You are on page 1of 7



Abstract: Many definitions have been formulated to characterize, detect, and measure active and nonactive current and power for non-sinusoidal and non-periodic waveforms in electric systems. This paper presents definitions and compensation of non-active current from the compensation standpoint.

Introduction Because of the widespread use of nonlinear loads and electronic power converters, non-sinusoidal and non-periodic wave-forms and voltage distortion are becoming more common in todays electrical systems. Many papers have dealt with the definition, identification, characterization, detection, measurement, and compensation of such nonsinusoidal and non-periodic current and power. Tolbert and Habetler have compiled a comprehensive technical survey of the published literature on the topic [8]. Everybody uses today the concept of reactive power, from the physicist to the electrical engineer. We are so accustomed with this term that we do not realize the contradictions associated with this concept. If one follows any textbook, the concepts of active and reactive power are introduced, usually at the beginning, as a product of the voltage and current measured at the same point of the circuit. The active power depends on the cos function and the reactive power on the sin function where the angle is taken between the voltage and the current. When the average in time is calculated, the two powers behave quite differently; the average active power is well defined while the average reactive power is zero, no matter of the network or state of the system. If for the active power (correctly the energy) exists a net flow from one point of the network to another, for the reactive power there is a continuously flow back and forth (to and from), but the net flow is zero for a complete cycle, as the amount of energy flowing in one direction for half a cycle is equal to the amount of energy flowing in the opposite direction in the next half of the cycle. The reactive power is exchanged by different parts of the network capacitors and reactors permanently, but is never consumed or produced. Instantaneous active power is defined as the time rate of energy generation, transfer, or utilization. It is a physical quantity and satisfies the principle of conservation of energy. For a single-phase circuit, it is defined as the instantaneous product of voltage and current: (1) p(t ) = u (t ) i (t ) As is not practical to work with instantaneous quantities because they are difficult to measure, averaged values are introduced. The average value of the instantaneous active power P is called the active power and is given by the time average of the instantaneous power over one period of the wave p(t). For a polyphased circuit with n phases, each phases instantaneous active power is still expressed as (1) and instantaneous total active power is the sum of the active powers of the individual phases: p (t ) = pi (t ) = u i (t ) ii (t ) .
i =1 i =1 n n



Non-active power can be thought of as the useless power that causes increased line current and losses, greater generation requirements for utilities, and other effects/burdens to power systems and connected/related equipment. For a single phase circuit with inductors, capacitors, and/or nonlinear elements, non-active power is the power that circulates back and forth between the source and loads and yields zero average active power over one period of the wave p(t). Therefore, the non-active power for single phase circuits is based on average values. For a polyphased circuit, non-active power is the power that circulates back and forth between the sources and loads and the power that circulates among the phases. The sourceload circulating power yields zero average over one period of the wave p(t) because of unbalanced storage elements in the circuit. The phase circulating power contributes no total instantaneous active power to the circuits because of balanced storage elements. Therefore, the non-active power for polyphased circuits includes two components: one is based on average values and the other is based on instantaneous value. Some theories are based on average values and restricted to frequency domain, while some others are formulated on time domain and instantaneous base. No matter what mathematical means are used, the goal of these theories is to improve the power factor and to minimize power losses and disturbances by identifying, measuring and eliminating the useless power. This paper presents definitions of non-active power or non-active current from the compensation standpoint. Almost all existing non-active power theories and definitions can be extended and deduced from the definitions presented. 2. Compensation system and definition of non-active power For a single or polyphased power system, a shunt compensator to minimize the useless power/current can be configured as in Fig. 1. Assuming that the shunt compensator only consists of passive (inductor and/or capacitor) and/or switching devices without any external power source and neglecting the compensators power loss, then the active power of the compensator should average zero according to the principle of conservation of energy. That is: (3) Ps (t ) t = PL (t ) t , PC (t ) t 1 Px (t ) = p x ( )d , for x=S, L or C. TC t C T


In (4), T C is the averaging interval that can be zero, one fundamental cycle, one-half cycle, or multiple cycles, depending on compensation objectives and the passive components energy storage capacity.

iS, p S Utility line from the source

iL, p L Load

iC, p C Compensator

Figure 1.: A shunt compensator configuration Equations (3) and (4) must hold true regardless of single-phase or polyphased, passive compensation or active compensation. Based on these physical and practical limitations, nonactive power and non-active current can be defined and formulated. Fryzes idea of non-active current/power [5] may be extended as follows: P(t ) (5) i p (t ) = 2 u p (t ), iq (t ) = i (t ) i p (t ) U P (t ) 1 U p (t ) = u 2 ( )d . p TC t TC


In the above formulae, i p (t) is the active current and i q (t) is the non-active current. P(t) is the average active power over the interval [t-T C , t], which can be calculated from (4). U P(t) is the rms value of the voltage u P (t) over the interval [t-T C , t], which is expressed by (6). u P (t) is the reference voltage that can be the voltage itself, u P(t)= u(t), or the fundamental component of u(t), where u(t)=u f (t)+u h (t) and u P (t)= u f (t), or something else, depending on compensation objectives. These definitions, (5) and (6) are valid for single- and poly-phase circuits. However, for polyphased circuits, voltages and currents are expressed as a vector, e.g., for a three-phase system, (7) u = [u a , u b , u c ]T, i = [i a , i b , i c ]T, and u2 = [u a , u b , u c ] [u a , u b , u c ]T.


Compensation Equation (5) provides the basic definitions of active and non-active current, from which most of the existing non-active power theories and definitions based on time-domain can be extended and deduced. The following discusses some deductions and compensation examples. A. Sinusoidal Single-Phase Circuits For a single phase circuit with sinusoidal waveforms, e.g.:

(8) u s = U s sin (t ) , i L = I L sin (t ) , the active and non-active currents are consistent with the traditional active and reactive currents and can be derived from (4), (5) and (6) by the following steps: 2 (i) choose T C to be one or half fundamental cycle, TC = or TC = ;

(ii) calculate the average active power, P L according to (4); (iii) calculate the rms value of the voltage u S , according to (6); (iv) calculate the active current and nonactive current, i Lp and i Lq according to (5). The result is: (9) i Lp (t ) = I L cos( ) sin (t ) and i Lq (t ) = I L sin ( ) cos(t ) . Equation (9) clearly shows the consistence with the traditional reactive power theory. An algorithm for active compensators can be easily implemented because the definitions as formulated in (4), (5) and (6) are all real-time based. In addition, it is easy to show that the compensator needs zero average power when it injects the non-active current because P C =0 when i C = i Lq . After compensation, the source current, i S will only contain the load active current, i Lp . B. Non-Sinusoidal Single-Phase Circuits For a single-phase circuit with non-sinusoidal waveforms, the active and non-active currents, which can be derived from (4), (5) and (6) using the steps described in section 3.A, are consistent with the traditional Fryzes active and non-reactive currents when choosing u P=u S in (5) and (6). For example, if : u S = u f (t ) + u h (t ) = U Sf sin (t ) + U Sh sin ( h t + h ) , (10) i L = I Lf sin (t ) + I Lh sin ( h t + h h ) then: U Sf I Lf cos + U Sh I Lh cos h i Lp (t ) = U Sf sin(t ) + U Sh sin( h t + h ) , (11) 2 2 U Sf + U Sh

i Lq (t ) = I Lf sin cos(t ) I Lh sin h cos( h t + h ) +

2 U Sh I Lf cos U Sf U Sh I Lh cos h +

sin(t ) 2 2 U Sf + U Sh

sin( h t + h ) 2 2 U Sf + U Sh Again, the average power of i Lq is zero, which satisfies the requirements in (3) for a compensator. However, it is observed from (11) that the active current contains harmonics because of the voltage distortion, which means that the source current will not become sinusoidal after the compensator injects (compensates) the non-active current expressed in (12). In most cases, it is desirable that compensation of non-active current results in a pure sine wave source current. In order to achieve that, one should choose u P = u Sf in (5) and (6). By doing so, one has: U I cos h sin(t ), (13) i Lp (t ) = I Lf cos + Sh Lh U Sf

2 U Sf I Lh cos h U Sf U Sh I Lf cos


U Sh I Lh cos h sin t + I Lh sin ( h t + h h ) . (14) U Sf Equation (13) shows that the active current is a sine wave, and (14) shows that the nonactive current contains all harmonic current and fundamental reactive current. After compensation, the source current will become sinusoidal and active. In addition, it is noticeable that the active and non-active currents expressed in (13) and (14) still meet the compensation energy conservation requirements in (3). This is a very important property of the definitions given in (4), (5), and (6), which is also necessary in order to implement compensation in Fig. 1. C. Single-Phase Circuits with Non-Sinusoidal and Non-Periodic Current In this case the active load current, I Lp , is sinusoidal and in phase with the voltage although the load current, I L , is highly distorted and nonperiodic, the calculated load nonactive current, I Lq , is highly distorted and out of phase with the source voltage, the average load power, P Lq , generated from the load nonactive current is zero. Therefore, when a compensator is used to compensate for the load non-active current it will consume average zero power and maintain the requirement in (3). In addition, the source current will become sinusoidal and in phase with the voltage. D. Three-Phase Circuits The definitions described in (5) and (6) are valid for a three-phase system as well regardless of whether the voltage and current waveforms are sinusoidal or non-sinusoidal, periodic or non-periodic, and balanced or unbalanced. Results are similar to those in the previous subsections. For polyphased (n-phase) systems, there is one interesting concept, instantaneous reactive (or non-active) power and current, which do not exist in single phase situations. This instantaneous non-active power theory [3, 7, 9] can be deduced from (5) and (6) as well. In (4) and (6), choosing T C .0 yields the instantaneous nonactive current (power) theory. That is: p (t ) (15) i p (t ) = 2 u p (t ) , iq (t ) = i (t ) i p (t ) , U p (t ) i Lq (t ) = I Lf sin cos t
2 2 2 u p (t ) = u1 p , u 2 p ,..., u np , i (t ) = [i1 , i2 ,..., in ] , U p (t ) = u12p + u 2 p + ... + u np . (16) Again, the definitions given in (4), (5) and (6) have great flexibility to meet all compensation objectives. Regardless of whether the voltage and current are sinusoidal or nonsinusoidal, balanced or unbalanced, and periodic or nonperiodic, the definitions give means to calculate any non-active current component that requires compensation. T T

iSa iSb iSc iSn

uSa uSb uSc

iLa iLb iLc iLn Nonlinear unbalanced load





Compensator with limited energy storage

Figure 2.: A compensation system for three-phase four-wire system Consider a three phase four wire system (Fig. 2). If the compensation objective is to make the source current sinusoidal and balanced, one can calculate compensation current as follows: - Separate the voltage into four components: fundamental positive-sequence, u Sf+ ; fundamental negative-sequence, u Sf- ; zero sequence, u S0 ; and harmonic component, u Sh , i.e.: u S = u SF + + u S 0 + u SF . (17) 1 T u S 0 = (u Sa + u Sb + u Sc ) [1,1,1] 3 - Choose T C =one-half or one fundamental cycle and u p = u Sf + in (4), (5) and (6), i.e.: T or TC = T , and u p = u Sf + . (18) 2 - Calculate the load non-active current, i Lq , as (5) and let i C =i Lq . From (4), (5), (6), (17), and (18), it is easy to show that the compensator consumes zero average power, P C =0, and satisfies the requirements in (3). The results clearly shows that the source current becomes sinusoidal, the source current becomes balanced as indicated by zero source neutral current, and the average power of the compensator, P c , equals zero. TC =

Conclusions In this paper, definitions of active and non-active power and current have been given from the compensation standpoint. Their definitions are consistent with the traditional reactive and non-active concept for single phase circuits. In addition, the instantaneous reactive power theories can be deduced from the proposed definitions for polyphased systems. The definitions also have the flexibility that any compensation objective can be achieved by choosing an appropriate averaging interval and reference voltage.


REFERENCES [1] Alexa, D., Ionescu, F., Gtlan, A., Lazr, A.: Convertoare de putere cu circuite rezonante, Editura tehnic Bucureti, 1998 [2] Alexa, D., Hrubaru, O.: Aplicaii ale convertoarelor statice de putere, Editura tehnic Bucureti, 1989 [3] Akagi H., Kanazawa Y., Nabae A., "Instantaneous Reactive Power Compensators Comprising Switching Devices without Energy Storage Components", IEEE Trans. Ind. Appl., vol.20, pp.625-630, May/June 1984. [4] Coma Dan: Utilizri ale energiei electrice, Editura didactic i pedagogic Bucureti, 1973. [5] Fryze S., Active, Reactive, and Apparent Power in Non-Sinusoidal Systems, Przeglad Elektrot, no. 7, pp. 193-203, 1931. [6] Ferrero A. and Superti-Furga G., "A New Approach to the Definition of Power Components in Three-Phase Systems Under Nonsinusoidal Conditions", IEEE Trans. Instrum. Meas., vol.40, no.3, June 1991. [7] Pcsi, R., Pierderi de putere n reelele electrice trifazate, referatul nr. 1, 2002. [8] Pcsi, R., Conversia static a energiei electrice, referatul nr. 2, 2002. [9] Peng F. Z. and Lai J. S., Generalized Instantaneous Reactive Power Theory for ThreePhase Power Systems., IEEE Trans. Instrum. Meas., vol.45, no.1, pp.293-297, February 1996. [10] ora, I., Utilizri ale energiei electrice, Editura Facla, Timioara, 1983. [11] Tolbert L. M. and Habetler T. G., Survey of Active and Non-Active Power Definition, IEEE International Power Electronics Congress, Acapulco Mexico, 2000. [12] Willems Jacques L., "A New Interpretation of the Akagi-Nabae Power Components of Nonsinusoidal Three-Phase Situations", IEEE Trans. Instrum. Meas., vol.41, no.4, August 1992.