. You can dynamically link C, C++ or FORTRAN code into PV-WAVE, and call external applications using pipes, Remote Procedure Calls (RPCs), or sockets communications. You can imbed PV-WAVE into your own C or FORTRAN application and share memory, or call PV-WAVE acting as a server using RPCs or sockets. PV-WAVE is unparalleled in the ways in which you can integrate it with in-house code or third-party applications.
Hardcopy and report generation
Many projects require either hardcopy output or charts saved in a format that can be easily imported into documents and presentations. PV-WAVE supports formats like Windows Metafile (wmf), Encapsulated PostScript (eps), and computer graphic metafile (cgm), as well as most common image formats (tiff, gif, jpeg, dicom, png, bmp, etc.). You can either send
output directly to printers or create image or raster files to allow you to import the charts into other applications.
Batch processing of reports is another common requirement, where preformatted plots are generated against new datasets and printed. This is easy to accomplish with the PV-WAVE programming language and is a built in feature in the TS-WAVE application.
Multi-platform support
PV-WAVE runs on Windows 98/NT/2000 as well as Linux, OpenVMS, and most UNIX platforms (Sun, HP, SGI, IBM, Compaq). Code written in PVWAVE, including user interface code, can be run on any platform without recompiling the code.
Broader Functionality
Some of the following functionality is used less in aerospace, but may be
important to your particular application. This shows the great breadth of functionality in PV-WAVE and how it is a tool that can meet even unforeseen needs down the road.
Accelerated 3D Graphics Using the VTK Toolkit
If you have requirements for photo-realistic 3D visualizations or the ability to rotate and fly through 3D scenes in real time, PV-WAVE can address these requirements using the Visualization Toolkit integration.  This is a best-of-breed tool, for high-end 3D visualizations, that is shipped as part of PV-WAVE with an easy to use interface to access the most commonly used functionality. Common needs for 3D visualizations in aerospace include surface plots, isosurfaces for volumetric data, and
mesh surface visualizations for computational fluid dynamics (CFD) and finite element analysis (FEA) data.
Mapping Functionality
For remote sensed data and satellite tracking displays, it can be useful to display information over cartographic maps using various projections. PVWAVE has extensive mapping functionality and built-in geographic and political map databases.
User Interface Design using PV-WAVE Widgets
For a standalone application, it is often necessary to ha ve a user interface developed both for the benefit of the developer of the application and for end-users who may not program in PV-WAVE at all. It is easy to build user interfaces in PV-WAVE using the PV-WAVE Widget routines. Higher level interactive tools are available in the Visual Data Analysis (VDA) tools that
can also be incorporated into a user interface. With PV-WAVE Widgets you can quickly put together ad hoc applications using high level routines, or go much deeper into the attributes of the Widgets to allow unlimited control of the exact look and behavior of your user interface.
Case Studies
Visual Numerics has been solving complex data problems for high profile
Aerospace customers for over a decade. All the top Aerospace companies around the world use our PV-WAVE Family of products to meet their data needs. While most of our customer applications are highly confidential, below are a few case studies we can share showing how customers in Aerospace have used PVWAVE.
| Customer: |
European Space Research Center |
| Situation: |
Developed a multipurpose data analysis software package for use in all phases of satellite systems from pre-launch testing through full commissioning. Had both satellite telemetry data and data processed at ground station, including synthetic aperture radar image data, thus both time series and image data was involved. |
| Solution: |
PV-WAVE provided the key components they needed: ability to handle large datasets, interactive 2D charts and image displays, user interface development, and image processing functions. |
| Benefits: |
They cite benefits including: performance, functionality, ease of use, multi-platform support, and outstanding technical support as key benefits in developing a true industrial quality application.
|
| Customer: |
Integral Systems |
| Situation: |
Developed an off-line analysis module for a satellite ground command and control system. Analyzed archived satellite data to track satellite performance and investigate problems. Large, complex datasets, analysis and charting, automated report generation, and multi-platform support were all requirements. |
| Solution: |
PV-WAVE provided the functionality and performance they needed. The open architecture made development easy. |
| Benefits: |
Replaced an in-house data analysis system. Saved time in development and had a positive ROI. Many of their customers using this system were so impressed that they purchased PVWAVE for their own use and to extend the capabilities of ABE.
|
| Customer: |
NASA Glenn Advanced Research Center |
| Situation: |
Research into performance of arc jet thrusters used to stabilize satellites using fluorescence image data. Parameters calculated include temperature, types of particles, energy conversion, and pattern of ionization. They also used PV-WAVE in material defect analysis using multiple lasers. |
| Solution: |
PV-WAVE provided the functionality and performance needed. Image, contour, and surface plots were used and custom algorithms for analysis. |
| Benefits: |
A better solution than in-house developed code. PV-WAVE provided a tool for rapid development of ad-hoc tools, new visualization techniques, and better performance. |
Visual Numerics - A Partner not Just a Vendor
Visual Numerics has for over 30 years provided Trusted Numerics and Scientific Graphic tools to thousands of users in aerospace. The PV-WAVE Family has the functionality you need now and in the future, including an open software environment to allow your investment to be integrated with new technologies.
Visual Numerics is your partner, not just a software vendor. We have unparalleled technical support that can answer the hard questions fast, and consulting that can provide in-depth expertise and fast delivery of time-critical solutions. Long time users can attest to the quality of support provided by Visual Numerics, which can be very important when you have difficult problems or tight schedules to meet. If you need specific expertise that is beyond what you have available in-house, or need to rapidly develop a solution and do not have the resources to develop it in-house, then the Technical Resources Group is available for contract development and are there to make sure you are successful.
 PV-WAVE has been an important tool in aerospace for many years. We work closely with many customers and are responsive to their particular needs and to the needs of the aerospace industry. The PV-WAVE Family provides you with standardized software that can provide rapid, high quality, cost-effective solutions for many kinds of problems, and deliver a positive return on your investment through maximizing your productivity.
|
|
