Wright, John D.
| 1. | Fluid velocity measuring apparatuε for indicating the average rate of fluid flow in an open channel conduit, compriεing: (a) εenεor meanε for producing local depth (Ds) and local velocity (Vs) signals at a given openchannel site; (b) memory means (56) for storing a plurality of setε of reference depth (DR) and reference velocity (VR) signals corresponding to a plurality of flow coefficients (CF) each of which iε repreεentative of the ratio between averaged to εenεed velocity, reεpectively; (c) compariεon means (58) responεive to εaid local depth and local velocity εignalε and to εaid reference depth and reference velocity εignalε for εelecting the corresponding flow coefficient (CF) , respectively; and (d) first multiplier meanε (64) for multiplying εaid local velocity εignal and εaid flow coefficient, thereby to produce an average velocity signal (V) . |
| 2. | Apparatus aε defined in claim 1, wherein εaid flow coefficientε correspond with different conduit sizeε and croεεεectional configurationε, reεpectively. |
| 3. | Apparatus as defined in claim 1, wherein εaid flow coefficientε correspond with different velocity εensor types. |
| 4. | Apparatus ?.s defined in claim 1, wherein said flow coefficient iε depthdependent; and further including modifying means (62) responsive to said site depth (Ds) for producing a diεtinctive correεponding flow coefficient CF') that iε εupplied to aid firεt multiplier means. |
| 5. | Apparat s aε defined in claim 4, wherein εaid comparison mea is responsive to the size and cross sectional confi uration of the conduit, and means (53,57) for supplying to said comparison means a conduit signaD (Ps) that is a function of the conduit εize and configuration. |
| 6. | Apparatuε as defined in claim 5, and further including second multiplying meanε (67) reεponsive to εaid condu; t εignal and to said local depth εignal for producing m area εignal (As) , and third multiplier means (66) responsive to said average velocity signal and to sa;.d area signal for producing an output flow signal (Qs) . |
| 7. | Apparatus as defined in claim 1, wherein said comparison meanε (158) iε responsive to the ratio of said local depth and local velocity εignals; (e) and further including divider meanε (153) responsive to said local depth and local velocity εignalε for producing a signal (Ds/Vs) that iε the ratio of εaid local depth and local velocity εignals, said compariεon meanε being reεponsive to the ratio of said local depth and local velocity signal to produce a diεtinctive flow coefficient. |
| 8. | Apparatuε aε defined in claim 1, wherein said senεor means comprises an electromagnetic senεor; and further wherein εaid memory means contains sets of depth and velocity reference signals representative of those produced under controlled conditions by electromagnetic senεor meanε. |
| 9. | Apparatuε aε defined in claim 1 wherein said sensor means compriεeε a εubmerged Doppler εignal generator; and further wherein εaid memory meanε contains sets of depth and velocity reference signals representative of those produced under controlled conditionε by εubmerged Doppler signal generator means. |
| 10. | Apparatus as defined in claim 1, wherein εaid εensor means is selected from the group consiεting of εubmerged acoustic ti eoftravel velocity εenεing meanε, acoustic lookdown velocity sensing meanε, microwave lookdown velocity sensing means, and lookdown optical sensing means; and further wherein εaid memory meanε containε setε of depth and velocity reference signals representative of those produced under controlled conditions by said velocity signal εourceε, reεpectively. |
| 11. | The method for meaεuring the flow rate of fluid flowing in an openchannel conduit, comprising: (a) generating a plurality of sets of reference depth (DR) and reference velocity (VR) signals correεponding to a plurality of flow coefficientε (CF) each of which is representative of the ratio between average velocity to εensed velocity, reεpectively. (b) εtoring said flow coefficientε in a εignal co pariεon means; (c) εupplying the εenεed depth (Ds) and velocity (Vs) signals to said comparison means to produce a correεponding flow coefficient (CF) from εaid εtorage means; and (d) multiplying said corresponding flow coefficient with said sensed velocity signal thereby to produce average velocity (V) . |
| 12. | The method as defined in claim 11, and further compriεing the εtepε of: (e) producing a pipe εize signal (Ps) that is a function of the size of the conduit; (f) multiplying said sensed depth signal (Ds) by said pipe εize εignal (Ps) , thereby to produce an area signal (As) ; and (g) multiplying εaid area εignal by εaid average velocity εignal to produce a flow εignal Q. |
| 13. | The method aε defined in claim 12, and further including the εtep of: (h) εtoring a plurality of sets of data including reference velocity (VR) , reference depth (DR) , pipe size (Ps) , sansor type (SÏ) and the ratio of average velocity to reference velocity corresponding with a different εite condition, reεpectively. |
| 14. | The method as defined in claim 13, and further including the εtepε of: (i) εtoring a plurality of reference depth to reference velocity ratioε correεponding to different εite conditionε; (j) dividing the εite depth εignal by the εite velocity εignal to produce a site ratio signal; and (k) comparing the εite ratio signal with the stored reference ratio signals; (1) producing a velocity coefficient (CF) for the reference ratio that most nearly correεponds with the εite ratio; and (m) multiplying the measured velocity εignal by εaid produced velocity coefficient. |
| 15. | The method as defined in claim 12, and further including: (h) generating a plurality of εetε of reference depth (DR) εignalε correεponding to a plurality of flow signals (Qs) each of which is representative of flow for various flow rates at a εpecific site, respectively: (i) εtoring εaid sets reference depth and flow signalε in a signal comparison means; (j) meaεuring εite depth (Ds) at εaid εpecific εite location; and (k) comparing εaid εite depth with said stored depth to produce a corresponding flow εignal. |
SELF-CALIBRATING OPEN-CHANNEL FLOWMETER
Specification
Statement of the Invention
The subject invention relates to a self- calibrating open-channel flowmeter including means for determining the flow coefficient for a particular site by examining the sensed velocity of the flowing fluid versus the level at one or more flow rates. The flow coefficient is then used to convert local velocity to average velocity over a wide range of fluid levels at that site.
Brief Description of the Prior Art
The need to monitor water and waste water over the last 20 years has increased significantly due to environmental laws and the cost of maintaining adequate water and waste water facilities. The measurement of open channel flow especially in sanitary and storm sewers is very difficult.
For electronic flow eters that are placed in a standard section of sewer pipe, both the level and velocity must be measured accurately because both parameters vary as flow varies in such a piping system under open channel flow conditions. The slope and roughness of the pipe control the relationship between velocity and level. The slope (or grade) of the pipe
chosen for a particular site is the result of civil engineering considerations determined by such factors as the terrain, whereas the roughness factor is the result of the surface of the pipe and obstructions such as bends and elbows.
Historically, flow in open channel piping has been measured through the use of the Manning equation or other empirical relationships where the pipe roughness and the slope of the water surface is used to yield a predictable relationship between the sensed level and flow through the pipe. Because it is very difficult to accurately measure either the roughness or the slope of the water surface directly, this technique has not been used where accurate flow measurements are required.
More recently the preferred method of measuring flow in open channel sewers has been the velocity/area technique. This technique utilizes the continuity equation which states that the flow Q is equal to the product of the mean (average) velocity V and the area A of the partially filled pipe, both which are measured at a common cross-section.
Instruments using the velocity/area technique typically cannot directly measure the average velocity accurately enough for commercial use, and hence they typically contain a modifier that acts on the sensed velocity (local velocity) to better approximate the
average velocity. This is true when the sensed velocities are localized to the bottom of the conduit, the sensed velocity is at the surface itself, on any other location that is not a direct measure of the mean velocity.
For any particular pipe size, the slope and roughness of the pipe establishes a velocity and level relationship under open channel flow conditions. As the slope of the pipe increases, the velocities tend to be higher for a given level and roughness, and as the roughness of the pipe increases the velocity tends to be lower for a given slope and lever. ' â :--'* Thus, in general, however, the relationship between velocity and level are controlled by the slope and roughness of the pipe and the pipe diameter.
Because it is very difficult to measure the actual average velocity directly, a variety of techniques have been developed where local velocity is measured (i.e., a sensed velocity that is related but not equal to the average velocity) , and then the local velocity is modified so as to provide an accurate estimate of average velocity.
In the prior Marsh patent No. 4,083,246 (which is assigned to the same assignee as the present invention) , a fluid flow measuring instrument is disclosed in which flow is determined by measuring both the level and the local velocity directly. The
sensed local velocity is modified to approximate average velocity, and flow is then calculated utilizing the continuity equation A = V X A, where V is the approximate average velocity of the flowing fluid and A is the partially filled area of the pipe. In the aforementioned Marsh patent, the local velocity is measured near the bottom of the pipe. An appropriate modifier is determined by profiling the site with a portable velocity meter or by other meanε where the average velocity is determined at one or more flow rates, and subsequently through empirical equations contained in computer software the sensed velocity is transformed to an accurate approximation of the average velocity V over a wide range of levels (i.e. flow rates) .
Others have proposed to measure velocities over a large portion of a cross-section of a flowing stream as an approximation of the true average velocity, together with the use of modifiers to improve the accuracy of the instrument. Refer for example, to the U.S. patent No. 5,371,686 to Nabity et al. In the Petroff patent No. 5,198,989, a velocity is measured that is the maximum observed in the cross-section, and in the Bailey patent No. 5,315,880, the velocity at the surface is measured. (See also Alan Petroff/ADS Technical Paper) .
It has been discovered that if one analyzes the relationship between the local (sensed velocity) and the level for any particular pipe size and site location and compares these on-site relationships with those relationships obtained under controlled conditions, the modifiers necessary for converting the sensed or local velocity to average velocity can be predetermined without performing a separate or independent determination of the actual on site flow rate or average velocity.
It should be noted that a simple velocity and level relationship obtained under just one flow condition, while providing a reasonably accurate calibration coefficient, would not be as good as that obtained by looking at the relationship of velocity and level over a broader range of flow rates so as to better establish what true conditions exist due to the slope and roughness of that particular site. If, for instance, one looks at the velocity/level relationship over a period of time such as a week, the flow coefficient can be chosen based on the velocity/level relationship at a more narrow range of flow rates obtaining higher accuracy where most of the data is occurring for that particular site. It is important to note that such a self- calibrating device or process is capable of reacting to various conditions that could result from downstream blockages or from tidal effects. Such conditions could cause the water level to rise and the velocity to slow down creating a different
velocity/level relationship for that period of time where the unusual conditions exists. Separating these data sets obtained from this changed condition allows for different coefficients to be applied to each of the sets of different flow regimes.
Summarv of the Invention
Accordingly, a primary object of the present invention is to eliminate the necessity of independent site profiling, which often must take place at an inconvenient time, at inappropriate water levels or under difficult, confined space condition thereby making it possibly less accurate and less representative of the flow condition where most of the important date is concentrated. A more specific object of the invention, is to provide an improved flowmeter in which any local velocity measurement â be it near the bottom, at the surface, or anywhere in between and utilizing different velocity sensing means â electromagnetic acoustic, microwave, optical or other is modified so as to cause it to better approximate the average or mean velocity. The modifying means utilizes velocity/level relationships for any particular velocity measuring techniques that have been gathered under controlled conditions and in comparison to local site characteristics provides information to determine
what the multiplier factor relating the sensed velocity to the average velocity should be for a particular location (site) . This multiplier factor (or flow coefficient) could either be a constant or vary with other parameters such as depth.
A further object of the invention is to provide a self-calibrating flowmeter having a default flow coefficient. The self-calibrating flowmeter may be installed with the default flow coefficient, and once exposed to flow, the measured level and velocity data obtained on site is compared to a library of data stored in the memory to arrive at an improved flow coefficient using data stored in a computer or archived in tabular form. Other inputs needed to perform the. self- calibration include the channel shape and channel dimensions. Direct knowledge of the conduit ' slope or roughness is not required.
Brief Description of the Drawings Other objects and advantages of the invention will become apparent from a study of the following specification, when viewed in the light of the accompanying drawings, in which:
Fig. 1 is a graph of fluid level (or depth) plotted against flow velocity for a conduit having a low slope and a high degree of internal roughness, and Fig. 2 is an accompanying table of graph values;
Fig. 3 is a graph of fluid level plotted against flow velocity for a conduit of medium slope and medium roughness, and Fig. 4 is an accompanying table of values; Fig. 5 is a graph of fluid level plotted against flow velocity for a conduit of high slope and low roughness, and Fig. 6 is an accompanying table of values;
Fig. 7 is a graph illustrating the scatter plot of sensed velocity vs. level under the three flow conditions of Figs. 1, 3 and 5;
Fig. 8 illustrates three scatter plots of velocity/level curves representative of the flow coefficient required to convert the sensed velocity to average velocity; and
Fig. 9 is a diagrammatic illustration of a sewer system incorporating the flowmeter of the present invention;
Figs. 10-13 are diagrammatic illustrations of various flowmeter sensor arrangements according to the present invention;
Fig. 14 is a block diagram of a preferred embodiment of the flowmeter of the present invention;
Fig. 15 is a block diagram of the means for obtaining and storing the reference data in the storage means of Fig. 14;
Fig. 16 is a block diagram of the means for processing the sensed signals in Fig. 14;
Fig. 17 is a block diagram of an alternate embodiment of the flowmeter system of Fig. 14; and
Fig. 18 is a block diagram of a level-only modification of the present invention.
Detailed Description
Referring first, more particularly, to Figs. 1-6, flow in open channels will exhibit a pattern of velocities (a velocity profile) that is approximately logarithmic in shape. As flow increases, the liquid depth and the velocity at each point in the profile will increase, leading to a "family" of curves as shown in Figs. 1, 3 and 5. For a given pipe characteristics (slope, roughness, etc.), as the flow (and depth) increase, the velocity profile is amplified exponentially, giving higher velocities at a given location in the flow. In Fig. 1, each curve represents a velocity profile that exists for different liquid depths. Figures 1, 3 and 5 illustrate how the family of profile curves varies with pipe characteristics. Fig. 1 represents a pipe with low slope/high roughness, Fig. 3 a pipe with moderate slope/moderate roughness, and Fig. 5 represents a pipe with high slope/low roughness. As can be seen from Figs. 1, 3 and 5, a low slope/high roughness pipe will exhibit lower velocities throughout the profile for a given depth of
flow than a high slope/low roughness pipe will. For instance, at a flow depth of 10 inches, the velocity is only 2.5 ft/s at a position 7 inches above the channel bottom for the pipe characteristics represented in Fig. 1. But for the pipe represented in Fig. 3, under the εame depth conditions, the velocity at 7 inches above the channel bottom is 6ft/s because the pipe is steeper or smoother or both, various combinations of pipe slope and roughness are found in most open channel piping systems.
Referring now to Fig. 7, the relationship between pipe characteristics and scatter plot characteristics of the conduit conditions of Figs. 1, 3 and 5 are shown for corresponding electromagnetic velocity sensor locations one inch off the channel bottom, which corresponds to a typical location for an electromagnetic sensor positioned to measure a sensed local velocity in a sanitary sewer without collecting debris. A plot of the senεed velocity versus the flow depth (a scatter plot) for the family of profile curves shown in Figs. 1, 2 and 5 is shown in Fig. 7 depicting sensed velocity vs. depth.
Fig. 7 demonstrates that the characteristics of a scatter plot (such as εlope and offset) are related to the characteristics of the pipe (pipe slope, roughness) where a level/velocity sensor is installed.
The scatter plot for the profiles in Fig. 1 depicts a smaller change in velocity for an incremental change in level (ÎV/ÎL) than the scatter plot of the profiles of Fig. 5 which have a greater velocity change for the same incremental change in level. This behavior is directly related to the nature of the family of profile curves, which, in turn, is directly related to the characteristics of the pipe size/slope/roughness etc.) . Other characteristics of the scatter plot provide additional (but less dramatic) insight into the site characteristics, the curvature of a best fit curve through the scatter plot being one such characteristic.
Referring now to Fig. 8, the pipe characteristics that influence scatter plot shapes also influence which flow coefficient is required to convert the sensed velocity (as measured by a level/velocity meter) to the average velocity. More particularly, the velocity multipliers (average velocity divided by sensed velocity) for the profile curves given in
Figs., 1, 3 and 5 have been plotted versuε the flow depth. Currently, use is made of a single "flow coefficient" to define a particular set of velocity multiplier values for various depths of flow. For the examples given, the pipe represented in Fig. 1 requires a flow coefficient to 1.5, while the Fig. 3 profiles require a flow coefficient of 1.35,
while the Fig. 5 profiles require a flow coefficient of 1.2. The reason for the difference in flow coefficient is that the slope/roughness differences in the pipe affect velocity profile shapes. By examining the velocity/level scatter plot characteristics the flow coefficient can be surmised at a particular site because it is directly related to the pipe εlope/roughness aε well.
Therefore, one can examine the characteristics of a scatter plot generated from a particular meter installation site and make a determination of what flow coefficient should be used to accurately site calibrate the meter.
Referring now to the flow meaεuring system of the present invention illustrated in Fig. 9, a portable flowmeter installation includes a flowmeter 2 installed inside a manhole 4 εo as to measure the flow velocity of liquid 6 in a conduit 8. Manholes are normally installed where there is a change in direction, slope, or pipe size, or otherwise where there is a need to have acceεε to the flow. Figs. 10- 13 illustrate various examples of different sensors for measuring the velocity of liquid flow in a conduit. Velocity sensors are generally of the electromagnetic, acoustic Doppler, microwave doppler, laser Doppler ^ correlation or scintillation type. Level transducers normally include bubbler type pressure transducers, submerged pressure transducers,
εubmerged acoustic level transducers, and look-down acoustic, laser or microwave level transducers. Fig. 10 shows an electromagnetic velocity sensor 10 mounted on the bottom of conduit 1. Such bottom-mounted transducers are often secured in place by mounting bands (not shown) . Thiε electromagnetic velocity εenεor alεo containε in the εame houεing a εubmerged preεεure tranεducer 12. Theεe combinations of velocity and level transducers are connected to the electronic processing contained in flowmeter 2 unit via cable means 14.
Fig. 11 illuεtrateε a bottom-mounted Doppler velocity sensor 16 and a submerged pressure transducer 18. Again a cable 20 connects the sensors to the electronic proceεεing unit.
Fig. 12 εhowε a bottom-mounted Doppler εurface velocity sensor 22 and a εubmerged presεαre tranεducer 24. Again cable 26 connectε the sensor to the flowmeter electronic procesεing meanε. Fig. 13 illuεtrateε a look-down velocity sensor
28 and a look-down depth senεor 13 connected by cable
32 to the electronic proceεεing module.
Referring now to the block diagram of Fig. 14, the signal processing of the depth and velocity signals includes the improved processing means of the present invention. The outputs of depth sensor 46 and velocity sensor 48 are converted by signal procesεorε 50 and 52 to the εite depth εignal Dε and the εite
velocity signal Vs, respectively. By use of the microprocesεor 54 with RAM memory, correεponding depth and velocity εignals are εtored aε pairε of data. Aε will be deεcribed in greater detail below, contained in the more permanent read-only (ROM) memory meanε 56 are previouεly obtained pairε of velocity and level data that have been collected under controlled conditionε in either a flow laboratory or in a carefully conducted field test. Data sets have been obtained for all pipe sizes expected to be seen by the device in the field under various εlope and roughnesε combinationε. Alεo contained in memory 56 are εite flow coefficientε that correεpond to the level dependent multiplier and which serve to convert the sensed velocity to mean velocity once the particular flow condition at the field has been recognized.
The flow condition at the field is recognized by comparison means 58 which includes a microprocessor that compares data contained in RAM 54 (particularly the site velocity V s and the depth velocity D s ) . The pipe size εignal P s from manual source 53 iε εupplied to compariεon meanε 58 via encoder 57. By obεerving the εcatter plot that exiεtε between εite velocity and εite depth for that particular pipe size, comparison meanε 58 εearcheε in memory meanε 56 to find an equivalent velocity/level relationship for the particular pipe size and that particular velocity measuring device (i.e., bottom-mounted electro¬ magnetic, bottom-mounted Doppler, surface-mounted
velocity εenεor, and the like) . For each type of velocity measuring device there must be stored the reference velocity and levelε and correεponding εite calibration coefficient C F . Once the correεponding curves have been matched, then the corresponding site calibration coefficient C F (which iε generally depth dependent) iε εupplied to the εite velocity modifying means 62 together with the depth signal D s .
The εite calibration coefficient C F and the depth signal D s are εupplied to the site velocity modifying means 62, whereupon the depth-modified coefficient C F ' iε εupplied to first multiplier meanε 64 together with the εenεed velocity V s , thereby to produce the average velocity εignal V which is supplied as one input to third multiplier means 66. To the other input of the third multiplier meanε 66 is applied an area signal A s which is the product of pipe size signal V. produced by second multiplier means 67 from manually-set pipe size signal generator 53 and pipe size encoder 57, and the site depth signal D s . The deεired flow signal Q s is produced at the output of the third multiplier means 66.
Referring now to Fig. 15, in order to obtain the reference data εtored in ROM data memory means 56, through a combination of mathematical modeling and actual testing of velocity and level sensors, under a variety of flow conditions a means of collecting
theεe various relationshipε iε achieved. In particular, inputε to a data acquisition memory consist of five different inputs (i.e., the average velocity obtained from the reference standard, the senεor type, the sensed velocity, the sensed depth, and the pipe size corresponding to each of the previouε inputε) . For each input the data iε run on each pipe εize, or on a range of pipe sizes, and are recorded on a ROM. The reference depth and reference velocity and the ratio R of average velocity to reference velocity at each depth are recorded in the ROM along with the corresponding pipe size and εensor type by meanε of which the meaεurementε were obtained. Of course, if only one sensor type is used, then the senεor type factor can be eliminated. Typically, not every condition is tested, but through mathematical modeling and interpolation a good estimation for the intermediate sizeε can be achieved and the data either εtored directly in ROM or, alternately an equation depicting the velocity and depth relationships stored. Referring now to Fig. 16, the processing of the εite depth εignal D s and the site velocity signal V s are processed together with pipe εize signal P s , and the senεor type εignal produced by manually-εet εensor type signal generator means 59 and senεor type encoder 61. Theεe εignalε are processed by microprocessor 54 and a corresponding RAM (Randon Acceεε Memory) and ROM. The local εenεed velocity and depth εignalε are digitized and εtored in RAM 54 along with information regarding the pipe size and the sensor type for that
particular site. Inputted to the comparison means 58 is the sensed velocity, sensed depth pipe size conduit εize and εensor type information. The comparison meanε εearches the ROM 56 so as to find that stored data that is pertinent for thiε εensor type. The next εtep iε that from the εtored data for the sensor type the stored data correεponding to the εite conduit εize iε located. From the stored information pertinent to both the pipe εize and the sensor type, the V s and D s of the site iε compared to the V R and D R in the appropriate memory section εo as to find the best fit V R and D R that match the data pair V s and D s . For each matching pair of data points, the correεponding flow coefficient C F iε choεen and subsequently provided to the εite velocity modifying meanε for multiplying by the εenεed local velocity V s .
An alternate meanε of modifying the sensed velocity utilizing velocity/level relationships is indicated in Fig. 17. In thiε alternate embodiment the depth signal D s is divided by the site velocity εignal V s via division module 153 producing at its output the ratio Ds Vs. The level/velocity relationship represents the level/velocity at a particular flow condition. Comparison means 158 is a microprocesεor that combineε thiε ratio of depth to velocity D s /V s and the depth D s and pipe εize P s to εearch for an equivalent pair of reference ratioε DR/VR. Once the εet of reference ratioε have been found in memory 156, the calibration coefficient C F for thiε particular εite condition iε proceεsed similar to that of Fig. 14.
Because of expense, quite often users do not wish to install a velocity/level flowmeter at all εiteε. In particular a "level only" type flowmeter will work as long as the velocity and level relationship are stable and repeatable for a particular site. In such caεeε the uεer may uεe a weir or flume which have predictable level-to-flow rate relationεhips or alternatively, by performing extensive site profiling to determine the level-to-flow rate relationship of a standard piece of conduit under variouε flow conditionε, and if εuch conditions are εtable one can utilize thiε information to obtain flow from just a level reading itself. However, such site profiling at different levels (flow rateε) is both time consuming and posεibly dangerous due to confined space entry conditionε. The preεent invention tendε itself to provide a much easier and more accurate means of
-I ', establiεhing the level of flow relationship in a standard conduit. Referring now to Fig. 18, the flow rate Q s1 at a particular site has been obtained by the instant invention and εtored along with the depth D s1 at the εite for each flow rate.
In other wordε pairε of data pointε Qε 1 and Ds 1 that were measured with the instant invention are stored in an electrical erasable prom EEPROM 260. The information in this EEPROM can either be copied electrically, or the EEPROM phyεically removed and
inεtalled in a second flowmeter. Thiε second flowmeter does not have a velocity εenεor but only haε a level or depth εenεor. The depth εenεor does not necessarily need to be of the same type as waε uεed to collect the original depth and flow rate data. Thiε depth signal D s2 is analyzed by a microprocessor in this level-only flowmeter. For every depth εignal D s2 , the memory meanε 260 iε εearched for an equivalent depth signal D sl , and once this is found, the flow corresponding to D s1 is read out of memory 260 and is outputted aε Q s2 .
The average velocity, the level and the εubsequent flow rates can either be stored for later uεe, or be channeled to an indicator for inεtant read out. It iε important to note that the εet of stored relationshipε in memory 56, 156 can be a computer program that compareε the velocity/level data that iε recorded over some period of time, or just one velocity/level part itself so as to provide the correction factor. In any event, the εtored general relationship of memory 56, 156, which have been theoretically or experimentally derived, are compared with the local velocity and depth relationship at any particular site εo aε to determine a correction factor that which iε tailored to that εite εo aε to modify the sense velocity to be actually an approximation of average velocity under thoεe εite conditions.