Method and distance unit for calculating distance to fault in a line of an electric power system

Abstract

The invention relates to a computer-implemented method for determining a distance at which a fault occurs in a protected power line of an electric power system, wherein the power line is protected from a weak end of the power line, comprising: measuring the following magnitudes at a measurement point (85) of the protected line from a weak end where the fault occurs: phase A current at the measurement point (Ia), phase B current at the measurement point (Ib), phase C current at the measurement point (Ic), phase A voltage at the measurement point (Va), phase B voltage at the measurement point (Vb) and phase C voltage at the measurement point (Vc); on the basis of the previous measurements, calculating the following magnitudes of the power line: homopolar sequence voltage at the measurement point (V0), direct sequence voltage at the measurement point (V1), reverse sequence voltage at the measurement point (V2), homopolar sequence current at the measurement point (I0), direct sequence current at the measurement point (I1), reverse sequence current at the measurement point (I2), equivalent voltage (Veq), equivalent current (Ieq) and polarisation voltage (Vpol); calculating the distance p at which a single-phase, two-phase or three-phase fault occurs on the protected power line. Unit of measurement used by the method.

Description

[0001]METHOD AND DISTANCE UNIT FOR CALCULATING THE DISTANCE TO FAULT ON A LINE OF AN ELECTRICAL POWER SYSTEM FIELD OF THE INVENTION The present invention belongs to the power electronics sector and, more specifically, to the units for calculating the distance to a fault that are used in protection and / or location equipment in electrical power systems. PRIOR STATE OF THE ART In electrical power systems, such as alternating current (AC) systems, it is necessary to identify - in order to subsequently correct or eliminate - any faults that may occur (such as short circuits or ground faults), since they represent anomalous behavior that can damage the equipment and endanger the health of users or operators. Locating the fault is a necessary step for its subsequent elimination. Electrical power systems are usually equipped with protection equipment.Protection equipment is responsible for responding to short circuits that occur on a specific power line. To do this, it is essential to locate the point on the network where the short circuit occurred. That is, when a short circuit occurs, it is necessary to determine whether it occurred in the direction of the line it is covering (forward fault) or in the opposite direction (backward fault), and the distance at which it occurred from the measurement point. Short circuits (generally, faults) can be single-phase (short circuit between one phase and ground), two-phase (short circuit between two phases), two-phase to ground (short circuit between two phases and ground), or three-phase (short circuit between the three phases). To simplify the analysis of asymmetrical faults (single-phase, two-phase, and two-phase to ground) and symmetrical faults (three-phase), so-called sequence networks are used.Using sequence networks, an asymmetrical network is generally broken down into three symmetrical networks (zero-sequence network, direct-sequence network, and inverse-sequence network). In the case of single-phase faults, to make the asymmetrical system equivalent to the three symmetrical sequence networks, the three sequence networks are connected in series, as shown in Figures 1A and 1B. Figure 1A shows a three-phase system (phases a, b, c) in which a fault has occurred in phase "a" (single-phase fault). The currents I. a Ib Ic represent the currents of phases a, b, and c at the measurement point, respectively. The fault in phase “a” is represented by the fault impedance Z f that connects phase “a” to ground. Figure 1B shows the three sequence networks connected in series: The (+) network represents the direct sequence network of the system up to the fault point. The voltage V a1is the direct sequence voltage of the system up to the fault point. The current I a1 is the direct sequence current of the fault current. The (-) network represents the negative sequence network of the system up to the fault point. The voltage V a2 is the negative sequence voltage of the system up to the fault point. The current I a2 is the inverse sequence current of the fault current. Network(0) represents the zero sequence network of the system up to the fault point. The voltage V a0 is the zero sequence voltage of the system up to the fault point. The current I a0 is the zero-sequence current of the fault current. 3Z frepresents the resistance at the fault point that connects the faulted phase to ground. This fault current is 3 times the Ia0 current, so the “3” is represented by multiplying it with the fault resistance Zf so that the sequence networks connected in series represent the asymmetrical network. In the case of two-phase faults, the zero-sequence network is zero. The direct and reverse sequence networks are connected in parallel, as shown in Figures 2A and 2B. Figure 2A shows a three-phase system (phases a, b, c) in which a fault has occurred in two phases: “b” and “c” (two-phase fault). The currents Ia Ib Ic represent the currents of phases a, b, c at the measurement point, respectively. The two-phase fault is represented by the fault impedance Zf, which connects phase “b” to phase “c”. Figure 2B shows the two sequence networks connected in parallel: The (+) network represents the direct sequence network of the system up to the fault point.Voltage Va1 is the direct sequence voltage of the system up to the fault point. Current Ia1 is the direct sequence current of the fault current. The (-) network represents the negative sequence network of the system up to the fault point. Voltage Va2 is the negative sequence voltage of the system up to the fault point. Current Ia2 is the negative sequence current of the fault current. The fault resistance Zf is the resistance that joins the two faulted phases. In the case of two-phase-to-earth faults, the three sequence networks are connected in parallel. Figure 3A shows a three-phase system (phases a, b, c) in which a two-phase-to-earth fault has occurred on phases “b” and “c”. The impedance Zff represents the fault resistance between phases “b” and “c”. The impedance Z. ftrepresents the resistance between phase “b” and ground and the resistance between phase “c” and ground. Figure 3B shows the three sequence networks connected in parallel: The (+) network represents the direct sequence network of the system up to the fault point. The voltage V a1 is the direct sequence voltage of the system up to the fault point. The current I a1 is the direct sequence current of the fault current. The (-) network represents the negative sequence network of the system up to the fault point. The voltage V a2 is the negative sequence voltage of the system up to the fault point. The current I a2 is the inverse sequence current of the fault current. Network(0) represents the zero sequence network of the system up to the fault point. The voltage V a0 is the zero sequence voltage of the system up to the fault point. The current I a0is the zero-sequence current of the fault current. In the case of balanced three-phase faults, only the direct-sequence network exists. Figure 4A shows a three-phase system (phases a, b, c) in which a balanced three-phase fault has occurred. The currents I a Yo b Yo crepresent the currents of phases a, b, and c at the measuring point, respectively. A balanced three-phase fault is represented by the fault impedance Zf. Figure 4B shows the direct-sequence network: The (+) network represents the direct-sequence network of the system up to the fault point. The voltage Va1 is the direct-sequence voltage of the system up to the fault point. The current Ia1 is the direct-sequence current of the fault current. Using sequence networks, the distance from the measuring point to the short-circuit can be calculated because the distance to the fault is proportional to the impedance from the measuring point to the fault point. In protection equipment, the unit responsible for determining the distance at which the fault occurred is commonly called the distance unit. The distance unit is located at one end of the line being protected.The distance unit plays a key role in the quality of the protection equipment in which it is integrated. Note that distance is proportional to the measured impedance, so the impedance calculation represents the distance to the fault. The distance to the fault is usually measured in values ​​per unit (p). The measurement point is the point from which voltages and currents are measured to calculate the distance. There are different types of distance units. Among the most popular is the quadrangular distance unit, so named because the characteristic is a four-sided polygon. In the quadrangular characteristic, it is necessary to work with the reactance measurement and the impedance resistance measurement. A distance unit determines the impedance at which a fault has occurred by dividing the voltage by the current on the protected line when a fault occurs.Impedance is a complex number where the real part represents the fault resistance and the imaginary part the reactance to the fault. The quadrangular distance function consists of a characteristic in which the inductive reach and the resistive reach must be adjusted. The distance function is traditionally measured in Ohms (Ω, a measure of impedance, to which the distance to the fault is proportional). The quadrangular distance function works on a plane where the X axis represents the fault resistance and the Y axis represents the fault reactance. The distance therefore has two dimensions, as illustrated by way of example in Figure 5: a resistive dimension (normally represented on the X or abscissa axis) and an inductive dimension (normally represented on the Y or ordinate axis). The impedance value (X, Y) is used to analyze whether the fault is inside or outside the characteristic.The characteristic used is a 4-sided polygon that delimits the characteristic in both X and Y. That is, the characteristic is delimited by the inductive reach and the resistive reach, as illustrated in Figure 5. In Figure 5, the characteristic has been defined with a dotted outline. If the operating point obtained from the value of the fault resistance (X axis) and the fault reactance (Y axis) is within the characteristic (i.e., the four-sided polygon), the fault is considered to be within the area of ​​operation of the quadrangular distance function (i.e., the fault is within the set reach). If the operating point is outside the characteristic, the fault is considered to be outside the area of ​​operation of the function (i.e., the fault is outside the set reach). This is the case for the fault whose point Pfault = (Rfault, Xfault) is illustrated in Figure 5.The illustrated ZR vector has two components: the Y axis represents the adjusted reactance, and the X axis represents the line resistance from the measurement point to the reactive reach. RF represents the resistive reach setting. Xfault is the reactance calculated by the function. Rfault is the resistance calculated by the function. For each protected line, the settings (inductive and resistive) are different. Depending on whether the operating point is inside or outside the characteristic, the distance unit will give or not an opening order to circuit breakers that isolate the fault and protect the line. Note that a fault generates high currents that can burn the line or break it due to the dynamic forces generated. Therefore, when a fault is detected, it must be isolated as soon as possible. On the other hand, a short circuit on the line can cause destabilization of the network (voltage dips, frequency increases or decreases, etc.).Therefore, the function (and therefore the equipment it integrates) also protects the rest of the network when it isolates the faulted line. If the operating point is outside this characteristic, the distance unit does not signal the breakers to open. However, if the operating point is within this characteristic, the distance unit signals the breakers to open. Typically, the inductive reach is set to 70–90% of the line's reactance. This percentage represents the amount of protection provided to the protected line. The resistive reach is normally set to no more than four times the inductive reach. The user adjusts these values ​​according to the impedance of the line they wish to protect. Since the line's impedance to the fault is proportional to the fault distance, it is the same as saying that we protect, for example, 70% of the line in kilometers or in ohms.That is, the reactive reach represents, in ohms, a percentage of the line to be protected. To correctly determine the reactance from the measurement point to the fault point in the case of resistive faults, the polarization current is usually the negative sequence current or the zero sequence current. The reactance X up to the fault point is calculated as follows: ∗ Im[^. ^^^ ∙ ^^ ^^^ ^ ] where ^ ^^ is the direct sequence impedance of the protected line, ^ ^^^ is the bias voltage, ^ ^^^ is polarization intensity e ^ ^^ It is the equivalent current, which in the case of single-phase faults corresponds to the compensated current. To calculate the resistance from the measurement point to the fault point, the real part of the division between the polarization voltage and the equivalent current is usually taken as the value: The decision rule for determining whether the fault is within or outside the scope is: -If X < X ^ and R < R ^ The fault occurred within the range of the distance unit. In other cases, the fault occurred outside the range of the unit. These distance units are highly accurate in radial systems where the line is protected from the generation side, even when the faults are resistive faults. In distributed generation, since the fault current is supplied from both ends of the line, protection begins to lose accuracy in heterogeneous systems and resistive faults. One solution to avoid overreaches is to use an adjustable tilt angle for calculating reactance: where ^ ^^^^^is the tilt angle, in degrees, and is adjusted with the value of the difference between the angles of negative sequence or zero-sequence current between the two ends of the line at the line reach value. To do this, it is necessary to know the source impedance of the remote end. This method has several drawbacks. The first is that the remote source impedance changes due to the topology changes that the system undergoes to optimize load flow, so the source impedance adjustment does not always represent reality. The second drawback is that, even if the value of the remote source impedance is known exactly, the distance calculation will be accurate only when the fault occurs right within the adjusted reach (i.e., at the limit of the adjustment). With faults outside or within the reach, the distance calculation is not accurate.The weaker the end from which the line is protected, the greater the inaccuracy in distance calculations in heterogeneous networks and in faults with non-zero resistance. This occurs because, when there are two ends feeding the fault, each end only measures its contribution. The weaker the end from which the line is protected, the lower its current contribution to the fault. The error is greater the greater the heterogeneity of the network and in faults with non-zero resistance. As an example, Figure 6 shows a diagram of a power network in which, in the lower right corner, a dot identifies the 60th position where the distance to the fault is measured. This 60th position is the point where currents and voltages are measured, for example, by a relay. Zl represents the impedance of the line to be protected. To its left is the strong end of the network.To the right of point 60 there may be a load (it is a passive element of the network) or there may be a weak source (such as renewable sources, generators with low generation capacity, etc.), represented as 65. On the other hand, for two-phase faults there is another way to calculate the distance to the fault, which requires determining the value of the remote source impedance in order to accurately calculate the distance (Chenxu Chao et. al., Adaptive distance protection based on the analytical model of additional impedance for inverter-interfaced renewable power plants during asymmetrical faults, 5 October 2022). This method first calculates the observed impedance by dividing the composite voltage by the composite current: where ^ ^^ is the seen impedance, ^ ^ is the voltage of phase A at the measurement point, ^ ^ is the voltage of phase B at the measurement point, ^ ^ is the intensity of phase A at the measurement point e ^^ is the intensity of phase B at the measurement point. This impedance seen ^ ^^ It is represented as a function of the fault resistance (R ^ ), direct sequence intensity of phase c (^ ^^ ), distance to the fault in per unit values ​​(^), direct sequence impedance of the line (^ ^^ ) and measured composite intensity (^ ^ − ^ ^ ). In the above equation, the second summand represents the error caused by the fault resistance in calculating the distance when calculating the observed impedance. This error is called the additional impedance ^ ^^^ : To calculate the distance to the fault, on the one hand, the angle φ is calculated add of the additional impedance ^ ^^^ by trigonometry, putting it in relation to the distance to the fault: ^ ( ^ ) = | ^ ^^ | ∙ without ^ ^^ − ^ ∙ | ^ ^^| ∙ without ^ ^ ^(^) = |^ ^^ | ∙ cos φ ^^ − ^ ∙ |^ ^^ | ∙ cos ^ ^ where ^ ^ is the angle of the impedance of the line to be protected. On the other hand, this angle φadd is calculated using the additional impedance ^ ^^^ . For this purpose, the fault current is represented as a function of the reverse sequence voltage of phase c at the measuring point (^ ^^ ), distance to the fault (p), direct sequence impedance of the protected line (^ ^^ ), direct sequence impedance of the remote source (^ ^^ ) and measured composite intensity (^ ^ − ^ ^ ). The direct sequence impedance of the remote source is calculated with the pre-fault voltage and current variations: ∆^ ^ ^^ = − ^ ∆^ ^^ When the two equations are equated, the only unknown is the distance to the fault. The resulting equation is a second-order equation: ^ = |^^^ | ^ ∙ cos(2 ∙ ^ ^ − ^) ^ = − | ^ ^^ | ∙ | ^ ^^ | ∙ cos(^ − ^ ^^ −^ ^ ) − | ^ ^^ | ∙ | ^ ^ | ∙ cos(^ − ^ ^ −^ ^ ) − | ^ ^^ |^ ∙ cos(^ − 2 ∙ ^ ^ ) ^ = | ^ ^^ | ∙ | ^ ^^ | ∙ cos(^ ^ +^ ^^ − ^) + | ^ ^^ | ∙ | ^ ^^ | ∙ cos(^ ^ +^ ^^ − ^) This method provides an accurate result for cases where there is no reverse current injection. In the case of reverse current injection, the protection makes an error in estimating the distance to the fault. For cases where reverse current injection is present, a correction angle is used, calculated using the set maximum reach value.^^^^ = arg where ^ ^^ is the reverse sequence intensity of phase c at the measurement point and ^ ^^ is the reverse sequence voltage of phase c at the measuring point. This correction angle ^ ^^^^ is applied to the reverse voltage. Therefore, the angle of the additional impedance will be equal to: However, even if the angle is corrected, the calculation will only be accurate for faults occurring right within the reach. For faults closer or further from the reach, the distance calculation will not be accurate. In short, the most advanced distance units currently are tilt-angle distance units, which represent a slight improvement over traditional reverse-sequence and zero-sequence current-polarized units. Even so, these units are unable to provide an adequate response when the line is protected from the weak end, in which case the response is lower the higher the fault resistance. Therefore, there is a need to develop new techniques for calculating the distance to a fault and distance units that implement these techniques, which overcome the drawbacks of the currently known and used distance techniques and units.DESCRIPTION OF THE INVENTION The present invention provides a method and a distance unit that attempt to resolve the drawbacks of the techniques described in the state of the art. Specifically, to improve the calculation of the distance to a fault in the case of protecting the line from the weak end in an electrical power system, a distance unit is provided that calculates the distance to the fault taking into account more network parameters (than those taken into account in conventional distance units). The distance unit and the method implemented therein are valid for protecting the line from the weak end, whether or not there is reverse current injection. The distance to the fault is therefore calculated more accurately.The new distance unit, which is a quadrangular distance unit, represents a new paradigm for distance units that uses a new formulation to accurately determine the distance to the fault for cases where a line is protected from the weak end of an alternating current (AC) power system. This new formulation takes into account all network parameters and is therefore capable of outperforming all existing units. Specifically, the distance is calculated using the direct-sequence impedance, which appears in all types of faults (single-phase, two-phase, two-phase-to-ground, and three-phase). Note that zero-sequence impedance does not appear in two-phase faults. Neither zero-sequence nor inverse zero-sequence impedance appears in three-phase faults. Due to its design adapted to weak ends, it is suitable for use at the ends of outgoing lines in, for example, inverter-based generating plants.Inverter-based generators can have the option of injecting only forward-sequence current or both forward-sequence and reverse-sequence current when a fault occurs. The new algorithm accurately calculates both cases. The proposed distance unit comprises a distance calculation module (which, for example, calculates the distance to the fault for each fault type and phase). The algorithm proposed in the present invention is preferably embedded in said distance calculation module. Thus, the new distance unit accurately determines the distance to the fault for cases where the line is protected from the weak end. The weaker the end from which the line is protected, the greater the inaccuracy in distance calculation in heterogeneous networks and for faults with non-zero resistance.This occurs because, when there are two ends feeding the fault, each end only measures its contribution. In the context of the present invention, one end from which a line is protected is called the "weak end" when it is the end that contributes less fault current than the other end. Thus, the other end from which the line is protected is called the "strong end" since it is the end that contributes more fault current. The distance unit is located at one of the ends or points of the line being protected. Of these two ends, the one that contributes much less fault current than the other end is the weak end. The other is the strong end. That is, the weak end has a much lower short-circuit power than the strong end. The weak end may be so because it does not have as high a generating capacity as the strong end.The distance unit may comprise other modules, typically present in conventional distance units, such as a fault detector (which indicates when a fault has occurred and whether it is a single-phase, two-phase, two-phase-to-earth, or three-phase fault), a fault selector (indicating which phases are involved in the fault), and / or a directional indicator (indicating whether the fault is in front of or behind the measuring point). The proposed distance calculation is based on the inventors' observation that, by protecting the line from the weak end, the distance to the fault can be accurately calculated for single-phase, two-phase, and three-phase faults, whether or not there is fault resistance. Thus, the inventors have developed an algorithm valid for weak ends whether or not there is negative sequence current injection in the event of fault detection.The distance to the fault is preferably calculated within the protection system itself, i.e., locally, for example, in the distance unit itself. The distance calculation can be sent, either in real time or not, to a control center, usually remote. Alternatively, the distance calculation can be performed remotely, based on measurements taken by other equipment, such as sensors located at different points in the power network. The distance unit determines the impedance at which a fault has occurred by dividing the voltage by the current on the protected line when a fault occurs. As explained in the prior art, the distance function consists of a characteristic defined by the inductive reach and the resistive reach. For each line, these adjustments are of a different value. The adjustment of the reaches is outside the scope of the present invention.Once the distance to the fault has been calculated, it is compared with preset values ​​(preset characteristic) to determine whether the fault is inside or outside the protected line. The proposed distance unit is implemented in a protection system. Once it has been established whether the fault is inside or outside the protected line, a decision can be made to act (isolate the fault, etc.) if it is within the protected line. Depending on whether the operating point (the point where the fault is estimated to be located with a given fault resistance) is within or outside this characteristic (the estimated reach), the distance unit (or, in general, the protection system) will or will not give an opening order to some circuit breakers to isolate the fault. If the operating point is outside this characteristic, the opening order will not be given to the circuit breakers.However, if the operating point is within this characteristic, the circuit breakers will be ordered to open. Normally, an attempt is made to protect up to a certain percentage of the line. For example, at least 70% of the line is protected, such as at least 80% or at least 90% of the line. Note that, since distance is proportional to the line impedance, a certain percentage of the line corresponds to that same percentage of the line impedance. Generally, when a fault occurs (for example, a short circuit), the proposed calculation method uses the direct, reverse, and zero-sequence voltages and the direct, reverse, and zero-sequence currents flowing through the point where the fault unit is located. Note that when reference is made to the voltages and / or currents used, it should be understood that said voltages and / or currents have been measured or calculated.The distance to the fault is calculated from the zero sequence voltages, direct and reverse ^. ^ , ^ ^ , ^ ^ , homopolar, direct and inverse sequence currents and homopolar and direct sequence impedances ^ ^^ , ^ ^^ and the reactive and resistive range adjustment values ​​X ^ , R ^ adjusted by the user. Note that ^ ^^ (direct sequence impedance of the protected line) = ^ ^^ (reverse sequence impedance of the protected line). The reactive reach setting value X ^ It is preferably set to a value within 70-90% of the reactance of the protected line. The resistive reach setting value R RIt is preferably set to a maximum of 4 times the reactive range.In a first aspect of the invention, there is provided a computer-implemented method (or digitally implemented method) for determining a distance at which a fault occurs on a protected power line of an electrical power system, wherein the power line is protected from a weak end of the power line, comprising: measuring the following quantities at a measurement point of the protected line from a weak end in a fault situation: phase A current at the measurement point (Ia), phase B current at the measurement point (Ib), phase C current at the measurement point (Ic), phase A voltage at the measurement point (Va), phase B voltage at the measurement point (Vb), and phase C voltage at the measurement point (Vc); from the above measurements, calculating the following quantities of the power line: zero sequence voltage at the measurement point (^. ^), direct sequence voltage at the measuring point (^ ^ ), reverse sequence voltage at the measuring point (^ ^ ), zero sequence current at the measuring point (I0), direct sequence current at the measuring point (I1), inverse sequence current at the measuring point (I2), equivalent voltage (^ ^^ ), equivalent intensity (^ ^^ ) and bias voltage Calculate the distance at which a single-phase, two-phase or three-phase fault occurs on the protected power line using the following formula: where ^ ^^ is the direct sequence impedance of the protected line and is known; and ^ ^^is the direct sequence impedance of the remote source and has been calculated before the fault occurs from measurements of direct sequence voltage at measurement point V1 and direct sequence current at measurement point I1 at two different instants in a non-fault situation; where the value of the distance to the fault is the minimum of the following two possible solutions: where In embodiments of the invention, the method further comprises calculating, from the distance to the fault p, the reactance to the fault point from the measurement point: ^ = ^ ∙ X ^^ where XL1 is the direct sequence reactance of the protected line. In embodiments of the invention, the method further comprises calculating, from the distance to the fault p, the fault resistance: where R is the resistance seen from the measurement point to the fault point. In embodiments of the invention the method further comprises: comparing the reactance to the fault point from the measurement point X with a previously chosen reactive reach adjustment value X R ; compare the resistance seen from the measurement point to the fault point R with a previously chosen resistive reach adjustment value R R ; and -if X < X ^ and R < R ^ : determine that the fault has occurred within a unit distance; optionally, operate a switch to isolate the fault from the rest of the electrical power system. In embodiments of the invention, if the fault is single-phase: -if the fault occurs in phase A, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ are obtained: ^ ^^^ = 3 ∙ [ −^ ^ + ^ ^ ∙ ( ^^^ + ^ ^^ )] ^ ^^ = ^ ^ where: ^ ^ is the neutral intensity at the measurement point, which is ^ ^ = 3 ∙ ^ ^ and ^ ^ is the hom sequence compensation factor ^ polar, ^ ^ = ^∙^ ^^ -if the fault occurs in phase B, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ are obtained ^ ^^^ = 3 ∙ ^ ∙ [ −^ ^ + ^ ^ ∙ ( ^ ^^ + ^ ^^ )] ^ ^^ = ^ ^ where: ^ is equal to 1∠120º, and is known in advance, ^ ^ is the neutral intensity at the measurement point, which is ^ ^ = 3 ∙ ^ ^ and ^ ^ is the zero-sequence compensation factor, ^ ^ ^ = ^ ^ ^^ ^^ ^ ∙^ ^^-if the fault occurs in phase C, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ are obtained: ^ ^^^ = 3 ∙ ^ ^ ∙ [ −^ ^ + ^ ^ ∙ ( ^ ^^ + ^ ^^ )] ^ ^^ = ^ ^ ^ ^^ = ^ ^ + ^ ^ ∙ ^ ^ where: ^ is equal to 1∠120º, and is known in advance, ^ ^ is the neutral intensity at the measurement point, ^ ^ = 3 ∙ ^ ^ and ^ ^ is the zero-sequence compensation factor, ^ ^ = ^ ^^ ^^ ^^ ^ ∙^ ^^ . In embodiments of the invention, if the fault is biphasic: -if the fault occurs in the BC phases, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ are obtained: ^ ^^^ = −2 ∙ [ −^^ + ^ ^ ∙ ( ^ ^^ + ^ ^^ )] -if the fault occurs in the AC phases, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ are obtained: ^ ^^^ = −^ ∙ 2 ∙ [−^ ^ + ^ ^ ∙ (^ ^^ + ^ ^^ )] ^ ^^ = ^ ^ √ ∙ ^ ^ − ^ ∙ ^ ^ = j 3 ∙ (^ − ^ ) 3 ^ ^ where: ^ is equal to 1∠120º, and is known in advance, -if the fault occurs in the BA phases, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ are obtained: ^ ^^^ = −^ ^ ∙ 2 ∙ [−^ ^ + ^ ^ ∙ (^ ^^ + ^ ^^ )] ^ ^ ^ ^ √3 ^ = ^ ∙ ^ − ^ ∙ ^ ^ = j ∙ (^ ^ − ^ ) 3 ^ ^ ^^ = ^ ^where: ^ is equal to 1∠120º, and is known in advance. In embodiments of the invention, if the fault is three-phase, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ are obtained: ^ ^^ = ^ ^ ^ ^^ = ^ ^ where ^ ^^^^^ is the direct sequence fault voltage at the measuring point with the prefault value removed, ^ ^^^^^is the direct sequence fault intensity at the measurement point with the pre-fault value removed. In a second aspect of the invention, there is provided a distance unit for determining the distance at which a fault occurs on a protected power line of an electrical power system, where the power line is protected from a weak end of the power line, comprising: means for measuring the following quantities at a measurement point of the protected line from a weak end in a fault situation: phase A current at the measurement point (I a ), phase B intensity at the measurement point (I b ), intensity of phase C at the measurement point (I c ), phase A voltage at the measurement point (V a ), phase B voltage at the measurement point (V b ) and phase C voltage at the measurement point (V c); means to calculate, from the above measurements, the following magnitudes of the power line: zero-sequence voltage at the measurement point (^ ^ ), direct sequence voltage at the measuring point (^ ^ ), reverse sequence voltage at the measuring point (^ ^ ), zero sequence current at the measuring point (I0), direct sequence current at the measuring point (I1), inverse sequence current at the measuring point (I2), equivalent voltage (^ ^^ ), equivalent intensity (^ ^^) and polarization voltage means for calculating the distance at which a single-phase, two-phase or three-phase fault occurs on the protected power line from the formula defined in the first aspect of the present invention, and where the value of distance to the fault is the minimum solution of the two possible solutions defined in the first aspect of the invention. In embodiments of the invention, the distance unit further comprises means for calculating, from the distance to the fault p, the reactance to the fault point from the measurement point: ^ = ^ ∙ X ^^where XL1 is the direct sequence reactance of the protected line. In embodiments of the invention, the distance unit further comprises means for calculating, from the distance to the fault p, the fault resistance according to the expression defined in the first aspect of the invention. In embodiments of the invention, the distance unit further comprises: means for comparing the reactance at the fault point from the measurement point X with a pre-set reactive reach setting value XR ; means for comparing the resistance seen from the measurement point to the fault point R with a pre-set resistive reach setting value R R ; and means for, if X < X ^ and R < R ^: determining that the fault has occurred within a distance unit; optionally, actuating a switch to isolate the fault from the rest of the electrical power system. In embodiments of the invention, the distance unit provides for specific calculations depending on whether the fault is single-phase, two-phase, or three-phase, similar to the method of the first aspect of the invention. In a third aspect of the invention, a protection system for an electrical power system is provided comprising a distance unit according to the second aspect of the invention. Because the proposed distance-to-fault calculation algorithm takes into account the aforementioned network parameters, it is able to overcome the drawbacks of conventional distance units.Due to its weak-end-adapted design, it is suitable for use, for example, at the outgoing line ends of inverter-based generating plants or at any other type of weak end of a power network, such as an end with high impedance or with little fault-current-generating capacity. In the case of inverter-based generators, these generators can have the option of injecting only direct-sequence current or both direct-sequence and negative-sequence current when a fault occurs. The proposed algorithm accurately calculates the distance to the fault in both cases. The proposed distance unit has the following improved features compared to existing units: It indicates the correct fault distance from the weak end of the line. It is valid for negative-sequence injection cases. It accurately calculates single-phase, two-phase, and three-phase faults.It is not affected by the fault resistance. It accurately calculates the remote source impedance. These and other advantages and features of the invention will become apparent in view of the figures and the detailed description of the invention. BRIEF DESCRIPTION OF THE DRAWINGS To complement the description and in order to aid in a better understanding of the features of the invention, in accordance with an example of practical embodiment thereof, a set of figures is attached as an integral part of the description in which, for illustrative and non-limiting purposes, the following has been represented: Figure 1A schematically shows a three-phase system (phases a, b, c) in which a fault has occurred in phase "a" (single-phase fault). Figure 1B shows a representation of the single-phase fault using sequence networks.Figure 2A shows a schematic representation of a three-phase system (phases a, b, c) in which a fault has occurred on two phases: “b” and “c” (two-phase fault). Figure 2B shows a representation of the two-phase fault using sequence networks. Figure 3A shows a schematic representation of a three-phase system (phases a, b, c) in which a two-phase to earth fault has occurred on phases “b” and “c”. Figure 3B shows a representation of the two-phase fault using sequence networks. Figure 4A shows a three-phase system (phases a, b, c) in which a balanced three-phase fault has occurred. Figure 4B shows a representation of the balanced three-phase fault using sequence networks. Figure 5 shows a representation on a coordinate diagram of the distance at which a fault has occurred. Distance has a resistive dimension (X or abscissa axis) and an inductive dimension (Y or ordinate axis).Figure 6 shows a schematic of a power network in which the distance unit is located at the weak end. The strong and weak ends of the network are shown with respect to the distance unit. Figure 7A shows a schematic of a network, in which the strong and weak ends are identified with respect to the protected line. In this case, the weak end is an end with a grounded star-connected transformer on the measurement side with no active generation sources on the delta side of the transformer. Figure 7B shows a schematic of a network, in which the strong and weak ends are identified with respect to the protected line. In this case, the weak end is an end with renewable sources. Figure 8 shows a representation or model of a power network.In this model, the network to the left of the strong end has been represented by the Thevenin equivalent with a voltage source and a remote source impedance. The network to the right of the weak end has been represented as a weak load or source. Figure 9 shows a schematic representation of the modeling of a fault-free network according to a possible embodiment of the invention. Figure 10 shows a schematic representation of the modeling of a fault-containing network according to a possible embodiment of the invention. Figure 11 shows a schematic representation of the modeling of a single-phase fault in a network according to a possible embodiment of the invention. The fault has been modeled with three sequence networks. Figure 12 shows a schematic representation of the modeling of a two-phase fault in a network according to a possible embodiment of the invention. The fault has been modeled with two sequence networks.Figure 13 shows a schematic representation of the modeling of a three-phase fault in a network according to a possible embodiment of the invention. The fault has been modeled with a sequence network. Figure 14 shows a schematic representation of the modeling of a sequence network in a pre-fault situation according to a possible embodiment of the invention. The network is represented by a strong end modeled as a voltage source with a source impedance and a weak end modeled by a current source and its source impedance (the weak end can also be modeled by a load, i.e. without a current source). The pre-fault situation is the correct situation (no fault) that exists before the fault occurs, but which does not provide information as to whether a fault will occur or not. Figure 15 shows a schematic representation of the modeling of a pure three-phase fault in a network, according to a possible embodiment of the invention.The fault has been modeled using a sequence network. The network is represented by a strong end modeled as a source impedance and a weak end modeled as a current source and its source impedance (the weak end can also be modeled as a load, i.e., without a current source). Note that the three-phase fault modeled in Figure 13 has the effect of the pre-fault, while in the pure three-phase fault modeled in Figure 15, the effect of the pre-fault has been eliminated. DESCRIPTION OF A METHOD OF CARRYING OUT THE INVENTION In the general diagram of a power network shown in Figure 6, the distance unit is located at one end of the line being protected, i.e., at point 60, where the voltages and currents are measured. The distance is calculated from these voltage and current measurements.In Figure 6, the strong end is to the left of the protected line impedance Zl, and the weak end is to the right of this impedance. The end that contributes much less fault current than the other end is the weak end. The other end is the strong end. As an example, a weak end can be an end where there are only loads, without generating power. For example, Figure 7A shows a diagram of a network (equivalent network), in which the strong and weak ends are identified with respect to the protected line (represented by its impedance Zl). Point 72, where voltages and currents are measured (e.g., measured by a relay), is the point where the distance unit is normally located. To the right of this point 72, i.e., at the weak end, there is no generating source. The equivalent network comprises a Thevenin equivalent voltage source 71 at the strong end.To the right of, for example, the relay, there is a star-to-ground transformer 73 on the measurement side and a delta transformer on the other side of the transformer. The load 74 at the weak end does not imply a generating source). Another example of a weak end is the case where there are renewable sources, which have little generating power, located behind the point 72 where the relay is placed, as illustrated in Figure 7B. In this case, the weak end is represented by a current source 76 representing a renewable source and the equivalent impedance 75 of the weak end (impedance between point 72 and the current source 76). The set of current source and equivalent impedance of the weak end has been represented as 77. Figure 8 shows a representation or model of a power network, such as the network in Figure 6.The network 81 to the left of the strong end can be represented by the Thevenin equivalent with a voltage source 82 and a remote source impedance Z. s 83, as illustrated in Figure 8. The dotted box 81 represents the strong end. The line being protected or to be protected is represented by a line impedance Z lThe distance unit is located at one end of the line being protected. In the diagram in Figure 8, the distance unit is connected to the relay located at point 85. This relay measures voltage and current. Of these two ends, the one that contributes much less fault current than the other end is the weak end. The other is the strong end. The weak end is represented by a weak load or source 86. The weaker the end from which the line is being protected, the lower its current contribution to the fault. The error is greater the greater the heterogeneity of the network and in faults with non-zero resistance. In perfectly homogeneous networks, the distance calculation would not be affected. The strong end contributes a large amount of fault current. The weak end has a much lower short-circuit power than the strong end.The weak end may be so because it does not have as high a generating power as the strong end, as shown in Figures 7A and 7B. The current passing through the fault point is the sum of the currents supplied by both ends. When a fault occurs, the remote source impedance Zs is normally small, and therefore the current flowing through it is high. The weak end contributes little current to the fault. Figure 9 shows a schematic representation of the modeling of a fault-free network in accordance with a possible embodiment of the invention. The network is represented by a strong end 91 modeled as a voltage source E 92 with a direct sequence impedance (which appears in all types of faults) of the remote source Zs193 and a weak end 97 (which may be a source or load 96) modeled by a current source and its source impedance. ^ is the Thevenin equivalent voltage at the remote end.In Figure 9, ^. ^ is the direct sequence voltage at the measuring point 85 e ^ ^ is the direct sequence current at the measuring point. In the diagram in Figure 9, the measuring unit is connected to relay 85. Figure 10 shows a schematic representation of the network modeling in Figure 9, with a fault. The network is represented by a strong end 91 modeled as a voltage source E 92 with a direct sequence impedance of the remote source Z s1 93 and a weak end 97 (which can be a source or load 96) modeled by a current source and its source impedance. The point where the fault occurred is located at a distance p from the measuring point 85. In the diagram in Figure 10, the measuring unit is connected to relay 85. The direct sequence impedance of the protected line Z L1 It is represented in two pieces (1-p) Z L1 yp Z L1to represent the fault at distance p. The proposed algorithm for calculating the distance to the fault requires as input the value of the direct sequence impedance of the remote source Z s1 . The remote source impedance Z sIt is variable, so it is necessary to recalculate this value. For example, in systems with renewable sources and in distribution networks, the injected current is usually variable due to variations in the load. The value of the remote source impedance can be recalculated when a variation in current (injected current) is detected. Alternatively, the periodic recalculation of this impedance value can be planned. To calculate the direct-sequence impedance of the remote source Zs1, two voltage and current values ​​are required. These two different voltage and current values ​​(direct-sequence) are measured at point 85 at different times in a non-fault situation.The first value used preferably corresponds to an instant before the detection of a variation in the load current I1 (direct sequence current at the measurement point 85) and the second value preferably corresponds to an instant after the detection of said variation. Analyzing figure 9 the calculation of the remote source direct sequence impedance Zs1 is obtained (see for example EP3361272A1 or Chenxu Chao et. al., Adaptive distance protection based on the analytical model of additional impedance for inverter- interfaced renewable power plants during asymmetrical faults, 5 October 2022, which are incorporated into this description by reference): ∆^ ^. ^^ = − ^ ∆^ ^^ where ^ ^^is the direct sequence impedance of the protected line, ∆^ is the variation in the direct sequence voltage measured at point 85 and ∆^ is the variation in the direct sequence current measured at point 85. When a short-circuit (generally a fault) occurs somewhere on the protected line from the weak end, the direct, reverse and zero sequence voltages and the direct, reverse and zero sequence currents flowing through the point where the distance unit is located (point 85) are used. The distance to the fault is then calculated from these voltages intensities and impedances ^ ^^ , ^ ^^ direct, inverse and homopolar sequence (note that ^ ^^ = ^ ^^^ ) and the reactive and resistive reach adjustment values ​​X ^ , R ^adjusted by the user, to determine the extent (distance) of the fault. Note that the remote source direct sequence impedance has been previously calculated ^ ^^ using direct voltage and current measurements ^ ^ , ^ ^at two different instants in a no-fault situation. When a variation occurs between these two magnitudes, two values ​​of V1 are obtained (one of these values ​​is prior to the variation and the other value is subsequent to the variation) and two values ​​of I1 (one of these values ​​is prior to the variation and the other value is subsequent to the variation). The algorithm for calculating the distance to the fault for single-phase, two-phase and three-phase faults is described in detail below. And, based on said algorithm developed for each of the three types of faults, a general equation valid for any of the three types of faults is obtained. SINGLE-PHASE FAULTS Figure 11 shows a schematic representation of the modeling of a single-phase fault in a network according to a possible embodiment of the invention. Following what fault theory dictates, the fault has been modeled with three sequence networks 1101, 1102, 1103.The network is represented by a strong end modeled as a voltage source E 110 with a source impedance Z. s (represented by three remote source impedances of zero sequence Zs0, direct sequence Zs1 and inverse sequence Zs2) and a weak end that can be represented as a source or load. In the diagram in Figure 11, the measurement unit is connected to the relay (point 85). Although physically there is only one fault point, when modeled as sequence networks, for each sequence network a fault point is modeled in the diagram, represented by py (1-p), as well as a measurement point 85. Measurement point 85 collects measurements of I a , YO b , YO c , V a , V b and V c At the measurement point, or alternatively remotely, if these measurements are sent to a control system, I0, I1, I2, V0, V1, V2, I are calculated. eq , V eq , V polfrom the measurements. Figure 11 shows: the zero-sequence impedance of the protected line ^ ^^ ; the direct sequence impedance of the protected line ^ ^^ ; the reverse sequence impedance of the protected line ^ ^^ ; the zero-sequence voltage at the measuring point ^ ^ ; the direct sequence voltage at the measuring point ^ ^ (source or load); the reverse sequence voltage at the measuring point ^ ^ (source or load); ^ ^ the zero-sequence current at the measurement point; the direct sequence current at the measurement point the reverse sequence intensity at the measuring point ^ ^ ; the zero sequence impedance of the remote source ^ ^^ ; the forward sequence impedance of the remote source ^ ^^ ; the reverse sequence impedance of the remote source ^ ^^ ; the fault resistance R ^and the fault intensity at the fault point ^ ^ . From Figure 11, it is clear that the equations involved are the following: ^ ^ − ^ ^ ∙ ^ ∙ ^ ^^ + ^ ^ − ^ ^ ∙ ^ ∙ ^ ^^ + ^ ^ − ^ ^ ∙ ^ ∙ ^ ^^ = ^ ^ ∙ 3 ∙ R ^ ^ ^^ = ^ ^^ ^ ^^ = ^ ^^ Developing the four equations we arrive at the following second order equation where the unknowns are the distance to the fault (^) and the fault resistance (R ^ ): ^ ∙ ^ ^^ [ ^ ^ ^ ^ ^^ ^^ ] ^ ^ ^ ( ^ ^^ + ^ ^ ) + ^ ∙ ^ ∙ −^ − ( ^ + ^ ∙ ^ ) ∙ ( ^ + ^ ) + ^ ∙ ^ ^^ ∙ ( ^ ^ + ^ ^ ∙ ^ ^ ) = [ −^ ^ + ^ ^ ∙ ( ^ ^^ + ^ ^^ )] ∙ 3 ∙ R^ where: V a is the voltage of phase A (where the single-phase fault has been assumed to have occurred); this voltage Va is equal to the sum of the zero-sequence voltage at the measuring point (V0), the direct sequence voltage at the measuring point (V1) and the reverse sequence voltage at the measuring point (V2): ^ ^ = ^ ^ + ^ ^ + ^ ^ Ia is the intensity of phase A which is equal to the sum of the zero sequence intensity at the measurement point (I0), the direct sequence intensity at the measurement point (I1) and the inverse sequence intensity at the measurement point (I2): ^ ^ = ^ ^ + ^ ^ + ^ ^ IN is the neutral intensity at the measurement point, which is equal to three times the homopolar current: ^ ^ = 3 ∙ ^ ^ K N is the compensation factor of the homopolar sequence that is calculated from ZL1 and ZL0: If we polarize with the following voltage V pol , the fault resistance will be eliminated (i.e., the effect of the fault resistance will be eliminated), so the only unknown in the equation will be the distance to the fault (note that choosing the polarization voltage ad hoc to eliminate the effect of the fault resistance is common in conventional methods for calculating distance to fault): 3 ∙ [ −^ ^ + ^ ^ ∙ ( ^ ^^ + ^ ^^ )] Polarizing by this voltage (it is the voltage used to eliminate the fault resistance from the equation) and taking the imaginary part, the previous equation will be as follows: Im{[−^ ^ + ^ ^ ∙ (^ ^^ + ^ ^^ )] ∗ ^ ^ ∙ (^ ^^ + ^ ^^ )} + ^ The equation is second order: ^ ∙ ^ ^ + ^ ∙ ^ + ^ = 0 where: ^ = Im{[−^ ^ + ^ ^ ∙( ^ ^^ + ^ ^^ ) ] ∗ ∙ ^ ^^ ∙ [−^ ^ - ( ^ ^ + ^ ^ ∙ ^ ^ ) ∙ ( ^ ^^ + ^ ^^ ) ]} ^ = Im{[−^ ^ + ^ ^ ∙ (^ ^^ + ^ ^^ )] ∗ ∙ ^ ^ ∙ (^ ^^ + ^ ^^ )} Note that, in the second order equation above, we can replace part of its expressions by the following parameters (calculable from the measurements taken previously): If the fault occurs in phase A, the following polarization voltage values ​​are obtained ^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ : ^ ^^^ = 3 ∙ [ −^ ^ + ^ ^ ∙ ( ^ ^^ + ^ ^^ )] ^ ^^ = ^ ^ ^ ^^ = ^ ^ + ^ ^ ∙ ^ ^If the fault occurs in phase B, the following polarization voltage values ​​are obtained ^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ : ^ ^^^ = 3 ∙ ^ ∙ [ −^ ^ + ^ ^ ∙ ( ^ ^^ + ^ ^^ )] ^ ^^ = ^ ^ ^ ^^ = ^ ^ + ^ ^ ∙ ^ ^ where: ^ is equal to 1∠120º (its value is known in advance). [Note that this “^” is different from the coefficient “a” that accompanies, along with the coefficients “b” and “c”, the previous second-order equation ^ ∙ ^ ^ + ^ ∙ ^ + ^ = 0]. If the fault occurs in phase C, the following polarization voltage values ​​are obtained ^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ :: ^ ^^^ = 3 ∙ ^ ^ ∙ [−^ ^ + ^ ^ ∙ ( ^ ^^ + ^ ^^ ) ] ^ ^^ = ^ ^ where: ^ is equal to 1∠120º. Returning to the second-order equation, since it is a quadratic equation, the distance to the fault will be equal to: As can be seen, there are two possible solutions. In homogeneous systems, one of these solutions is equal to the sum of the direct sequence impedance of the protected line Z L1 plus the forward sequence impedance of the remote source Z s1 In heterogeneous systems, one of the solutions is similar to the sum of the direct sequence impedance of the protected line Z L1 plus the direct sequence impedance of the source impedance Z S1 In both cases, the distance to the fault is equal to the minimum value between the two values: In the previous example, it has been assumed that the single-phase fault has occurred in phase A. If the fault occurs in phase B, the equations will be the same, except that instead of using the magnitudes of phase A, the magnitudes of phase B will be used. If the fault occurs in phase C, the magnitudes of phase C will be used. Once the distance ^ to the fault has been accurately calculated, the fault resistance is calculated. To do this, the following equation is solved: R ^ ^ ∙ ( ^ + ^ ) + ^ ∙ ^ ∙ −^ − ( ^ + ^ ∙ ^ ) ∙ ( ^ + ^ ) + ^ ^ ∙ ^ ^ ∙ ( ^ + ^ ∙ ^ =^ ^^ ^^ ^^ [ ^ ^ ^ ^ ^^ ^^ ] ^^ ^ ^ ^ ) 3 ∙ [−^ ^ + ^ ^ ∙ (^ ^^ + ^ ^^)] TWO-PHASE FAULTS Figure 12 shows a schematic representation of the modeling of a two-phase fault in a network according to a possible embodiment of the invention. The fault has been modeled with two sequence networks 1201, 1202. The network is represented by a strong end modeled as a voltage source E 120 with a source impedance Z s (represented by the direct sequence remote source impedances Z s1 and reverse sequence Z s2 ) and a modeled weak end that can be represented as a source or load. In the schematic in Figure 12, the measurement unit is connected to the relay (point 85). In Figure 12, the same variables and parameters denote the same concepts referred to previously, for example, in Figure 11. From Figure 12, the equations involved are as follows: ^ ^ − ^ ^ ∙ ^ ∙ ^ ^^ + ^ ^ ∙ R ^ = ^^ − ^ ^ ∙ ^ ∙ ^ ^^ ^ ^^ = ^ ^^ ^ ^^ = ^ ^^ Developing the four equations we arrive at the following second order equation, in which the unknowns are the distance to the fault (^) and the fault resistance (R ^ ): If we polarize with the following voltage, the fault resistance will be eliminated, so the only unknown in the equation will be the distance to the fault: [ ^^ − ^^ ∙ ( ^^^ + ^^^ )] Polarizing by this tension and taking the imaginary part, the previous equation will be as follows: The equation is second order: ^ ∙ ^ ^ + ^ ∙ ^ + ^ = 0 where: ^ = Im {[ ^ ^ − ^ ^ ∙ ( ^ ^^ + ^ ^^ )]∗ ∙ ^ ^^ ∙ [ - ( ^ ^ − ^ ^ ) − (^ ^ − ^ ^ ) ∙ ( ^ ^^ + ^^^ )]} ^ = Im {[ ^ ^ − ^ ^ ∙ ( ^ ^^ + ^ ^^ )]∗ ∙ ( ^^ − ^^ ) ∙ ( ^^^ + ^^^ )} Note that, in the second order equation above, we can replace part of its expressions by the following parameters (calculable from the measurements taken previously): If the fault occurs in the BC phases, the following polarization voltage values ​​are obtained ^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ : If the fault occurs in the AC phases, the following polarization voltage values ​​are obtained ^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ : ^ ^^^ = −^ ∙ 2 ∙ [−^ ^ + ^ ^ ∙ (^ ^^ + ^ ^^ )] ^ ^ ^ √3 ^ = ^ ∙ ^ ^ − ^ ∙ ^ ^ = j ∙ (^^ − ^ ^ ) 3 where: ^ is equal to 1∠120º, (its value is known in advance). [Note that this “^” is different from the coefficient “a” that accompanies, along with the coefficients “b” and “c”, the previous second-order equation ^ ∙ ^ ^ + ^ ∙ ^ + ^ = 0]. If the fault occurs in the BA phases, the following polarization voltage values ​​are obtained ^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ : ^ ^^^ = −^ ^ ∙ 2 ∙ [−^ ^ + ^ ^ ∙ ( ^ ^^ + ^ ^^ ) ] ^ ^^ = ^ ^ where: ^ is equal to 1∠120º, (its value is known in advance). Returning to the second-order equation, since it is a quadratic equation, the distance to the fault will be equal to: As can be seen, there are two possible solutions. In homogeneous systems, one of these solutions is equal to the sum of the protected line impedances Z L1plus the remote source impedance Z s1 In heterogeneous systems, one of the solutions is similar to the sum of the protected line impedance Z L1 plus the impedance of the source impedance Z S1 In both cases, the distance to the fault is equal to the minimum value between the two values: In the previous example, it has been assumed that the two-phase fault has occurred in phases B and C, so the magnitudes of phase A are used. If the fault occurs in phases A and B, the equations will be the same, except that instead of using the magnitudes of phase A, the magnitudes of phase C will be used. If the fault occurs in phases A and C, the magnitudes of phase B will be used. Once the distance to the fault has been accurately calculated, the fault resistance is calculated. To do this, the following equation is solved: THREE-PHASE FAULTS Figure 13 shows a schematic representation of the modeling of a three-phase fault in a network according to a possible embodiment of the invention. The fault has been modeled with a sequence network 1301. The network 1301 is represented by a strong end modeled as a voltage source E 130 with a source impedance Zs1131 and a weak end that can be represented as a source or load. Two voltage sources (of opposite values) are also represented, representing the pre-fault voltage (which was present before the fault) of direct sequence at the fault point V Fpref . In the diagram in Figure 13, the measurement unit is connected to the relay (point 85). To calculate the distance to three-phase faults, it is necessary to have measured the values ​​that were present before the fault, since the current that was present before the fault occurs has an effect on the calculation of the distance to the fault. When referring to I pura, the pre-fault values ​​are not being taken into account. The current that occurred before the fault has an effect on the distance calculation. Figure 14 shows a schematic representation of the modeling of a sequence network in a pre-fault situation according to a possible embodiment of the invention. The network is represented by a strong end 1401 modeled as a voltage source E 140 with a remote source direct sequence impedance Zs1141 and a weak end 1402 that can be represented as a source or load. In the schematic of Figure 14, the measurement unit is connected to the relay (point 85). In Figure 14, I 1pref denotes the direct sequence prefault intensity at the measurement point and V 1prefdenotes the direct sequence prefault voltage at the measurement point (source or load). Figure 15 shows a schematic representation of the modeling of a pure (i.e., after removing the prefault effect) three-phase fault in a network, according to a possible embodiment of the invention. The fault has been modeled with a sequence network. The network is represented by a strong end 1501 modeled as a remote source direct sequence impedance Z s1 and a weak end 1502 which can be represented as a source or load. In the diagram in Figure 15, the measuring unit is connected to the relay (point 85). In Figure 15, I1remote denotes the direct sequence fault current at the remote end in the case of a three-phase fault, with the prefault value, I, removed. 1pura denotes the intensity of the direct sequence fault at the measurement point with the prefault value, V, removed 1puradenotes the direct sequence fault voltage at the measuring point with the prefault value removed and V Fpref denotes the direct sequence prefault voltage at the fault point. As explained, measurements of Ia, Ib, Ic, Va, Vb, and Vc are collected at measurement point 85. At the measurement point, or alternatively remotely if these measurements are sent to a control system, I0, I1, I2, V0, V1, V2, Ieq, Veq, and Vpol are calculated from the measurements. In Figures 13 to 15, the same variables and parameters denote the same concepts as those referred to above, for example, in Figures 11 and 12. From Figures 13, 14, and 15, it is clear that the equations involved are as follows: ^ ^^^^^ − ^ ^^^^^ ∙ ^ ∙ ^ ^^ = ^ ^^^^^ ^ ^^^^^ − ^ ^^^^^ ∙ ^ ∙ ^ ^^ = ^ ^ ∙ R ^ − ^ ^^^^^ ^ ^ = ^ ^^^^^^^ + ^ ^^^^^Developing the four equations, we arrive at the following second order equation, in which the unknowns are the distance to the fault (^) and the fault resistance (R ^ ): If we polarize with the following voltage, we will be able to eliminate the fault resistance, so the only unknown in the equation will be the distance to the fault: Polarizing by this tension and taking the imaginary part, the previous equation will be as follows: If we consider that: ^ ^ = ^ ^^^^^ + ^ ^^^^^ ^ ^ = ^ ^^^^^ + ^ ^^^^^ The equation is second order: ^ ∙ ^ ^ + ^ ∙ ^ + ^ = 0 where: Note that, in the second-order equation above, we can replace part of its expressions by the following parameters (calculable from the measurements taken previously): ^ ^^ = ^ ^ ^ ^^ = ^ ^Returning to the second order equation, since it is a quadratic equation, the distance to the fault will be equal to: As can be seen, there are two possible solutions. In homogeneous systems, one of these solutions is equal to the sum of the protected line impedances Z L1 plus the remote source impedance Zs1. In heterogeneous systems, one solution is similar to adding the protected line impedance ZL1 plus the source impedance Zs1. In both cases, the distance to the fault is equal to the minimum value between the two: Once the distance to the fault has been accurately calculated, the fault resistance is calculated. To do this, the following equation is solved: From the three previous developments relating to the respective single-phase, two-phase and three-phase faults, in turn derived from the sequence network theory, it can be stated, in general, that the value of the distance to the fault in per unit values ​​^ is calculated using the following expression, in which the polarization voltage parameters Vpol, equivalent voltage Veq and equivalent current Ieq are calculated specifically depending on whether the fault is single-phase, two-phase or three-phase: Which is a second-order equation where: where ^ ^^^ is the bias voltage; ^ ^^ is the direct sequence impedance of the protected line; ^ ^^ is the equivalent intensity, which in the case of single-phase faults corresponds to the compensated intensity; ^ ^^ is the equivalent voltage; ^ ^^ is the forward sequence impedance of the remote source; ^ ^^is the direct-sequence impedance of the protected line. Note that the polarization voltage is used to remove the fault resistance from the previous equation; and the equivalent current and voltage are calculated, respectively, from the phase current and voltage measurements. The values ​​of ^ ^^^ , ^ ^^ and ^ ^^ They are calculated, with the measurements taken from the fault situation, specifically depending on whether the fault is single-phase, two-phase or three-phase. ^ ^^It is a value that must be calculated before the fault occurs. This value is calculated with two different values ​​of V1 (direct sequence voltage at the measurement point) and two different values ​​of I1 (direct sequence current at the measurement point). These values ​​are obtained when, in a non-fault situation, a variation in V1 or I1 occurs. When a variation occurs between these two magnitudes, two values ​​of V1 are obtained (one of these values ​​is prior to the variation and the other value is after the variation) and two values ​​of I1 (one of these values ​​is prior to the variation and the other value is after the variation).V1 and I1 are calculated from the measured values ​​of phase A current at the measurement point (Ia), phase B current at the measurement point (Ib), phase C current at the measurement point (Ic), phase A voltage at the measurement point (Va), phase B voltage at the measurement point (Vb) and phase C voltage at the measurement point (Vc). And, as explained in each specific case, mathematically there are two possible solutions: The correct solution is the minimum of the two: Multiplying the distance in unit values ​​with the line reactance gives the reactance X between the measurement point and the fault: ^ = ^ ∙ X ^^ where XL1 is the direct sequence reactance of the protected line. The value of the fault resistance R F It is calculated as follows: where R is the resistance seen from the measurement point to the fault point. The decision rule for determining whether the fault is within the zone is: If X < X ^ and R < R ^The fault has occurred within the range of the distance unit. Note that, to locate the fault on a plane, it is necessary to calculate R and X. The reactance X represents the distance to the fault (location). The weaker the end from which the line is protected, the greater the inaccuracy in the distance calculation in heterogeneous networks and in faults with non-zero resistance. This occurs because, when there are two ends feeding the fault, each end only measures its contribution (see, as an example, the electrical system in Figure 6). It will be demonstrated below that the value of the distance to the fault is the minimum of the two possible values. Let us assume that the coefficients ^, ^ and ^ are purely imaginary, that is, they do not contain a real part. In that case, the coefficients can be represented as follows: ∗ ^ = ^^ ^^^ ^ ^ ^ ^ ^ ∙ ^ ^^ ^ = ^ ^ ∗ ^ ^^^ ∙ ^ ^^ ∙ ( ^ ^^ + ^ ^^ ) With these coefficients, the distance to the fault is calculated: Replacing the coefficients: ^^ ^ + ^^^ ∙ ^^^ + ^^^ ^ ± ^ ^^^ + ^^^ ∙ ^^^ + ^^ ^ ( ) ^ ( ^^ )^ − 4 ∙ ^ ^^ ∙ ^ ^^ ∙ ( ^ ^^ + ^ ^^ ) ^ = 2 ∙ ^ ^^ ∙ ^ ^^ ^ ^^ ^^ + ^ ^^ ∙ ( ^ ^^ + ^ ^^ ) ^ ± ^ ^^ ^^ − ^ ^^ ∙ ( ^ ^^ + ^ ^^ ) ^ ^ = 2 ∙ ^ ^^ ∙ ^ ^^ Simplifying the previous expression, we arrive at the two possible distance values ​​being p1, p2: One of the two possible values ​​is always positioned at the opposite end (point between the equivalent voltage source of the strong end and the source impedance of the remote end), that is, at point (^ ^^ + ^ ^^ ). This value is the incorrect value and will always be greater than the other value. Therefore, the correct value will be the minimum value between the two: The above equations are fulfilled when the faults are frank, that is, whose resistive component is equal to zero. In the case of resistive faults, at the maximum value the distance is not exactly equal to (^ ^^ + ^ ^^), but it will be very close. In any case, the correct value will always be the minimum between the two values. An example is detailed below in which a single-phase fault has been simulated at 80% of the line (that is, if the line were, for example, 100 km long, the fault would be simulated at 80 km), and which has a fault resistance equal to 3 Ω. The simulation scheme is represented in Figure 10. It will be demonstrated that the algorithm of the invention, and the distance unit that implements it, are capable of identifying the distance to the fault (80%) and the fault resistance (3 Ω). The simulation has been carried out in Matlab version 2020b. For the simulation, the following values ​​have been chosen (imposed, by design): The direct and reverse sequence impedances of the line are: ^ ^^ = 6.989∠70.734º Ω and ^ ^^ = 22.436∠68.398º Ω The source impedance is: ^ ^^ = 9.579∠74.393º Ω The homopolar compensation factor is: ^ ^= 0.737∠ − 3.392º Ω With these simulated values, the fault currents and voltages collected are the following: ^ ^ = 963.507∠ − 11.549º A, ^ ^ = 367.176∠ − 52.603º A, ^ ^ = 104.37∠ − 129.452º A ^ ^ = 13.291∠25.458º kV , ^ ^ = 18.901∠ − 80.459º kV, ^ ^ = 19.52∠149.275º kV The direct and reverse sequence currents and voltages are: ^ ^ = 3.114∠ − 129.604º kV The neutral current is: ^ ^ = 1237∠ − 27.18º A The polarization voltage, the equivalent voltage and the equivalent intensity are: ^ ^^^ = 3 ∙ [ −^ ^ + ^ ^ ∙ ( ^ ^^ + ^ ^^ )] = 21.424∠51.414º ^^ , ^ ^^ = ^ ^ = 13.291∠25.458º ^^ , ^ ^^ = ^ ^ + ^ ^ ∙ ^ ^ = 1850∠ − 20.797º ^ The coefficients are equal to: ^ = ^ ∙ ^ ^^ ∙ ^ ^^ ^ = 1.81 × 10^ × 10 ^ The two measured distances are equal to: 2.374 0.801 The minimum between the two values ​​is 0.801, which shows that it correctly calculates the distance to the fault (note that a distance to the fault of 80% had been defined): ^ = min ( ^ ^ , ^ ^ ) = ^ ^ = 0.801 Now the fault resistance is calculated: It can be seen that the algorithm correctly determines the value of the fault resistance (note that a fault resistance equal to 3Ω had been defined). The following table shows the results obtained in other simulations, in different faults with and without fault resistance and in the case of single-phase, two-phase and three-phase faults: Type Fault point Calculated Fault (pu) Calculated resistance Distance resistance (Ω) (Ω) AG 0.8 0 0.8 -0.002 AG 0.8 3 0.801 2.957 BC 0.8 0 0.805 -0.004 BC 0.8 3 0.79 3.044 ABC 0.8 0 0.802 0.012 ABC 0.8 5 0.813 4.9 This demonstrates the excellent distance and resistance calculation accuracy of the proposed algorithm. In this text, the term “comprise” and its derivations (such as “understanding”, etc.) should not be understood in an exclusive sense, that is, these terms should not be interpreted as excluding the possibility that what is described and defined may include additional elements, steps, etc. In the context of the present invention, the term “approximately” and terms of its family (such as “approximate”, etc.) should be interpreted as indicating values ​​very close to those accompanying said term. That is, a deviation within reasonable limits with respect to an exact value should be accepted, because a person skilled in the art will understand that such deviation from the indicated values ​​may be unavoidable due to measurement inaccuracies, etc. The same applies to the terms “about”, “about” and “substantially”.The invention is obviously not limited to the specific embodiment(s) described, but also encompasses any variation that may be considered by any person skilled in the art (for example, in relation to the choice of materials, dimensions, components, configuration, etc.), within the general scope of the invention as defined in the claims.

Claims

CLAIMS 1.- A computer-implemented method for determining a distance at which a fault occurs in a protected power line of an electrical power system, where the power line is protected from a weak end of the power line, comprising: - measuring the following quantities at a measurement point (85) of the protected line from a weak end in a fault situation: intensity of phase A at the measurement point (I a ), phase B intensity at the measurement point (I b ), intensity of phase C at the measurement point (I c ), phase A voltage at the measurement point (V a ), phase B voltage at the measurement point (V b ) and phase C voltage at the measurement point (V c ); -from the previous measurements, calculate the following magnitudes of the power line: zero-sequence voltage at the measurement point (^ ^), direct sequence voltage at the measuring point (^ ^ ), reverse sequence voltage at the measuring point (^ ^ ), intensity of homopolar sequence at the measurement point (I 0 ), direct sequence intensity at the measurement point (I 1 ), reverse sequence current at the measuring point (I2), equivalent voltage (^ ^^ ), equivalent intensity (^ ^^ ) and bias voltage (^ ^^^ ); -calculate the distance at which a single-phase, two-phase or three-phase fault occurs on the protected power line using the following formula: where ^ ^^ is the direct sequence impedance of the protected line and is known; and ^ ^^ is the direct sequence impedance of the remote source and has been calculated before the fault occurs from direct sequence voltage values ​​at the measurement point V 1and direct sequence current at the measurement point I 1 at two different times in a no-fault situation; where the distance to the fault is the minimum of the following two possible solutions: where 2.- The method of claim 1, further comprising calculating, from the distance to the fault p, the reactance to the fault point from the measurement point: ^ = ^ ∙ X ^^ where X L1 is the direct sequence reactance of the protected line.

3. The method of any of claims 1 or 2, further comprising calculating, from the distance to the fault p, the fault resistance: where R is the resistance seen from the measurement point to the fault point.

4. The method of claims 2 and 3, further comprising: - comparing the reactance at the fault point from the measurement point X with a previously chosen reactive reach adjustment value XR ; - comparing the resistance seen from the measurement point to the fault point R with a previously chosen resistive reach adjustment value R R ; and -if X < X ^ and R < R ^ -determining that the fault has occurred within a distance unit; -optionally, actuating a switch to isolate the fault from the rest of the electrical power system. 5.- The method of any of the preceding claims, wherein, if the fault is single-phase: -if the fault occurs in phase A, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ are obtained: ^ ^^^ = 3 ∙ [ −^ ^+ ^ ^ ∙ ( ^ ^^ + ^ ^^ )] ^ ^^ = ^ ^ where: ^ ^ is the neutral intensity at the measurement point, which is ^ ^ = 3 ∙ ^ ^ ^ ^ ^ is the zero-sequence compensation factor, ^ ^ = ^ ^^ ^^ ^^ ^ ∙^ ^^ -if the fault occurs in phase B, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ are obtained ^ ^^^ = 3 ∙ ^ ∙ [ −^ ^ + ^ ^ ∙ ( ^ ^^ + ^ ^^ )] ^ ^^ = ^ ^ ^ ^^ = ^ ^ + ^ ^ ∙ ^ ^ where: ^ is equal to 1∠120º, and is known in advance, ^ ^ is the neutral intensity at the measurement point, which is ^ ^ = 3 ∙ ^ ^ ^ ^ ^ is the compensation factor d^ ^^ ^^ ^^ ^ and homopolar sequence, ^ ^ = -if the fault occurs in phase C, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ are obtained: ^ ^^^ = 3 ∙ ^ ^ ∙ [ −^ ^ + ^ ^ ∙ ( ^ ^^ + ^ ^^ )] ^ ^^ = ^ ^ where: ^ is equal to 1∠120º, and is known in advance, ^ ^ is the neutral intensity at the measurement point, ^ ^ = 3 ∙ ^ ^ ^ ^ ^ is the compensation factor ^ homopolar sequence action, ^ ^ = ^∙^ ^^ 6.- The method of any of claims 1-4, wherein, if the fault is biphasic: -if the fault occurs in the BC phases, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^and equivalent intensity ^ ^^ are obtained: -if the fault occurs in the AC phases, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ are obtained: ^ ^^^ = −^ ∙ 2 ∙ [−^ ^ + ^ ^ ∙ (^ ^^ + ^ ^^ )] where: ^ is equal to 1∠120º, and is known in advance - if the fault occurs in the BA phases, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ are obtained: ^ ^^^ = −^ ^ ∙ 2 ∙ [ −^ ^ + ^ ^ ∙ ( ^ ^^ + ^ ^^ )] √ ^ ^^ = ^ ∙ ^ ^ − ^ ^ ∙ ^ ^ = j 3 ∙ (^ ^ − ^ ^ ) 3 ^ ^^ = ^ ^where: ^ is equal to 1∠120º, and is known in advance. 7.- The method of any of claims 1-4, wherein, if the fault is three-phase, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ are obtained: ^ ^^ = ^ ^ ^ ^^ = ^ ^ where ^ ^^^^^ is the direct sequence fault voltage at the measuring point with the prefault value removed, ^ ^^^^^is the direct sequence fault current at the measuring point after the pre-fault value has been removed. 8.- A distance unit for determining the distance at which a fault occurs on a protected power line of an electrical power system, where the power line is protected from a weak end of the power line, comprising: - means for measuring the following quantities at a measuring point (85) of the protected line from a weak end in a fault situation: phase A current at the measuring point (Ia), phase B current at the measuring point (Ib), phase C current at the measuring point (I c ), phase A voltage at the measurement point (V a ), phase B voltage at the measurement point (V b ) and phase C voltage at the measurement point (V c); -means to calculate, from the above measurements, the following magnitudes of the power line: zero-sequence voltage at the measurement point (^ ^ ), direct sequence voltage at the measuring point (^ ^ ), reverse sequence voltage at the measuring point (^ ^ ), zero sequence current at the measuring point (I0), direct sequence current at the measuring point (I1), inverse sequence current at the measuring point (I 2 ), equivalent voltage (^ ^^ ), equivalent intensity (^ ^^ ) and bias voltage -means for calculating the distance at which a single-phase, two-phase or three-phase fault occurs on the protected power line using the following formula: where ^ ^^ is the direct sequence impedance of the protected line and is known; and ^ ^^is the direct sequence impedance of the remote source and has been calculated before the fault occurs from the direct sequence voltage values ​​at the measuring point V1 and the direct sequence current values ​​at the measuring point I1 at two different instants in a non-fault situation; where the distance to the fault value is the minimum of the following two possible solutions: where 9.- The distance unit of claim 8, further comprising means for calculating, from the distance to the fault p, the reactance to the fault point from the measurement point: ^ = ^ ∙ X ^^ where XL1 is the direct sequence reactance of the protected line.

10. The distance unit of any of claims 8 or 9, further comprising means for calculating, from the distance to the fault p, the fault resistance: where R is the resistance seen from the measurement point to the fault point.

11. The distance unit of claims 9 and 10, further comprising: - means for comparing the reactance to the fault point from the measurement point X with a pre-set reactive reach adjustment value X R ; - means for comparing the resistance seen from the measuring point to the fault point R with a pre-set resistive reach adjustment value R R ; and - means for, if X < X ^ and R < R ^ -determining that the fault has occurred within a distance unit; -optionally, actuating a switch to isolate the fault from the rest of the electrical power system.

12. The distance unit of any of claims 8-11, wherein, if the fault is single-phase: -if the fault occurs in phase A, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^^^ are obtained: ^ ^^^ = 3 ∙ [−^ ^ + ^ ^ ∙ (^ ^^ + ^ ^^ )] ^ ^^ = ^ ^ ^ ^^ = ^ ^ + ^ ^ ∙ ^ ^ where: ^ ^ is the neutral intensity at the measurement point, which is ^ ^ = 3 ∙ ^ ^ ^ ^ ^ is the zero-sequence compensation factor, ^ ^ = ^ ^^ ^^ ^^ ^ ∙^ ^^ -if the fault occurs in phase B, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ are obtained ^ ^^^ = 3 ∙ ^ ∙ [ −^ ^ + ^ ^ ∙ ( ^ ^^ + ^ ^^ )] ^ ^^ = ^ ^ ^ ^^ = ^ ^ + ^ ^ ∙ ^ ^ where: ^ is equal to 1∠120º, and is known in advance, ^ ^is the neutral intensity at the measurement point, which is ^ ^ = 3 ∙ ^ ^ ^ ^ ^ is the fact ^ ^^ ^^ ^^ ^ or zero-sequence compensation, ^ ^ = ^∙^ ^^ -if the fault occurs in phase C, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ are obtained: ^ ^^^ = 3 ∙ ^ ^ ∙ [−^ ^ + ^ ^ ∙ (^ ^^ + ^ ^^ )] ^ ^^ = ^ ^ where: ^ is equal to 1∠120º, and is known in advance, ^ ^ is the neutral intensity at the measurement point, ^ ^ = 3 ∙ ^ ^ ^ ^ ^ ^ is the zero-sequence compensation factor, ^ ^ = ^ ^^ ^^ ^^ ^ ∙^ ^^13.- The distance unit of any of claims 8-11, wherein, if the fault is biphasic: -if the fault occurs in the BC phases, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ are obtained: -if the fault occurs in the AC phases, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ are obtained: ^ ^^^ = −^ ∙ 2 ∙ [ −^ ^ + ^ ^ ∙ ( ^ ^^ + ^ ^^ )] where: ^ is equal to 1∠120º, and is known in advance - if the fault occurs in the BA phases, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ are obtained: ^ ^^^ = −^ ^ ∙ 2 ∙ [ −^ ^ + ^ ^ ∙ ( ^ ^^ + ^^^ )] ^ ^ ^ ^ √3 ^ = ^ ∙ ^ − ^ ∙ ^ ^ = j ∙ (^ − ^ ) 3 ^ ^ ^ ^^ = ^ ^ where: ^ is equal to 1∠120º, and is known in advance. 14.- The distance unit of any of claims 8-11, wherein, if the fault is three-phase, the polarization voltage values ​​^ ^^^ , equivalent voltage ^ ^^ and equivalent intensity ^ ^^ are obtained: ^ ^^ = ^ ^ where ^ ^^^^^ is the direct sequence fault voltage at the measuring point with the prefault value removed, ^ ^^^^^ is the direct sequence fault intensity at the measurement point after the prefault value has been removed. 15.- A protection system for an electrical power system comprising a distance unit according to any of claims 8-14.

Images