Autonomous power expert system

The Autonomous Power Expert (APEX) system has been designed to monitor and diagnose fault conditions that occur within the Space Station Freedom Electrical Power System (SSF/EPS) Testbed. The APEX system is being developed at the NASA Lewis Research Center by the Space Electronics Division (SED) in conjunction with the Space Station Directorate and Power Technology Division (PTD). APEX is designed to interface with SSF/EPS testbed power management controllers to provide enhanced autonomous operation and control capability.

The APEX architecture consists of three components: (1) a rule-based expert system, (2) a testbed data acquisition interface, and (3) a power scheduler interface. Fault detection, fault isolation, justification of probable causes, recommended actions, and incipient fault analysis are the main functions of the expert system component. The data acquisition component requests and receives pertinent parametric values from the EPS testbed and asserts the values into a knowledge base. Power load profile information is obtained from a remote scheduler through the power scheduler interface component.

This paper will discuss the current APEX design and development work. Operation and use of APEX by way of the user interface screens will also be covered.     

Autonomous power system: Intelligent diagnosis and control

The Autonomous Power System (APS) project at NASA Lewis Research Center is designed to demonstrate the abilities of integrated intelligent diagnosis, control, and scheduling techniques to space power distribution hardware. Knowledge-based software provides a robust method of control for highly complex space-based power systems that conventional methods do not allow. The project consists of three elements: the Autonomous Power Expert System (APEX) for fault diagnosis and control, the Autonomous Intelligent Power Scheduler (AIPS) to determine system configuration, and power hardware (Brassboard) to simulate a space based power system. APEX is a software system that emulates human expert reasoning processes to detect, isolate, and reconfigure in the case of a power system distribution fault. The APEX system continuously monitors the operating status of the Brassboard and reports any anomaly (either static or incipient) as a fault condition. APEX functions as a diagnostic advisor aiding the user in isolating the probable cause of the fault. Upon isolating the probable cause, APEX automatically reconfigures the Brassboard based upon internal knowledge as well as information from the scheduler. APEX provides a natural language justification of its reasoning processes and a multilevel graphical display to depict the status of the Brassboard. AIPS is an intelligent scheduler used to control the efficient operation of the Brassboard. A database is kept of the power demand of each load on the Brassboard and its specified duration and priority. AIPS uses a set of heuristic rules to assign start and end times to each load based on priorities, as well as temporal and resource constraints. When a fault condition occurs, AIPS assists APEX in reconfiguring the system. The APS Brassboard is a prototype of a space-based power distribution system and includes a set of smart switchgear, power supplies, and loads. Faults can be introduced into the Brassboard and, in turn, be diagnosed and corrected by APEX and AIPS. The Brassboard also serves as a learning tool for continuously adding knowledge to the APEX knowledge base. This paper describes the operation of the APS as a whole and characterizes the responsibilities of the three elements: APEX, AIPS, and Brassboard. A discussion of the methodologies used in each element is provided. Future plans are discussed for the growth of the APS.     

The correspondence between U.K. 'action levels' for lead in blood and in water

This paper considers whether the Department of the Environment's water lead concentration criterion for lead pipe replacement and action in individual cases, i.e. 50 micrograms/l in any sample, is too high when set against the Department of Health's advisory action limit for blood lead concentration of 25 micrograms/100 ml. The relationships between blood lead and water lead concentrations found in the Glasgow and Ayr duplicate diet studies, together with unpublished data from Glasgow and Liverpool, indicate that over 10% of people exposed to an average water lead concentration of 100 micrograms/l (the earlier action level) would have blood lead concentrations above 25 micrograms/100 ml, as would about 4% of those exposed to 50 micrograms/l (the Maximum Admissible Concentration in an EEC Directive). For adults, average water lead concentrations should not exceed 30 micrograms/l to ensure compliance with the limit for blood lead, i.e. so that not more than 2% exceed 25 micrograms/100 ml. However, for one of the critical groups, bottle-fed infants (whose diet is 90% water), average water lead concentrations should not exceed 10-15 micrograms/l. The WHO's Provisional Tolerable Weekly Intake (PTWI) for children (25 micrograms/kg body weight) also implies that their water lead concentrations should not exceed 10-15 micrograms/l.     

Data-parallel programming on a network of heterogeneous workstations

We describe a compiler and run-time system that allow data-parallel programs to execute on a network of heterogeneous UNIX workstations. The programming language supported is Dataparallel C, a SIMD language with virtual processors and a global name space. This parallel programming environment allows the user to take advantage of the power of multiple workstations without adding any message-passing calls to the source program. Because the performance of Individual workstations in a multi-user environment may change during the execution of a Dataparallel C program, the run-time system automatically performs dynamic load balancing. We present experimental results that demonstrate the usefulness of dynamic load-balancing In a multi-user environment These results suggest that initially allocating the same amount of work to each processor and letting the dynamic load balancing algorithm adjust the load during program execution yields very good performance. Hence neither the compiler nor the run-time system need a priori knowledge of the speeds of the machines that will participate in a program execution.     

Messenger Materials Moving Forward: The Role of Functional Polymer Architectures as Enablers for Dynamic Nano-to-Macro Messaging

Identifying changes in the nanoscopic domain is a key challenge in the physicochemical sciences, where great interest is on sensing complex processes that involve cellular biochemical reactions, chemical heterogeneities, contact forces, and other interfacial phenomena. This has stimulated the development of diverse materials that allow subtle nanoscopic environments to be "seen". The challenge in the nano-domain has always been the ability to sense changes on the minute scale and rapidly transduce the information out for macroscopical observation. Ideally, materials should inform when processes are occurring. Recently, new systems that leverage established concepts with fluorescence- and plasmonic-based sensing have been devised, which has reinvigorated the domain, where functional polymers coupled in specific architectures to transducing motifs allow for a new basis of messenger materials to be realized. The key aspect in this regard is that the polymers allow for sensing to be achieved only when they are carefully coupled to the amplification system. In this perspective, the role of specific functional polymer architectures for the realization of nano-to-macro sensing of subtle nano-messengers is discussed and where the exciting field of messenger materials is seen moving forward is pointed out.