API TR 2567-2005 Evaporative Loss from Storage Tank Floating Roof Landings《浮顶罐的蒸发损耗》.pdf

1、Evaporative Loss from Storage Tank Floating Roof Landings TECHNICAL REPORT 2567 APRIL 2005 Evaporative Loss from Storage Tank Floating Roof Landings Measurement Coordination TECHNICAL REPORT 2567 APRIL 2005 Prepared by: Robert L. Ferry The TGB Partnership SPECIAL NOTES API publications necessarily a

2、ddress problems of a general nature. With respect to particular circumstances, local, state, and federal laws and regulations should be reviewed. Neither API nor any of APIs employees, subcontractors, consultants, committees, or other assignees make any warranty or representation, either express or

3、implied, with respect to the accuracy, completeness, or usefulness of the information contained herein, or assume any liability or responsibility for any use, or the results of such use, of any information or process disclosed in this publication. Neither API nor any of APIs employees, subcontractor

4、s, consultants, or other assignees represent that use of this publication would not infringe upon privately owned rights. API publications may be used by anyone desiring to do so. Every effort has been made by the Institute to assure the accuracy and reliability of the data contained in them; howeve

5、r, the Institute makes no representation, warranty, or guarantee in connection with this publication and hereby expressly disclaims any liability or responsibility for loss or damage resulting from its use or for the violation of any authorities having jurisdiction with which this publication may co

6、nflict. API publications are published to facilitate the broad availability of proven, sound engineering and operating practices. These publications are not intended to obviate the need for applying sound engineering judgment regarding when and where these publications should be utilized. The formul

7、ation and publication of API publications is not intended in any way to inhibit anyone from using any other practices. All rights reserved. No part of this work may be reproduced, stored in a retrieval system, or transmitted by any means, electronic, mechanical, photocopying, recording, or otherwise

8、, without prior written permission from the publisher. Contact the Publisher, API Publishing Services, 1220 L Street, N.W., Washington, D.C. 20005. Copyright 2005 American Petroleum Institute FOREWORD The methodologies presented in this report were presented previously in a January 23, 2002, report,

9、 Tentative Method for Determining Storage Tank Evaporative Losses from Floating Roof Landings, prepared for API by Robert L. Ferry of The TGB Partnership. The purpose of this revision is to address a scenario that was not addressed in the earlier report. The earlier report addressed storage tanks th

10、at retain a heel of stock liquid across the entire bottom of the tank when the floating roof is landed, and storage tanks that are drained dry. This revision addresses the intermediate case of a partial liquid heel, in which pools of stock liquid remain in the tank (and thus it is not drained dry),

11、but the free standing liquid does not cover the entire bottom of the tank. There have been no changes to the equations or factors presented in the earlier report, other than the addition of the case for a partial liquid heel. API publications may be used by anyone desiring to do so. Every effort has

12、 been made by the Institute to assure the accuracy and reliability of the data contained in them; however, the Institute makes no representation, warranty, or guarantee in connection with this publication and hereby expressly disclaims any liability or responsibility for loss or damage resulting fro

13、m its use or for the violation of any federal, state, or municipal regulation with which this publication may conflict. Suggested revisions are invited and should be submitted to API, Standards department, 1220 L Street, NW, Washington, DC 20005, standardsapi.org. ii TABLE OF CONTENTS Section Page 1

14、. EXECUTIVE SUMMARY 1 1.1 Statement of Purpose .1 1.2 Summary of the Investigations1 1.3 Scope and Limitations of the Model2 1.4 Proposed Estimating Methods 3 1.4.1 Standing Idle Loss.3 1.4.2 Filling Loss 3 1.4.3 Total Landing Loss 4 1.4.4 Landing Loss Estimation Equations.4 2. DESCRIPTION OF CONCEN

15、TRATION AND SATURATION.9 3. DESCRIPTION OF FLOATING-ROOF LANDING LOSSES.9 3.1 Landing Loss Events.9 3.2 Standing Idle Loss Mechanisms .10 3.2.1 Internal Floating-Roof Tanks With a Liquid Heel .10 3.2.1.1 Confidence in the Breathing Loss Model.10 3.2.1.2 Derivation of the Breathing Loss Model.11 3.2.

16、2 External Floating-Roof Tanks With a Liquid Heel 13 3.2.2.1 Confidence in the Wind Effects Loss Model 14 3.2.2.2 Derivation of the Wind Effects Loss Model14 3.2.3 Internal or External Floating-Roof Tanks That Drain Dry.16 3.2.3.1 Confidence in the Clingage Model.16 3.2.3.2 Derivation of the Clingag

17、e Model 17 3.3 Filling Loss Mechanism.18 3.3.1 Internal Floating-Roof Tanks With a Liquid Heel .18 3.3.1.1 Confidence in the Submerged-Fill Loading Model.18 3.3.1.2 Derivation of the Submerged-Fill Loading Model.18 3.3.2 External Floating-Roof Tanks With a Liquid Heel 19 3.3.2.1 Confidence in the Wi

18、nd-Affected Loading Model.19 3.3.2.2 Derivation of the Wind-Affected Loading Model 19 3.3.3 Internal or External Floating-Roof Tanks That Drain Dry.20 3.3.3.1 Confidence in the Drain Dry Loading Model20 3.3.3.2 Derivation of the Drain Dry Loading Model20 4. SAMPLE CALCULATIONS.20 4.1 Accounting for

19、Cone-Down Bottoms.20 4.2 Worked Examples.22 5. CONCLUSION 23 6. REFERENCES 25 iii Figures Figure 1. Vacuum-Breaker Vent .6 Figure 2. Standing Idle Loss (Emissions) .7 Figure 3. Filling Loss (Emissions).7 Figure 4. Tank with a Liquid Heel .8 Figure 5. Drain-Dry Tank 8 Figure 6. Wind Action .13 Figure

20、 7. Wind Effects 14 Figure 8. Saturation Pattern of Drain-Dry Tanks versus Tanks with a Liquid Heel .16 Figure 9. Volume of Vapor Above a Liquid Heel.21 Figure 10. Volume of Vapor in a Cone-Down Bottom21 Tables Table 1. Summary of Floating-Roof Landing Loss Estimation Methods by Tank Type 4 Table 2.

21、 Properties of Selected Petroleum Stocks . 6 Table 3. Effective Height of Liquid, hle, and Height of Vapor Space, hv22 Appendix I Examples. 26 iv Evaporative Loss from Storage Tank Floating Roof Landings 1. EXECUTIVE SUMMARY 1.1 Statement of Purpose The purpose of this study was to investigate stora

22、ge tank emissions that may result from landing and subsequently refloating a floating roof. The existing emission factors for floating-roof tanks1,2are based on the assumption that the floating roof is continuously floating on the stored stock liquid. Additional emissions may occur, however, if the

23、tank is emptied such that the floating roof is no longer floating. When the liquid level approaches the bottom of the tank, the floating roof lands on deck legs or other supports which prevent it from dropping any further as the stock liquid continues to be removed. Further withdrawal of stock liqui

24、d could then potentially form a partial vacuum beneath the landed floating roof. If the receding liquid were to create an excessive partial vacuum, the floating roof could collapse. To avoid this condition, a vacuum-breaker vent on the floating roof opens automatically as the floating roof lands (se

25、e Figure 1 on page 6). The vapor space created under the floating roof is thereby freely vented to the space above the floating roof. Vapor loss (and the corresponding emissions to the atmosphere) may occur while the tank remains nominally empty and the floating roof continues to stand idle in this

26、landed condition (see Figure 2 Standing Idle Loss). Additional emissions may occur during the refilling of the tank, as the vapor space beneath the floating roof is displaced by the incoming stock liquid (see Figure 3 Filling Loss). This study sought to quantify these floating-roof landing loss emis

27、sions. 1.2 Summary of the Investigations Part I of this study proposed a methodology for estimating floating-roof landing losses for tanks storing refined products. Steps pursued in the development of this methodology included a literature search,1-14a survey of manufacturers and owners of petroleum

28、 storage tanks, computer modeling, and analyses of available emissions data from fixed-roof tank breather vents and from barge loading operations. The results of Part I were presented in the report, Determining Product Evaporation from Tank Turnovers,15October 1, 1997, prepared for API by Robert L.

29、Ferry and J. Randolph Kissell of The TGB Partnership (TGB). Data collection for Part II of the study involved field tests that were conducted in 1998 and 1999. Data were obtained from four tanks, each representing a different combination of tank construction and type of stock liquid. It was not econ

30、omically feasible to test a sufficient number of tanks to empirically determine emission factors for floating-roof landing losses, given the spectrum of construction configurations and storage conditions found in the industry. Furthermore, limitations on accessibility to the space under the floating

31、 roof impose constraints on the field methods used. These limitations were found to impact the absolute accuracy of the data gathered. While it was recognized that the testing of a single tank is insufficient for determining a typical value for the entire population of similar tanks, and that the da

32、ta gathered should be interpreted as relative indicators rather than as absolute measures, it was expected that comparison of the data from each of the test tanks would indicate relative trends for the different tank configurations. The results of these field tests were presented in the report, Dete

33、rmining Product Evaporation from Tank Turnovers, Part II - Tank Testing,16May 1999, prepared for API by Sue Sung and Yousheng Zeng of Trinity Consultants. 1 2 API Technical Report 2567 Part II of the study also included a review by TGB of the methodology proposed in Part I, in light of the field tes

34、t results and other available data. This review specifically considered the following: a. whether wind effects should be included in the model, b. extension of the model to include crude oil, and c. reasonableness of the model in light of the relative trends exhibited in the field data. Part III of

35、the study incorporated the results of Parts I and II into a revised model for estimating floating-roof landing losses. This revised model was presented in the January 23, 2002 report, Tentative Method for Determining Storage Tank Evaporative Losses from Floating Roof Landings,17prepared for API by R

36、ob Ferry of TGB. API sponsored an additional field study in 2003 that sought to overcome the testing difficulties experienced in the earlier field work. The first step of this study was to develop field and laboratory protocols for measuring the concentration of vapors displaced from under a landed

37、floating roof during the refilling process. These protocols were then applied to obtain data from three internal floating-roof tanks, each of which was in gasoline service. This 2003 field study provided spot validation of the 60% vapor saturation value proposed in the 1/23/02 TGB Report for interna

38、l floating-roof tanks with a full liquid heel (see Table 1). The results of the 2003 study were presented in the January 28, 2004 report, Floating Roof Landing Loss: Field Study of Saturation Factors for Refilling of Internal Floating-Roof Tanks,18prepared for API by Rob Ferry of TGB. In addition to

39、 spot validation of the full liquid heel case, the 2003 study collected samples from tanks for which the liquid heel did not extend across the entire bottom of the tank. The free-standing liquid in these partial liquid heel cases was confined to the area in or near the sump. This case of a partial l

40、iquid heel was not addressed in the 1/23/02 TGB Report, in that none of the supporting data were applicable to such a case. In order to better quantify a saturation level for the case of a partial liquid heel, API commissioned additional testing in 2004. The purpose of the 2004 study was to obtain a

41、dditional data points, to be combined with the data collected from similar tanks in 2003, in order to develop a saturation factor for the condition of a partial liquid heel. The results of the 2004 study were presented in the November 15, 2004 report, Floating Roof Landing Loss: Field Study of A Ref

42、ill Saturation Factor For An IFRT With A Partial Liquid Heel,19prepared for API by Rob Ferry of TGB. This report retains the methodology presented in the 1/23/02 TGB Report, and adds the case of a partial liquid heel. 1.3 Scope and Limitations of the Model The emissions characterized as floating-roo

43、f landing losses in this study are those that would be expected to occur if a floating roof is landed in the course of normal operations, and subsequently refilled. This study does not address emissions that may result from additional activities, such as degassing or tank cleaning, that may occur wh

44、ile the tank is empty. The model is intended for use with any petroleum liquid. The inclusion in the model of the stock liquids physical properties (i.e., true vapor pressure, vapor molecular weight, and liquid density) appears to effectively differentiate crude oil from gasoline, and therefore no f

45、urther differentiation was made in the form of product factors or other product-specific adjustments. The model assumes that the stock liquid used to refill the tank is the same as that stored prior to landing the floating roof. Situations in which there is a change of service (i.e., the tank is to

46、be filled with a different Evaporative Loss from Storage Tank Floating Roof Landings 3 stock than it had been storing) may warrant differentiating between the stock vapor properties for the arrival and generated components of filling loss. The model does not address standing idle losses for partial

47、days. It would be conservative (i.e., potentially overestimate emissions) to apply the model to episodes during which the floating roof remains landed for less than a day. Any emission factor is properly understood as representative of the actual emission rates that are typical for a population of e

48、mission points. For a non-uniform population, however, there is an inherent level of uncertainty associated with the application of the general emission factor to any individual emission point. Some of the critical sources of uncertainty in this model of floating-roof landing losses are addressed in

49、 the comments on the confidence associated with each step of the model. As noted in these comments, some of the variables have not been well defined, and the values shown are intended to serve only as placeholders pending further research. 1.4 Proposed Estimating Methods Floating-roof tanks were segregated into the following categories for purposes of estimating landing losses: a. internal floating-roof tanks (IFRTs) with a full or partial liquid heel, b. external floating-roof tanks (EFRTs) with a full or partial liquid heel, and c. IFRTs and EFRTs that drain dry. The two mod