Journal of Energy Engineering

A mathematical model for the performance of spray ponds is developed using a detailed computational approach. The model considers heat transfer and evaporation from individual droplets of sprayed water and the pond surface, and drift loss from the sprays. Modification of the temperature and humidity of the surrounding air is taken into account in calculating heat and mass transport from the sprays. Flow of air through the spray field is considered to be either driven by the ambient wind, or by natural convection. Relationships for heat and mass transfer from the drops are derived from the empirical studies of Ranz and Marshall. Numerical experiments demonstrate that simplified models can still be useful for spray pond performance assessments. The models are validated with data on several industrial and nuclear power plant spray ponds, and with data from an extensively instrumented, experimental spray pond. Results of the validation studies demonstrate generally good agreement. These models are used in a complex methodology for predicting the performance of spray ponds used for nuclear power plant service, although it is not limited to this application. A companion paper describes the overall assessment methodology.

This paper presents a complex methodology for assessing the performance of spray ponds in ultimate heat sink service at nuclear power plants. A spray pond performance model, developed in the companion paper, is used in conjunction with on‐site and off‐site meteorological data to predict the highest temperature and greatest 30‐day water loss which can reasonably be expected to occur during the lifetime of the plant. The performance model for heat and mass transfer is used to develop an efficient phenomenological model used to scan the long‐term meteorological records. Refined estimates of temperature or water loss may then be based on more complicated models if necessary. Short‐term onsite data are correlated to the long‐term off‐site data to formulate correction factors for the difference in location. Cumulative distribution functions for temperature and water loss are determined from the longterm meteorological records to predict the occurrence of these quantities which are less severe than the peak. The methodology is demonstrated using data and parameters from the Palo Verde nuclear plant as an example.

A $5\text{,}000{\mathrm{km}}^2$ catchment in Transkei, Southern Africa is monitored with 10 rain gages which transmit data to a microcomputer model. The numerical model is based on the kinematic equations with both an overland flow/stream flow component and a subsurface component of flow for 14 sub‐catchments. Both flood flows and base flows from the catchment are predicted using the model on a real‐time basis. In addition to being able to predict flows ahead by the lag time of the catchment (of the order of one day), projection can be made further ahead using rainfall projections. The low cost of the receiving, transmitting and processing system is made possible by coupling simple tipping bucket type rain gages through a processor to a standard microcomputer. The numerical model is simple enough to operate on the microcomputer and can readily be extended to assist in operation of the hydroelectric station to optimize the available water.

Experimental and analytical studies are conducted to develop a novel water intake design for the prevention of fish impingement and excessive damage to fish larvae. Requirements by the regulatory agency lead to the use of cylindrical screens mounted in pairs on “risers” connected to four manifolds, which in turn were connected to an 18 ft diam withdrawal pipe carrying a total flow of $23.2m^3/s$ (818 cfs) to a powerplant condenser. Separate model studies conducted on individual screen panels, individual and multiple risers and manifolds are described. The studies conducted at scale ratios ranging from 1:20‐1:3 are described. The system buit in 1981 continues to work well with only minor damage sustained by ice.

The applications of a generalized numerical model of coal characteristics are described. The model simulates the mixing, blending, diffusion, and shuffling of coal lots during mining, transportation, storage and handling to generate frequency distributions of as‐delivered coal characteristics (such as heat value and ash, metal or sulfur content) given one or more mine characteristics (e.g., sulfur content distribution). The model demonstrates that procedures used throughout coal‐handling operations affect the mean and variance of selected characteristics of as‐delivered coal. As an example of the model's usefulness, it is shown that a reduction in sulfur content or its variance by modification of coal‐handling procedures may result in a reduction in the amount of sulfur dioxide (air pollutant) emission control required on coal combustion sources. In some cases, the marginal cost of modifying coal‐handling procedures may be far less than the cost of added emission controls.

This paper reports on an experimental effort to generate S‐N (dynamic stress‐fatigue life) curves for a Drake ACSR (aluminum conductor steel reinforced) conductor. In the testing, aeolian vibrations were simulated by mechanical means in a controlled laboratory environment for one set of line conditions. Fatigue data for the first five strand breaks were evaluated, and S‐N curves for the first through fifth strand breaks were fit to the data using linear regression analyses. A minimum of four duplicative tests at each stress level tested were conducted to allow an evaluation of the statistical parameters and to provide statistical credibility to the results. A pseudo endurance limit of 21.31 MPa (3.09 ksi) was estimated from the S‐N data. The S‐N curves generated in this study provide much needed information for transmission line designers on the fatigue characteristics of Drake ACSR conductors. It is recommended that the S‐N curves be used in conjunction with a probabilistic approach in designing transmission lines for aeolian fatigue.