GB2117913A - Vector KVA meter - Google Patents

Abstract

A meter apparatus for determining Vector VA demand, and other electrical quantities, over a predetermined time period, uses a watthour meter and a Q-hour meter with pulse initiators. The time period is divided into a plurality of sub- intervals and the VAh or other quantity is determined for each sub- interval DELTA t. Thus DELTA VAh is determined using the relationship <IMAGE> where DELTA Wh is the output from the watthour meter and DELTA VARh is a value for DELTA VAR hours determined generally from the relationship <IMAGE> where Qh represents Q-hours and phi represents an angle by which the voltage vectors are rotated for measurement by a Q meter of Q- hours.The values of DELTA VAh for the sub- intervals in the time period are summed to provide total VAh for the time period, and from which can be calculated the Vector VA demand for the time period. By making the determination based on relatively short subintervals, the inaccuracies introduced by changing power factor over the time period are reduced or eliminated for all practical purposes. <IMAGE>

Description

SPECIFICATION Improved vector KVA meter This invention relates to meters for measuring Vector volt ampere demand or more commonly kilo volt ampere (KVA) demand and other electrical quantities.

The measurement of Vector KVA demand is of importance in the evaluation of energy consumption for billing to accommodate the billing to loads which fluctuate and may use large amounts of energy for short periods. It is also useful in assessing the heating of generating, transmitting and distributing equipment. The apparatus may also be of value for determining load power factor. It may also be used for determining other electrical values.

In the following description the expressions Vector VA and Vector KVA or watthour and kilowatt hour and similar expressions, may be used. It will be understood that the expressions are similar except for the factor of 1000 and both forms may be used when describing apparatus. The alternate forms will be recognized by those skilled in the art, and the meaning understood.

In the past a common way to calculate Vector KVA demand was by measuring watts (W) demand and reactive volt amperes (VAR) demand and using the relationship

Where in a three phase system having phases designated A, B and C (t) =WA(t) + wB(t) + WC(t) (2) 5 VAR(t)= VARA(t) + VARB(t) + vARc(t) (3) and

It can be demonstrated that the relationship (1) provides a correct determination of Vector VA demand only if the power factor and the reactive factor do not vary over the demand interval.

For example:

at any instant of time t.

However, Vector VA demand is measured over a time interval or time period which we can call t2-t1. The value of Wilt) and the value of VAR(t) will normally vary over the time period t2-t1 and they will not vary in the same manner over this time period since Wit)=VAlt) Cos oIt) (6) and VAR(t,=VA(t, Sin (t) (7) Thus, for the time period t2-t1, the Vector VA demand is given by:

Now to determine under fhat conditions the equations (8/ and (1/ are equal. Vector VA demand=VA

Since cos2 Olt+sin2 0it=1 we have equality.

Thus the equations are equal only if Cos O (the Power factor) and Sin 0tuts (the Reactive factor) are constant over the time period during which Vector VA demand is being measured. It follows that for an accurate determination of demand by the method of equation (1), the power and reactive factors must be constant over the time interval or time period of the determination. Inasmuch as a time period used in practice could be a 1 5 minute interval or greater, then it is unlikely the power and reactive factor will remain constant over the interval and the result will not be accurate.

In summary thus far, by using W demand and VAR demand to determine Vector VA demand, accurate results are obtained only if the power and reactive factor do not change significantly over the time interval of the measurement.

Another quantity or value which is useful in the measurement of electrical energy in a polyphase circuit is the quantity Q. The quantity Q is defined as Q=E.I. Cos ) (9) where the voltages in a system are rotated by an angle 0 (which is not zero or not 900). Further information on measuring the quantity Q or on "Q metering" can be found, for example, in an article in "Distribution", January 1970, by C. E. Gambell.

Another relationship, which can be developed from equation (9) is Q W VAR - (10) Sin) TanX The relationship of equation (10) is general. A more convenient relationship is available when ss=60 lagging.The quantity 0 is customarily measured by a O-hour meter (i.e., O-hour or in abbreviated form Qh) and it is possible to connect a watthour meter cross phase in a three phase system to provide an angle 0=60 . If this is done, the relationship of equation (10) becomes:

where Qh is Q-hours or the output of a watthour meter type of instrument connected as explained above to read Hours and having an angle 0=600, Wh is watthours as obtained from a watthour meter, and VARh is VAR hours.

The present invention makes use of the relationship of equation (12). In particular, in a preferred form, it makes use of relationship as represented generally by equation (12) with (8). Thus, the present invention in its preferred form, determines demand, with reference to a time period which we will now define as T2-T0, by using the output of a watthour meter, Wh, and the output of a O-hour meter, Oh. In other words, in a preferred form, a determination of demand is obtained by the relationship below (equations (13) and (14) are modified form of (8) and (12)).

where N is the number of subintervals in the time interval of the measurement, T2-T0

where the determination of equation (14) is used in equation (1 3).

It should be noted that, as with the previously discussed arrangement for accurate results, the power factor must be constant over the time interval during which demand is measured. The present invention achieves the desired accuracy by making measurements over very short time intervals. If the time intervals are made quite small, the accuracy can be improved as the power and reactive factors will be substantially constant over small time intervals. However, it is difficult to make the time intervals small and obtain accurate results. The present invention provides means for doing this as will subsequently be described.

In accordance with the invention, as use is made of a watthour meter, then Wh or watthours can be determined by summing the Wh determined for each subinterval. Thus, Similarly watts W can be determined as follows:

The quantity VARh or VAR hours is available as set forth in equation (14)

with 2A0h-AWh > 0 for inductive VARh and 2AQhAWh < O for capacitive VARh If AVARh for the subintervals in a given time period T2-T0 are summed, then an output representing VAR demand can be derived according to the following relationship:

The VAR demand can be calculated for total VAR (i.e., the sum of inductive VAR and capactive VAR), or for capacitive or inductive VAR only.

A distinction should be noted between present demand, peak demand, and accumulated demand.

Present demand is the demand at any particular time, for example, in any time interval. Peak demand is the highest demand value obtained in a predetermined period of time. Accumulated demand or cumulative demand is a running total of all peak demands recorded for any desired billing period. While peak demand is the value normally required and which is a value obtainable with the present apparatus, the other values may be provided where they are of use.

Also, it is possible to determine a value for AVAh, volt-ampere hours, for the energy provided to the system in each of the subintervals according to the relationship:

By summing the values for the subintervals, VAh can be determined

It will be apparent that values can also be obtained for, instantaneous power factor, average power factor over the current time period, average power factor for the billing period, and average power factor for the particular time periods in which occurred a peak demand (i.e., peak demand KW, KVA or KVAR).

Thus power factor for a subinterval is the ratio for the subinterval of KW/KVA. Power factor for a subinterval is, for all practical purposes, instantaneous power factor.

Average power factor in a timer period is the ratio of KW to KVA demand for that time period.

Similarly average power factor in a billing period may be calculated as the ratio of KWh to KVAh for the billing period.

Average power factor can, of course, be stored for the time periods in which occurred the peak demands of KW, KVA or KVAR (total, capacitive or inductive).

It is therefore a feature of the present invention to provide an improved Vector KVA demand meter.

It is a feature of the present invention to provide a Vector KVA meter based on values obtained from a watthour meter and a O-hour meter.

It is another feature of the present invention to provide accurate measurements of KWh, KVAh, KQh and KVARh (total, capacitive and inductive), KW, KVA and KVAR demand (total, capacitive and inductive) from 900 lagging to 300 leading power factor from a watthour meter and a Q-hour meter.

It is yet another feature of the present invention to provide time of use metering capabilities of the KWh, KVAh, KVARh (capacitive and inductive), KW, KVA and KVAR demand quantities with the same apparatus.

It is another feature of the invention to provide apparatus which determines values for subintervals and sums them over required time periods to obtain values for electrical quantities including instantaneous power factor, average power factor for any time period, average power factor for a billing period and average power factor for a time period in which a peak demand occurred.

Therefore, in accordance with one of the basic forms of the invention there is provided apparatus for determining an electrical value related to consumption of electrical energy over a predetermined time period, comprising means to provide a pulse train having a pulse frequency related to said consumption, means to divide said time period into a number of consecutive sampling periods, means to define a number of consecutive subintervals each terminating within a respective sampling period, means to determine from the number of pulses occurring in each subinterval said electrical value for each said subinterval, and means for summing said electrical values determined for each said subinterval over said time period to provide an output representing said electrical value for said time period.

The present invention will be described with reference to the accompanying drawings, in which Figures 1 and 2 are diagrams of the pulse output of a watthour meter and a O-hour meter, useful in explaining the invention, and Figure 3 is a simplified block diagram outlining the invention.

Referring to Figure 1, the diagram 10 represents the pulse output of a watthour meter as the meter disc rotates. The rotation of the disc past a sensor gives a series of pulses 14 shown along a time base. Similarly the diagram 11 represents the pulse output of a Q-hour meter as the meter disc rotates, and the pulses 1 5 are representative of the disc rotation. It is known to derive pulse trains from rotating discs of meters, for example, for use in accumulating counts for remote reading systems.

The pulse trains represented by pulses 14 and by pulses 1 5 are asynchronous as is normally obtained from a watthour meter and a O-hour meter under a varying load condition. It was previously indicated that accuracy could be increased by shortening the time interval. For example, a time interval could be divided into equal subintervals which could be evaluated and the evaluation results summed.

The shortening of the time interval, i.e., the selection of a subinterval is, of course, limited to the smallest time difference possible between two pulses coming from the same meter (i.e., the O-hour meter or the watthour meter when running at full load). This smallest possible subinterval is in turn limited by the number of pulses that can be provided in one turn of the disc. In addition, when the pulses are asynchronous, it is not possible to choose a constant short time interval where the pulses in the interval will always be a whole number of pulses. If this was possible the accuracy would depend on physicai mislocations of pulse-producing holes on the meter disc and changing power factor. Thus it is desirable to be able to predict fractional pulses.This is of particular importance at light load when the power factor changes rapidly or where the power factor approaches 900 lagging or 300 leading. It would also be desirable for maximum accuracy to avoid estimating fractional pulses in each subinterval and for both meters (i.e., watthour meter and Q-hour meter) as this is a source of error. The arrangement provided by the present invention minimizes such errors.

In Figure 1 , the integration period or time period T2-T0 is divided into equal time intervals Bx. The time interval Bx or sampling period Bx is the sampling period in which a computation is to be made. It extends between boundaries (i.e., between times) 1 6 and 1 7. Boundary line 1 7 and 1 6 are also designated as time Tx and Tx~,. In like manner the sampling period Bx~, represents the time period in which the previous computation was done. The sampling period Bxi is defined by boundaries or times 18 and 1 6. The sampling period Bx+, is the period following Bx.

The last pulse generated in the current sampling period is pulse 20. The last pulse 20 could be from either the watthour meter or the Q-hour meter. In this instance the last pulse 20 is from the watthour meter and it is at time tA. The last pulse in the sampling period Bx from the other meter, in this case the Q-hour meter, is pulse 21 which occurs at time TBT also boundary line 22. The line 22 defines one boundary of a time interval (i.e., a subinterval) used in the computation.

In the preceding time period Bx~, the last pulse was pulse 23 occurring at time TK which could be from either meter but is shown as coming from the watthour meter, and the last pulse from the other meter is pulse 24. Pulse 24 occurs at time TD also shown as time boundary line 25. The lines 25 and 22 define a time interval or subinterval TBTD which is of interest. Other time intervals are of interest.

Pulses 20 and 26, 26 and 27, define time intervals T-Tc and T,-T,, and pulses 21 and 28,28 and 29 define time intervals TBTH and THTG. Pulses 30 and 31 define a time interval TLTM.

These five time intervals are of interest as will be evident later. The apparatus tracks the times at which the last 3 pulses from the watthour meter and O-hour meter occurred, and the times at which pulses 30 and 31 occurred in relation to pulse 21. In other words, as pulses are produced the apparatus will always remember the time at which have occurred pulses 20, 26, 27, 30 and 31 for the watthour meter and pulses 21,28 and 29 from the Q-hour meter.

Of course, at the beginning of the first sampling (i.e., sampling period B1) or time To the values of TA, Tc, TF, TBT THT T5 and TD, T, and TM are zero and as pulses are produced they will become defined.

The apparatus also assumes a computation has been executed at time T,.

The two pulse trains comprising pulses 14 and 1 5 are received by computational apparatus according to the invention, and at the end of the current sampling period Bx, that is at time Tx (boundary line 17) the computational apparatus does the following: a) It checks to see if both meters have generated at least one pulse in the time interval TxTx~a. If they have not, it checks at the end of each of the subsequent sampling periods till both meters have generated at least one pulse in the time interval TX+nTx~, where TX+N represents the end of sampling period 8x+N.

b) The meter which generated the last pulse before time boundary 1 7 is identified. In this instance, the watthour meter generated pulse 20 at time TA. Once this pulse has been identified, the apparatus also identifies pulse 21 at time T8 (i.e., the last pulse of the other meter). The apparatus then checks to see if the meter that generated pulse 20 (the last pulse) also generated at least one pulse after the end of the previous integration period and before pulse 21. If this condition is not met, the apparatus waits until it is met before executing a computation as in step (a). This ensures, as will be apparent hereinafter, that after an integration period or time period ends the next calculation will not result in a negative value. This can only occur at the first computation. If the condition is met, it identifies pulses 30 and 31 which occurred at times T, and TM, respectively. Pulse 30 is the pulse from the watthour meter which follows the pulse 21 from the Q-hour meter, and pulse 31 is the pulse from the watthour meter which precedes pulse 21 from the O-hour meter. It is possible to identify these pulses because the apparatus always tracks the times at which occurred the last pulse from each meter.

Furthermore, each time a meter sends a pulse to the apparatus, it stores the time at which occurred the preceding pulse from the other meter so as to be able to identify time TM. Moreover, when receiving one pulse from the said meter, the apparatus also verifies if it is the first pulse received from this meter since the other meter sent its last pulse. If so, the time at which the present pulse occurs is identified at least temporarily as time T, in case no further pulses would be received until the end of the integration period. The apparatus also tracks the times at which occurred the last three pulses from each meter because these will be used for the last computation of readings in the integration period or time period as will be explained in steps g) through j).

c) The apparatus determines Nc the number of pulses in the time interval TBTD generated by the watthour meter. The determination of Nc is as follows: NC=NA1 +P3+P, (15) where NA=Whole number of pulses accumulated in the time interval TBTD, and where

The value AP3 represents a fraction of pulse for the meter that generated pulse 20, and where Asp1, is known from the previous calculation and represents a fraction of pulse carried over from the preceding subinterval.When a demand value (i.e., W., VA or VAR) is calculated for a predetermined time period T2-T0 using many subintervals as is usually the case, and the maximum value retained for billing purposes, the value of AP, can be added to the subsequent time period or be made zero at the beginning of each predetermined time period or time T,. Of course, for cumulative quantities like VA hour or VAR hour, the value of AP, cannot be made zero.

d) The apparatus determines ND the number of pulses generated by the other meter, that is, by the Qh meter that generated pulse 21. The determination of ND is as follows: ND=NB+AP4+AP21 (17) where NB=whole number of pulses accumulated from the pulse train in the time interval defined by TD to TB (a pulse coincident with TD is always counted, and a pulse coincident with TB is not counted), where:

AP4 represents a fraction of a pulse calculated for the meter that generated pulse 21 (It will be apparent that this is included for completeness of description and that no calculation is necessary to determine "1".The value is "1" because the subinterval is assumed to terminate just before the pulse). and where AP2 is known from the previous calculation and represents a fraction of pulse carried over from the preceding subinterval. In this instance AP2=0 and P4=1. At the beginning of the first sampling period or time T,, the value of AP2 can be set to zero or retain its value as was said for AP1.

e) The value of TD is made equal to TB and stored for use in the subsequent computation.

f) The apparatus calculates AWh, AW, AVARh, AVAR, AVAh and AVA for the subinterval based on the following relationships:

where K is a pulse constant for both meters and where Nw and NQ have for value either NC or ND depending on the meter which generated the last pulse in the sampling period TxTx~,.

For continuous integrated quantities like VAR hour, volt ampere hours and watthour, the calculated values are added to existing values of VARh, VAh and Wh. For demand quantities like VAR, Watt and VA, the current calculations are added to the accumulation of the other subintervals from the beginning of a predetermined period starting at time T,. The new values so obtained are then compared to maximum values obtained in previous predetermined periods. In cases where they are greater, the registers are updated to the new maximum demand values.

The steps (a) through (f) are repeated until the end of the predetermined time period is approached.

For the calculation and summing of demand values, we have used a subinterval TBTD that is not identical to a sampling period. While neglecting the partial subinterval at the end of the final sampling period may provide sufficient accuracy for some requirements, it is not a constant interval and is not equal to the sampling period Bx. It is preferable to account for this at the end of the final sampling period, i.e., at the end of a predetermined time period over which the various outputs are to be derived.

Referring to Figure 2, time boundary line 30 represents time T2 the end of the predetermined time period and time boundary 31 with time boundary 30 defines the last sampling period Bw. Time boundary line 32 with time boundary line 31, defines the next to last sampling period BN-1- The time TÁ is the time the last pulse 33 is generated in sampling period period BN. The time Tc is the time of the preceding pulse 34 generated by the same meter, and the time Tr is the time of the pulse 35 preceding that, i.e., for the third last pulse of the same meter.The time TB is the time of the last pulse 39 of the other meter; the time TH is the time of the preceding pulse 38 generated by the same meter; and the time T5, is the time of the next preceding pulse 37 (i.e., the third last pulse) generated by the same meter.

The time TD (pulse 36) is the time at which was executed the calculation in the sampling period Bn~1. In the sampling period BN two calculations have to be executed. The apparatus executes a computation for the subinterval TBTD according to the procedure described previously. This is for the final subinterval, that is the final complete subinterval for the final sampling period. There remains a partial subinterval T2-T5, which is between boundary 30 and pulse 39.

The apparatus now does the following: g) The apparatus determines the minor or partial subinterval T2-TB' which extends to time boundary line 30, i.e., time T2. The time T2 is the end time in the initial time period T2-T0.

h) The apparatus calculates NC the number of pulses generated by the meter that generated the last pulse, i.e., the pulse 33 at time T,,. The determination of NC as follows:

where NÁ is the whole number of pulses occurring in the time interval between the end T2 of the final sampling period and the end of the last subinterval, for the pulse train which had the last pulse in the time period or integration period, i.e., the pulses accumulated in the time interval T2-TB' by the meter that generated the last pulse, where

This is the time interval between the end of the final sampling period and the last pulse, multiplied by the time interval between the second last and the third last pulses for the pulse train which had the last pulse, divided by the square of the time interval between the last pulse and the second last pulse for the pulse train which had the last pulse and where AP1, is known from the previous calculation as before. It will be seen that this determination is similar to that described in (c) above, except that the value of AP3 is changed to that for åP3. If åP3 > 1 then åP3=1.

i) The apparatus determines ND the total number of pulses generated between the end of the last subinterval and the end of the final sampling period for the pulse train which does not have the last pulse, that is, from the other meter or Oh meter. The determination of ND is as follows::

where, as before, NB is the whole number of pulses accumulated by the other meter in the time interval T2TB (the number of whole pulses is zero as the pulse coincident with TB is not counted),

that is, the time interval between the end of the final sampling period and the end of the last subinterval multiplied by the time interval between the second last and the third last pulses for the pulse train which did not have the last pulse, divided by the square of the time interval between the last and the second last pulse of the pulse train which did not have the last pulse, and where AP2,, is known from the preceding calculation, as before.It will be seen that this determination is somewhat similar to (d) above, except that the value of AP4, is changed to that for AP4,. If P4, > 1, then AP4,=1.

j) The apparatus now calculates AWh, AW, AVARh, AVAR, AVAh and AVA using the new values of NC and ND exactly as outlined in (f) above. As was said before, the calculated values in (j) are added to existing values of Varhour, VA hour and watthour. For demand quantities, the calculated values in (j) are added to the accumulation of respective values for the other subintervals. The new demand values of W, VAR and VA so obtained are compared to the stored maximum values of the same quantities and the stored maximum values updated if the new values are larger.As the value for VA (and other electrical values) is calculated for each of a number of short subintervals, error caused by changing power factor is minimized. If the subintervals are short, the power factor will be substantially constant over the subinterval and error caused by power factor change is negligible for practical purposes.

For W and VAR demand the error is almost nil since the only cause of error is when estimating AP; and AP47.

The usual practice is, as those familiar with metering demand knows; to calculate or measure demand values over many predetermined time periods of 1 5 minutes, or 30 minutes or 60 minutes, for example, for a period of one month, and to bill the customer based on the maximum demand attained in that period. Since demand is calculated over many predetermined time periods the values åP,, and AP2 can be added to the subsequent time period or be made zero at the beginning of each predetermined time period T,. Of course, for continuous integrated values (e.g., VARhour or VAhour), the values of åP, and AP2 cannot be made zero and are always saved. The calculated values of these last two quantities is never in error.

It will be apparent that power factor information can be derived from the stored values for KW, KVA, KWh and KVAh. As was previously described, the instantaneous power factor can be determined from the values of KW and KVA for a subinterval as the ratio of KW/KVA; average power factor for a time period can be determined from KW and KVA for that time period as the ratio of KW to KVA demand for that time period; and average power factor for a billing period may be determined from KWh and KVAh for the billing period as the ratio of KWh to KVAh.

Referring now to Figure 3, there is shown a simplified block diagram of apparatus according to the invention. The apparatus has been shown in a polyphase three phase, 4 wire system.

In Figure 3 there is shown a watthour meter 35 and a O-hour meter 36 is connected to a power system represented by lines 37, 38, 39 and 40 to receive the required voltages and current inputs. The meter 35 and pulse initiator 41 provide pulse output 10 (Figure 1). The meter 36 and pulse initiator 42 provide pulse output 11 (Figure 1).

The pulse initiators 41 and 42 are connected to pulse process circuitry 44 and 45 respectively, with input ports 43 being connected to both initiators 41 and 42. When a pulse is received by pulse process circuitry 44 or 45, it is shaped into a square pulse and latched or retained. The same pulse is also fed to block circuitry 43. The output from pulse process circuitry 44 and 45 is applied to a central processing unit 46. The output from input ports 43 is also applied to processing unit 46. The central processing unit accepts pulses only when it receives a signal from either circuitry 44 or 45 and the same signal is still present from port 43 after a small time-delay. This helps to differentiate between valid and spurious pulses.

The central processing unit or CPU 46 does all the calculations. The read-only-memory or ROM 55 is connected to CPU 46 and it contains all the machine code instructions necessary for the CPU 46 to process all the data and make the desired calculations. A chip selector 51 is connected with CPU 46 to select the various chips, i.e., to select the various circuitry utilized by the CPU. A time base generator 48 is connected to CPU 46 and provides a time base or clock for the CPU. Also connected with CPU 46 is a watchdog timer or faulty operation mode detector 47. While CPU 46 is operating properly, in accordance with internal tests, it sends regular periodic signals to the faulty operation detector 47 which in effect retriggers or resets the detector 47. The detector 47 has a time interval longer than the time between the periodic reset signals it receives from CPU 46.If a periodic signal is not received by detector 47 in its predetermined time interval, it begins a re-initializing routine in the CPU which starts it running again. Such watchdog circuits are well known.

In addition to ROM 55, a random access memory or RAM 54 is connected with CPU 46 via chip selector 51, and a non-volatile random access memory or NVRAM 50 is connected to CPU 46 via a NVRAM select logic circuit 49. The random access memory 54 is used to store and recall data as required by CPU 46 while data is being processed. The non-volatile memory 50 is used to store and retain pertinent data during a power failure or an interruption by the watchdog circuit (faulty operation mode detector 47). The NVRAM 50 is write protected, and each time the central processing unit addresses this memory it first must go through a sequence which disables the write protecting feature.

Thus, a signal on conductor 60 disables the write protect feature, and a signal on conductor 61 informs NVRAM select protect logic 49 that the non-volatile RAM 50 has been selected. In turn, the NVRAM select protect logic 49 applies a select signal over conductor 62 permitting the reading of data from or the writing of data to NVRAM 50 over conductor 63.

The output comprises output ports 52 selected by chip selector 51 and the output is to peripherals 53.

Input may be provided from input peripherals 56 over conductor 64 when required. One example is to reset the peak demand register to zero.

The operation will be apparent from the preceding description. Very briefly, pulses from pulse initiators 41 and 42, representing the operation of the watthour meter 35 and O-hour meter 36 respectively, are received by the central processing unit 46. The pulse count is stored in RAM 54 and at the end of a sampling period (e.g., time period Bx of Figure 1) the pulses necessary for defining a subinterval and for making the necessary calculations are identified. This is in accordance with instructions stored in ROM 55. The calculations are all made as set forth in (a) to (f) of the preceding description. From the number of pulses received from the clock 48, the CPU identifies the end of a time period and then makes the calculations as set forth in (g) to (j) of the preceding description.The new demand values calculated for the time period are compared to previous demand values stored in NVRAM 50 and, if greater, they replace the stored values. The other values, that is, the continuous values, are added to the respective stored values to provide a total quantity and these totals may be stored in NVRAM 50 or fed to an output peripheral 53.

The meter of the invention has provision for storing the desired values until the stored amounts can be transferred to a reading device which can be plugged into the meter to transfer the stored amounts to the reading device and reset the memory. Such reading devices are known. It is to be noted that only watt-hours, Q-hours or VAR-hours (total, inductive or capacitive) and VA-hours may be output continuously to an output peripheral 53 which, for example, could be a general purpose computer or even display devices.

It is believed that the operation will be clear from the preceding description.

Claims (23)

Claims

1. Apparatus for determining an electrical value related to consumption of electrical energy over a predetermined time period, comprising means to provide a pulse train having a pulse frequency related to said consumption, means to divide said time period into a number of consecutive sampling periods, means to define a number of consecutive subintervals each terminating within a sampling period, means to determine, from the number of pulses occurring in each subinterval said electrical value, and means for using said electrical value determined for each said subinterval to provide an output representing said electrical value for said time period.

2. Apparatus for determining an electrical value related to power consumption over a predetermined time period, comprising means to provide a pulse train having a pulse frequency related to power consumption, means to divide said time period into a number of consecutive sampling periods, means to define a number of consecutive subintervals each terminating within a sampling period, means in the final sampling period of said time period to define a partial subinterval extending from the end of the last subinterval to the end of said final sampling period, means to determine from the number of pulses occurring in the subintervals and the said partial subinterval, said electrical value, and means for summing said electrical value determined for each said subinterval and said partial interval to provide an output representing said electrical value for said time period.

3. Apparatus as defined in claim 2 in which said subintervals are variable.

4. Apparatus for determining an electrical value related to power consumption over a predetermined time period and based on a first and a second type of meter, comprising means to provide a first and a second pulse train each having a number of pulses related to the meter output, means to divide said time period into a number of consecutive sampling periods.

means for identifying the last pulse in each said sampling period for each said first and second pulse trian and for defining a subinterval terminating with the earlier occurring of the last pulse for each said first and second pulse train in a sampling period and beginning at the termination of the subinterval in a preceding sampling period, means for determining for each said first and second pulse train, the number of pulses in each subinterval, means in the final sampling period of said predetermined time period to define a partial subinterval extending from the end of the last subinterval to the end of said final sampling period, means for determining for each said first and second pulse train, the number of pulses in each partial subinterval, means for receiving the count of the number of pulses for each said subinterval and said partial subinterval for each said first and second pulse train, and for determining for each said subinterval and partial subinterval said electrical value, and means'for using the electrical value for each said subinterval and partial subinterval to provide an output representing said electrical value for said predetermined time period.

5. Apparatus as defined in claim 4 in which the first meter is a watthour meter and the second meter is a O-hour meter.

6. Apparatus as defined in claim 5 in which the electrical value is VA demand.

7. Apparatus as defined in claim 6 in which said time period is defined by T2-T0, the subinterval is defined by t2-t1 and in which the means for determining for each said subinterval and partial subinterval the value for VA demand operates according to the relationship

where Wh is the value of watthours and VARh is the value for VAR hours.

8. Apparatus as defined in claim 7 in which the O-hour meter is connected to have the voltage inputs 600 lagging and the value for VAR hours is determined from the relationship

9. Apparatus as defined in claim 5 in which the electrical value is VA hours (VAh).

10. Apparatus as defined in claim 9 in which said subinterval is defined by t2-t1, in which the Qhour meter is connected to have voltage inputs 600 lagging and the value for VAR hours is determined by the relationship

and in which the means for determining for each subinterval and partial subinterval the value for VAh is in accordance with the relationship

11. Apparatus as defined in claim 5 in which the electrical value is power factor.

12. Apparatus as defined in claim 11 in which a subinterval is defined by t2-t1, in which the means for determining for each subinterval and partial subinterval includes means for determining VA demand for the subinterval according to the relationship

where Wh is the value of watthours and VARh is the value for VAR hours, in which the means for determining for each subinterval and partial subinterval includes means for determining watts according to the relationship

where W represents watts, K is a constant, Nw is the number of pulses representing for the subinterval watts, and in which the means for determining for each subinterval and partial subinterval includes means for determining power factor according to the relationship W Power factor= VA

13.Apparatus for providing an output representing VA demand for a predetermined time period T2-T0, comprising means for determining a value Wh representing watthours of energy provided to a system in each of a number of subintervals t2-t1 of said predetermined time period, means for determining a value VARh representing VAR hours for said energy provided to said system in each of said subintervals, means for determining a value'representing VA demand for each said subinterval according to the relationship

and means for summing the values of VA demand for the subintervals to provide said output representing VA demand for said predetermined time period.

14. Apparatus for determining vector VA demand for a predetermined time period T2-T0 provided by a power system to a load, comprising a watthour meter providing a first pulse train representing watthours, Wh, a Q-hour meter connected to the system to have a rotation angle of 600 lagging, for providing a second pulse train representing Hours, Qh, means for dividing said time period into a number of equal sampling periods, means for providing consecutive subintervals t2-t1, each subinterval terminating in a sampling period, means for receiving said first and second pulse trains and deriving for each subinterval a value for VAR hours, VARh, in accordance with the relationship

means for receiving said first pulse train and said value for VARh and deriving therefrom for each subinterval a value representing VA demand in accordance with the relationship

and means for receiving the value representing VA demand for each subinterval and summing the values over said predetermined time period.

1 5. Apparatus for providing an output representing VA demand for a predetermined time period T2-T0, comprising a watthour meter for providing a first pulse train whose pulse frequency represents watthours of energy provided to an electrical system, a O-hour meter for providing a second pulse train whose pulse frequency represents Q-hours of energy provided to said system, said O-hour meter being connected with a rotational angle 4 of 600 lagging, divider means for dividing said time period into a number of consecutive and substantially equal sampling periods, means for identifying for any sampling period having pulses from both meters the last pulse for each said first and second pulue train.

means for defining a subinterval t2-t1 terminating with the earlier occurring of the said two last pulses in a sampling period and beginning with the termination of the last subinterval, means for determining for each subinterval the number of pulses NC in the subinterval for the pulse train which had the last pulse, and the number of pulses ND in the subinterval for the pulse train which had the earlier of the said two last pulses, calculating means for calculating for each subinterval a value for VA demand according to the relationship

where K=a constant, means for storing and summing the values for VA demand for each subinterval in said time period, and output means for receiving the summed values for VA demand and providing as an output signal the total of said summed values representing VA demand for said time period.

16. Apparatus as defined in claim 1 5 and further comprising means for defining a partial subinterval beginning with the termination of the last subinterval and terminating at the end of said time period, means for determining for said partial subinterval the number of pulses Nc for the pulse train which had the last pulse, and the number of pulses ND for the other pulse train, calculating means for calculating VA demand for said partial subinterval according to the relationship

where K=a constant, and means for including the value for VA demand for the partial subinterval with the summed values for VA demand for the subintervals in the same time period.

1 7. Apparatus as defined in claim 1 6 in which the last pulse in the pulse train having the last pulse in a sampling period is represented by TA, in which the last pulse in the other pulse train in a sampling period is represented by TB, and further comprising ~ means for identifying and storing the respective times representing Tc and TF of the second and third last pulses in each sampling period for the pulse train having the last pulse, the respective times represented by TH and TG of the second and third last pulses of the other pulse train and the respective times represented by T, and TM of the pulse in the pulse train having the last pulse which follows the last pulse TB in the other pulse train and the pulse preceding it, in which Nc is further defined as

where Nthe whole number of pulses in the respective subinterval for the pulse train having the said last pulse,

QP=a fraction of a pulse carried over from the preceding calculation, and

where N1=the whole number of pulses in the respective subinterval for the other said pulse train,

AP2=a fraction of a pulse carried over from the preceding calculation, and

where N,,=the whole number of pulses in the partial subinterval for the pulse train having the last said pulse,

AP,=a fraction of a pulse carried over from the preceding calculation, and

where NB=the whole number of pulse in the partial subinterval for the other said pulse train

and AP2, is a fraction carried over from the preceding calculation, and where TÁ, TC, and TF, are times representing the last, second last and third last pulses in the pulse train having the last pulse in said time period and TB TH and Td are times representing the last, second last and third last pulses in the other pulse train.

1 8. A method for determining VA demand for a predetermined time period T2-T0 comprising the steps of receiving a first pulse train representing watthours Wh, and a second pulse train representing 0hours Oh, dividing said time period into a number of consecutive sampling periods, identifying for sampling periods having pulses the last three pulses for each pulse train and determining which pulse train had a last pulse in the sampling period occurring later than the other, defining for said sampling periods a subinterval t2-t1 terminating with the last pulse of the pulse train which did not have the later of the two said last pulses and beginning with the end of the last proceeding subinterval, determining the number of pulses from both the first and second pulse train in each subinterval, calculating VA demand for each subinterval based on the relationship

where K =a constant Nc=number of pulses determined in the subinterval for the pulse train which had the last pulse Number of pulses determined in the subinterval for the other pulse train and summing the values VA demand calculated for each subinterval in said time period to determine VA demand for said time period.

19. A method as defined n claim 18 in which the step of determining the number of pulses NC and ND comprises obtaining a count NA of the whole number of pulses occurring in the subinterval for the pulse train which had the last pulse, determining a value AP3 which is the time interval between the end of the subinterval and the pulse from the same meter preceding the end of the subinterval, divided by the time interval between the pulse which follows the end of the subinterval and the pulse preceding the end of the subinterval for the pulse train which had the last pulse in the sampling period determining NC according to the relationship

where AP ,=a fraction of a pulse carried over from the preceding subinterval for the pulse train having the last pulse, obtaining a count NB of the whole number of pulses occurring in the subinterval for the other said pulse train determining a value AP4 which is the time interval between the end of the subinterval and the second last pulse for the other said pulse train, divided by the time interval between the last pulse and the second last pulse of the same pulse train, and determining ND according to the relationship

where AP2=a fraction of a pulse carried over from the preceding subinterval for the other said pulse train.

20. A method as defined in claim 1 9 and further comprising the steps of determining the final sampling period in said time period and defining a partial subinterval beginning with the termination of the last subinterval and terminating at the end of said time period, calculating a value Nc representing the number of pulses in said partial subinterval for said pulse train which had the last pulse, according to the relationship

where NÁ =the whole number of pulses occurring in said partial subinterval for the pulse train which had the last pulse, AP3=the time interval between the end of the final sampling period and the last pulse, multiplied by the time interval between the second last and third last pulses for the pulse train which had the last pulse, divided by the square of the time interval between the last pulse and the second last pulse for the pulse train which had the last pulse in said time period, and AP;=is a fraction of a pulse carried over from the last preceding subinterval for the pulse train having the last pulse, calculating a value ND representing the number of pulses occurring in said partial subinterval for said pulse train which did not have the last pulse, according to the relationship

where NB =the whole number of pulses in said partial subinterval for the pulse train which did not have the last pulse, åP4=the time interval between the end of the final sampling period and the end of the last subinterval, multiplied by the time interval between the second last and the third last pulses for the pulse train which did not have the last pulse, divided by the square of the time interval between the last and second last pulse of the pulstrain which did not have the last pulse, AP2,=is a fraction of a pulse carried over from the preceding subinterval for the pulse train which did not have the last pulse, calculating VA demand for said partial subinterval based on the relationship

and adding VA demand for the partial subinterval to the summed values for VA demand for the subintervals in said time period to determine said VA demand for said time period.

21. The invention as defined in any of the preceding claims including any further features of novelty disclosed.

22. Apparatus for determining an electrical value related to consumption of electrical energy over a predetermined time period, substantially as herein described with reference to the accompanying drawings.

23. A method of determining an electrical value related to consumption of electrical energy over a predetermined time period, substantially as herein described with reference to the accompanying drawings.