| dc.description.abstract | The primary focus of this thesis is to investigate two particular heat exchangers in the steam power plant located at the University of Kansas. The secondary focus of this thesis is to compare the power consumption of two specific condensate pumps, also in the steam power plant located at the University of Kansas. The power plant generates and supplies steam (not electricity) to the buildings on the KU campus. The boilers in the plant require fuel (natural gas) to convert the feed water into steam. The two heat exchangers have been investigated to see how they affect the amount of natural gas that is required by the boiler during normal plant operation. Also, the power consumption of two specific condensate pumps, one constant speed and the other variable speed, has been compared during normal plant operation. Heat exchangers have been widely used to substantially contribute to energy consumption savings, especially in power plants. There are presently two Bell & Gosset SU type heat exchangers in the steam power plant located on the University of Kansas campus. One of the heat exchangers is located in the basement of the power plant while the other heat exchanger (also known as the “vent condenser”) is located on the first floor of the power plant. A comparative study has been conducted to investigate the amount of natural gas saved in one year by each of these heat exchangers. There are two types of pumps that are used to supply condensate water in the power plant. One of these pumps is a Worthington D-824 constant speed pump, while the other is a pair of Grundfos CRE 15-3 variable speed pumps. These pumps provide water to the deaerator tank. The DA tank uses steam to preheat the condensate water and remove air and other non- condensable gases from that water before the water flows to the boilers. Both of these types of condensate pumps also move the condensate water up to the vent condenser on the first floor. Different instruments were used for the measurement and recording of data in this thesis. The temperature data of the water was recorded by installing temperature sensors on the outside surfaces of the pipelines. These temperature sensors were installed at the inlets and outlets of the vent condenser and the basement heat exchanger. Temperature was recorded on a per-minute basis. Pump data such as discharge pressure, discharge flow rate and power consumption were also recorded on a per-minute basis. Pressure transducers were used to record pressure; magnetic flow meters were installed to record flow rate; and a power measuring device associated with the pumps helped in measuring power consumption. There was a common data acquisition system for all components. A Gateway laptop with HOBOWARE software installed was used to record and plot the data obtained from the respective data measuring instruments. The basement heat exchanger’s function is to heat the cold makeup water (i.e., the water received from the city of Lawrence) before it goes to the condensate storage tanks, which are also located in the basement. Also, this heat exchanger helps to reduce the boiler blowdown water temperature to a temperature less than 1400 F (for environmental safety standards) before this water is drained to the sewers for disposal. In contrast, the vent condenser receives steam (along with associated non-condensable gases) from an open feed water heater in the basement (also called the Dearator tank or DA tank). This steam heats the condensate water flowing through the vent condenser. The vent condenser then returns this heated condensate water back to the condensate storage tanks located in the basement. Natural gas is used in the boiler to convert the condensate water into steam. The use of these two heat exchangers saves natural gas which can be assigned a dollar value. It was found that the vent condenser saves approximately $27,680 for one year, while the basement heat exchanger saves approximately $7,660. In continuation of the work done by a previous KU-ME graduate student, Raoof Alabdullah, the power consumption of both the Worthington pump and the Grundfos variable speed pumps was investigated for each month from June of 2015 through November of 2015. The average power consumption for the Worthington pump over this time period was found to be 5.29 kW. The Grundfos pumps can operate in either of two modes: pressure control or level control; and the average power consumption of the Grundfos pumps is considerably different, depending on operation mode. The pressure control mode of the Grundfos pumps is similar to that of the Worthington pump’s operation, i.e., the Grundfos pumps run at a constant discharge pressure. Operation in level control mode depends on the water level of the DA tank. Based on how close this water level is to the target level (which is 52% of tank capacity), the Grundfos pumps either speed up or slow down, causing fluctuations in the discharge pressure. Also, while running in level control mode, the Grundfos pumps do not have enough pressure head to supply water to the vent condenser on the first floor. On October 21 and October 29, 2015, level control mode was employed from 1:30 pm to 3:30 pm; and the power consumption of the Grundfos pumps during this mode was compared to that of the pressure control mode runs made from 12 pm to 1:30 pm on those same days. The Grundfos pumps consumed 1.33 kW on October 21 and 3.66 kW on October 29, during the time periods when they ran in level control mode. This was considerably less than the 5.2 kW and 5.6 kW consumed by the pumps when they ran in pressure control mode on those same days. Thus, the level control mode allowed for significant decreases in power consumption. A theoretical comparative study was also done between the Worthington pump and a less powerful condensate pump that would provide condensate water to the DA tank only. A Dayton constant speed pump was selected for this purpose. The vent condenser was assumed to be absent from the power plant, so no condensate water flowed to the first floor. This was compared to the current system with the Worthington constant speed pump supplying water to both the DA tank and the vent condenser. It was estimated that, for a 3% rate of interest, the presence of the vent condenser yields $20,296 in yearly savings as compared to just $2,470 if there were no vent condenser present and the pump were smaller. The savings calculations are with respect to the baseline: the Worthington pump being used without the vent condenser. The yearly costs of the Worthington pump and the Dayton pump were factored into the calculation of these savings, as was the cost of the vent condenser. It is thus concluded that the vent condenser and the basement heat exchanger should continue to be used because of the large amount of natural gas savings. Also, the Grundfos pumps should be run in level control mode whenever possible, so that there is less power consumption by the pumps. In addition, it is recommended that the vent condenser be moved from the first floor to the basement near the DA tank, so that the Grundfos pumps can be operated in level control mode while the vent condenser is also in operation. By the simultaneous use of the vent condenser and the Grundfos pumps in level control mode, natural gas savings of an additional $15,000 for a period of 20 years might be achieved. Different methods can be attempted in the future so as to improve calculation accuracy. To get even more accurate temperature data, instead of the external temperature sensors presently used, temperature sensors which are inserted into the pipelines could be installed at the inlets and outlets of the vent condenser and the basement heat exchanger. This would allow the sensors to be in direct contact with the condensate water. The data acquisition system for such sensors would thus record the most accurate temperatures. Presently, only the flow rate of the condensate water for the vent condenser and makeup water flow rate for the basement heat exchanger are known. Flow meters could be installed in the shell side of both the vent condenser and the basement heat exchanger. The flow rate data of the condensed steam in the vent condenser’s shell side and of the flash tank water in the basement heat exchanger’s shell side would then be known. This would help to improve the accuracy of calculated natural gas savings. Also for future purposes, a flow meter with an associated data acquisition system could be installed in the makeup water line. This would help in recording the actual flow rate of makeup water, rather than estimating the makeup water flow rate based on temperature rise of the water as has been done in this thesis. Presently, there is no flow meter installed in the makeup water line. So, the makeup water’s average flow rate is estimated based on the temperature rise in the makeup water across the heat exchanger and the hourly volumetric usage of water which is recorded “by hand” by the power plant staff. | |
