This paper describes the progress made in developing geothermal resources at Ulumbu Flores, Indonesia for utilization mini geothermal power generation. Two deep exploratory wells drilling drilled by PLN confirmed the existence of the resources. The well measurement carried out during drilling and after completion of the well indicated that the major permeable zone at around 680 m depth and that this zone is a steam cap zone, which is likely to produce high enthalpy steam. The above information indicates that well ULB-01 will produce a mass flow at least 40 tonnes per hour, which will ensure a 3 MW (E) Ulumbu mini geothermal power plant.

PREFACE The Twenty-First Workshop on Geothermal Reservoir Engineering was held at the Holiday Inn, Palo Alto on January 22-24, 1996. There were one-hundred fifty-five registered participants. Participants came from twenty foreign countries: Argentina, Austria, Canada, Costa Rica, El Salvador, France, Iceland, Indonesia, Italy, Japan, Mexico, The Netherlands, New Zealand, Nicaragua, the Philippines, Romania, Russia, Switzerland, Turkey and the UK. The performance of many geothermal reservoirs outside the United States was described in several of the papers. Professor Roland N. Horne opened the meeting and welcomed visitors. The key note speaker was Marshall Reed, who gave a brief overview of the Department of Energy's current plan. Sixty-six papers were presented in the technical sessions of the workshop. Technical papers were organized into twenty sessions concerning: reservoir assessment, modeling, geology/geochemistry, fracture modeling hot dry rock, geoscience, low enthalpy, injection, well testing, drilling, adsorption and stimulation. Session chairmen were major contributors to the workshop, and we thank: Ben Barker, Bobbie Bishop-Gollan, Tom Box, Jim Combs, John Counsil, Sabodh Garg, Malcolm Grant, Marcel0 Lippmann, Jim Lovekin, John Pritchett, Marshall Reed, Joel Renner, Subir Sanyal, Mike Shook, Alfred Truesdell and Ken Williamson. Jim Lovekin gave the post-dinner speech at the banquet and highlighted the exciting …

Inertial confinement fusion driven by beams of heavy ions is an attractive route to controlled fusion. In the U.S., induction accelerators are being developed as {open_quotes}drivers{close_quotes} for this process. This paper is divided into two main sections. In the first section, the concept of induction-accelerator driven heavy-ion fusion is briefly reviewed, and the U.S. program of experiments and theoretical investigations is described. In the second, a {open_quotes}taxonomy{close_quotes} of space-charge-dominated beam physics issues is presented, accompanied by a brief discussion of each area.

Motor current and power analysis methods have been developed to assist in the condition monitoring of a variety of motor-driven devices. The early work in this area was conducted at Oak Ridge National Laboratory (ORNL) on motor-operated valves in the mid-to-late 1980`s in support of the US Nuclear Regulatory Commission`s Nuclear Plant Aging Research Program. The successful implementation of motor current signature analysis (MCSA) as a diagnostic for valves led to its application to other devices and to refinements in the methodologies used. Motor current and power analysis have been found to provide information that is complementary to that available from conventional diagnostics, such as vibration and pressure pulsation analysis. Inherent signal filtering associated with rotor to stator magnetic field coupling does limit the high frequency response capability of the motor as a transducer; as a result, certain phenomena, such as pump or fan vane pass energy, is not readily apparent in the motor electrical signals. On the other hand, the motor-monitored parameters have often been found to be much more sensitive than vibration transducers in detecting the effects of unsteady process conditions resulting from both system and process specific sources.

Many automatic text indexing and retrieval methods use a term-document matrix that is automatically derived from the text in question. Latent Semantic Indexing, a recent method for approximating large term-document matrices, appears to be quite useful in the problem of text information retrieval, rather than text classification. Here we outline a method that attempts to combine the strength of the LSI method with that of neural networks, in addressing the problem of text classification. In doing so, we also indicate ways to improve performance by adding additional {open_quotes}logical sensors{close_quotes} to the neural network, something that is hard to do with the LSI method when employed by itself. Preliminary results are summarized, but much work remains to be done.