Method for determining adjustment coefficients of a measuring system

Abstract

A measuring system (5) is proposed for measuring current values of an alternating current flowing through a conductor (1) and correcting the measured current values for capacitive interference. This system comprises a Rogowski coil (2) designed to induce a voltage by alternating current, a signal processing device (7) for detecting and processing values of the voltage induced in the Rogowski coil, a device (6) for determining a current value by integrating the voltage processed by the signal processing device, and a device (6) for correcting the current value for capacitive interference. The device (6) for correcting the current value is designed to perform the correction using an approximation (UADCcap1(t)) for a voltage drop caused by the capacitive interference.The correction according to the invention improves measurement accuracy without requiring additional hardware. This allows measuring devices constructed with Rogowski coils to be made smaller and more compact.

Description

[0001] The invention relates to a method for determining adjustment coefficients of a measuring system formed with a Rogowski coil for measuring current values of an alternating current flowing through a conductor and correcting the measured current values with respect to capacitive interference.

[0002] Rogowski coils are used to measure alternating currents (AC current measurement). The voltage at the output of the Rogowski coil's measuring winding is used as a sensory variable. The voltage at the output of the Rogowski coil's measuring winding is influenced not only by the time-varying magnetic field of the current to be measured (current in the conductor to be measured (primary current)), but also by electric fields - especially alternating electric fields. The main source of the alternating electric fields is the conductor (primary current) itself, as it is at mains potential with respect to the measuring circuit (e.g., at a voltage of 230 Vrms and a frequency of 50 Hz; Vrms: root mean square of the alternating voltage) (see Fig. 1).

[0003] In general, the conductive winding of the Rogowski coil forms a more or less complex capacitive structure with all the surrounding conductive surfaces. Due to the large distances between the winding of the Rogowski coil and the conductive surfaces (e.g., the surface of the primary conductor) compared to discrete capacitors, the corresponding coupling capacitances are usually very low. However, since the potential difference between these surfaces is very large compared to the measuring voltage at the ends of the Rogowski coil, even small coupling capacitances can lead to significant measurement errors when measuring current with Rogowski coils under normal operating conditions in low-voltage networks.

[0004] This problem is traditionally solved by using shielding surfaces, e.g. made of expanded metal, by using an additional shielding turn made of thin wire or strips of shielding material, or by shielding the entire sensor housing. If these shielding surfaces are kept at the measuring potential of the coils by a suitable electrical connection to the measuring circuit, the shielding surfaces provide the necessary "countercharges" on their surface to shield the electric field. This means that the surface of the Rogowski coil does not have to be recharged, and no capacitive interference occurs at the measuring outputs of the Rogowski coil. Since good shielding must not influence the time-varying magnetic field in the Rogowski coil and at the same time should not significantly increase the space required by the Rogowski coil, there are certain limits to the use of capacitive shields in small Rogowski coils.When using shielded windings, it is important to note that both sides must not be connected to earth or reference potential (to avoid circulating currents).

[0005] In the past, larger Rogowski coils were often used to measure larger currents. For this purpose, Rogowski coils with a significantly larger diameter (diameter of the ring) but a similar ring height were used. Generally, these larger Rogowski coils can be designed to achieve a better signal-to-noise ratio. In simple terms, increasing the distance between the winding of the Rogowski coils and the current conductor (primary current) leads to a reduction in the coupling capacitance. However, this is only possible if the necessary space is available in the electrical power distribution system. In general, alternating currents with an effective value of less than 100 A can also be measured in low-voltage applications using a toroidal current transformer. In addition to some advantages over Rogowski coils, these also have some disadvantages, such as higher costs, greater weight and a larger design or significantly larger volume.

[0006] US 2015 / 0 362 532 A1 (Chartouni; Itron France) 17.12.2015 describes a measuring system for measuring current values of an alternating current flowing through a conductor and correcting the measured current values with regard to capacitive interference.

[0007] US 2020 / 0 141 981 A1 (Hurwitz; Analog Devices International Unlimited Company) 07.05.2020 describes an electrostatic shield for providing electrostatic shielding for a current sensing coil.

[0008] The invention aims at efficiently correcting interference coupling in measuring systems based on Rogowski coils without increasing the space requirements of conventional systems.

[0009] The problem is solved by a method according to claim 1. An advantageous further development is specified in the subclaim.

[0010] According to the invention, a method is proposed for determining adjustment coefficients of a measuring system for measuring current values of an alternating current flowing through a conductor and for correcting the measured current values for capacitive interference. This can also involve a system with multiple phases or conductors, and the capacitive interference from the other phases on the conductor in question can be taken into account. The measuring system can be designed as a measuring device, e.g., as an energy monitoring device or PMD (Power Monitoring Device). However, the system can also be designed with multiple components, e.g., by transmitting measured values to a central evaluation point, where the correction is performed.

[0011] The measuring system comprises a Rogowski coil designed to induce a voltage through alternating current. The monitored conductor typically passes through the Rogowski coil. The system also includes a signal processing unit for recording and processing values of the voltage induced in the Rogowski coil. This unit is formed, for example, by a circuit through which the induced voltage is tapped, filtered, and possibly amplified, before being fed to an analog-to-digital converter. Post-processing of the digital signal may also be provided. Furthermore, the system includes a device for determining a current value by integrating the voltage processed by the signal processing device, e.g., a microcontroller.

[0012] Preferably, the system is formed with an analog-to-digital converter and provides digital integration of the processed voltage. However, the invention is also applicable in principle to systems based on a Rogowski coil with analog integration. It is conceivable that analog systems could be supplemented by a (possibly external) current correction device that corrects the current values obtained by analog integration.

[0013] According to the invention, the measuring system also includes a device for correcting the current value with respect to capacitive interference, which is preferably identical to the device for determining a current value by integration. This device is configured or programmed to perform the correction with the aid of an approximation. (UADCkap1(t)) for a voltage drop caused by capacitive interference.

[0014] This approximation (UADCkap1(t)) has a form in which it calculates the voltage drop caused by the capacitive interference couplings as a function of at least one adjustment coefficient (C K1,1 ), wherein the at least one adjustment coefficient (C K1,1 ) can be determined by one or more calibration measurements.

[0015] According to an embodiment not according to the invention, the approximation may contain at least one term which is proportional to the product of the adjustment coefficient (C K1,1 ), the derivation of a voltage attributable to the monitored conductor (dUμ(t)dt) and a resistance assigned to the Rogowski coil (R i_RoGo ). A number of phases can be considered, and a corresponding term can be provided for each phase considered.

[0016] According to an embodiment not according to the invention, the approximation may have the form −14∑μ(dUμ(t)dtCK1,μ)⋅Ri_RoGo where the index µ runs over the considered phases, C K1,µ is a balancing coefficient, dUμ(t)dt a derivative of a voltage attributable to the monitored conductor and (R i_RoGo ) a resistance (R i_RoGo ).

[0017] According to the invention, the correction by means of the approximation (UADCkap1(t)) for a voltage drop caused by the capacitive interference couplings in such a way that an expression obtained from it by integration (if necessary with sign adjustment) is used in a formula for current values (I1(t)) of the monitored conductor.

[0018] In general, the form of the approximation (UADCkap1(t)) is designed in such a way that its integral is a linear function term depending on a voltage assigned to the conductor (U µ(t)) and, if applicable, other phases considered in the approximation. This has the advantage that calibration measurements for determining coefficients result in a linear equation or a linear system of equations that is comparatively easy to solve.

[0019] According to an embodiment not according to the invention, the expression for current values (I1(t)) of the conductor can have the form l1(t)=14∑μ=13(Uμ(t)CK1,μ)⋅RiRoGo+1M11∫0tUADC1(τ)+l1(0) where the index µ runs over the considered phases, C K1,µ is a matching coefficient, U µ (t) a voltage attributable to a conductor (the conductor under investigation or another phase considered), R i_RoGo a resistor associated with the Rogowski coil and ∫0tUADC1(τ) is the integral of a voltage induced in the Rogowski coil. The term "integral of a voltage induced in the Rogowski coil" refers to voltage values that can be processed using the signal processing device after induction in the Rogowski coil.

[0020] The inventive method for determining adjustment coefficients (C K1,1 ) of a measuring system using calibration measurements is characterized by the fact that an expression of the form l→(t)=(cK11cK12cK13cK21cK22cK23cK31cK32cK33)∗U→(t)+(S11S12S13S21S22S23S31S32S33)∗∫0tU→ADC(τ)+I→(0) is assumed, and the adjustment coefficients (C K1,1) are determined by measurements at known currents applied to the conductor and, if applicable, other phases. This involves the use of simplified expressions, e.g., by considering only one conductor or neglecting secondary diagonals of the matrices. The measurements are preferably carried out at low voltages applied to the conductor or the other phases. (U→(t)) made.

[0021] The invention has the advantage that additional fixture elements, which increase space requirements in conventional systems, are not required. Furthermore, a mathematical expression can be provided that allows the correction to be carried out simply, efficiently, and with minimal effort.

[0022] The invention is described in more detail below within the framework of an exemplary embodiment. Fig. 1: a sketch of the formation of capacitive interference on the signal of Rogowski coils, Fig. 2: a schematic representation of a measuring device (power monitoring device) for the acquisition of measurement data related to an electrical circuit, Fig. 3: a simple electrical equivalent circuit for the combination of a Rogowski coil and an antialiasing low-pass filter, Fig. 4: a simplified electrical equivalent circuit for the combination of a Rogowski coil and an antialiasing low-pass filter and Fig. 5: Simulation result for capacitive coupling based on the equivalent circuit diagram from Fig. 3.

[0023] Fig. Figure 1 shows a conductor 1 surrounded by a Rogowski coil 2. Electric field lines 3 are drawn to illustrate the effect of fields generated by the conductor 1 on the Rogowski coil 2, which influences the voltage signal induced on signal lines 4 of the Rogowski coil 2.

[0024] More specifically, the alternating electric field that develops between the primary conductor 1 and the Rogowski coil 2, which is used for current measurement, causes potential fluctuations on the signal lines 4 or the signal output of the Rogowski coil relative to ground potential. These potential fluctuations can lead to various interferences during current measurement. These are referred to below as capacitive interference. This is a common term that derives from the fact that the surface of the conductor and the winding of the Rogowski coil form a capacitor. In addition to overloading or overdriving the inputs of the measuring circuit, these capacitive interferences primarily lead to differential mode interference in the measuring circuit.

[0025] Power monitoring devices are often referred to as PMDs (Power Monitoring Devices). Compact PMDs are primarily used to measure electrical power and thus current and voltage at line frequency.

[0026] Fig. Figure 2 shows a PMD 5 for acquiring measurement data related to an electrical circuit. This can also be configured to determine consumption data. Determining consumption data can, for example, be designed for standardized billing of energy consumed (e.g., according to the MID Directive EN 50470-1 / 3 or the standards IEC TR 63213 or IEC 61557-12).

[0027] The PMD 5 is connected to the circuit or measuring network to be monitored and receives measurement data recorded by a sensor (in this case, a Rogowski coil) and processed by measurement electronics 7. The measurement electronics usually comprise a low-pass filter 71 (antialiasing), an analog-to-digital converter or ADC 72, and a filter 73 for post-processing the digitized data, e.g., a high-pass filter. These components process the sensor data in the order listed above (see also DE 102015216981 B4). The processed measurement data is further processed by a microcontroller or MCU 6 (e.g., calculation of consumption data). The MCU 6 is also provided with an interface 9, via which data can be transmitted or read out. A power supply or supply circuit 8 is provided for the power supply of the MCU 6.This is powered from the monitored circuit and supplies 3.3 V DC voltage to the MCU 6. To safeguard the supply, PMDs usually have an additional power supply for the supply circuit.

[0028] If Rogowski coils are used for current measurement at line frequency, these coils are often designed such that the internal impedance of the Rogowski coil is dominated by the ohmic series resistance of the coil at line frequency. At the same time, the impedance of the anti-aliasing filter 71 should be selected to be significantly greater than the ohmic series resistance of the Rogowski coil. To avoid negative influences on the signal quality of the Rogowski coil, e.g., due to temperature drift, the value of the "ohmic component" of the anti-aliasing low-pass filter is generally selected to be significantly larger than the value of the ohmic series resistance of the Rogowski coil (RAA > Ri_Rogo → 1 / (ωmess*C)> RAA > Ri_Rogo) (see Equation 1). The capacitance CK of the capacitive coupling and the values RAA and CAA of the anti-aliasing low-pass filter result in the following relationship: 12πf⋅ck>>12πf⋅cAA>>RAA>Ri_Rogo

[0029] If the value of the impedance of the capacitive coupling between the primary conductor and the Rogowski coil in the range of the measuring frequencies (in the case of PMDs or energy monitoring devices, measurements are mainly taken at mains frequency) is much larger than the value of the ohmic series resistance of the Rogowski coil, the equivalent circuit can be Fig. 3. This simplified equivalent circuit is shown in Fig. 4 shown.

[0030] With the equivalent circuit diagram from Fig. 4 and the approximation from Eq. 1, the following approximate formula can be given for the current through the coupling capacitance ICK=U⋅jωCK

[0031] The underscore makes it clear that these are complex quantities.

[0032] Thus, the capacitive couplings between conductor 1 and Rogowski coil 2 at the analog-to-digital converter (ADC) 72 approximately lead to the following voltage drop: U_ADC=U_ADC_p−U_ADC_n=−14U_⋅jωCK⋅Ri_Rogo

[0033] Ri_RoGo is the value of the ohmic series resistance of the Rogowski coil. Fig. The simulation result shown in Figure 5 clearly confirms the relationship in Eq. 3 through its frequency response.

[0034] The relationship in Eq. 3 can be extended to include the influence of all phases of the distribution system (since these are often located in close proximity to each other): U_ADC_1=U_ADC_p_1−U_ADCn1=−14∑μ3(U_μ⋅jωCK1,μ)⋅Ri_RoGo

[0035] This results in the following relationship for the voltage at the input of the measuring system or the analog-to-digital converter (ADC) in the time domain: UADkap1(t)=−14∑μ=13(dUμ(t)dtCK1,μ)⋅Ri_RoGo

[0036] This includes: CK1,1>>CK1,2; CK1,1>>CK1,3

[0037] For the total voltage at the output of the combination of a Rogowski coil 2 and an anti-aliasing low-pass filter 71, and thus at the input of the analog-to-digital converter (ADC) 72, the equation Eq. 7 follows. UADC1(t)=UADkap1(t)+URoGo1(t)=−14⋅RiRoGo∑μ=13dUμ(t)dtCK1,μ+M11⋅dI1(t)dt

[0038] To analyze the current to be measured through Rogowski coil 2, the output signal of Rogowski coil 2 is usually subjected to analog or digital integration. The usefulness of integrating the output signal of Rogowski coil 2 or the input signal of the analog-to-digital converter ADC 72 can be directly derived from Equation 7. This integration yields the following relationship for the current, assuming that inductive interference can be neglected: I1(t)=14∑μ=13(Uμ(t)CK1,μ)⋅RiRoGo+1M11∫0tUADC1(τ)+I1(0)

[0039] As can be seen from Eq. 7 and Eq. 8, in order to correct the capacitive interference, the values of the voltages of the conductors which are located in the direct vicinity of the Rogowski coil 1 in a multi-phase system must be known (in the following, the equations are based on a three-phase system, e.g. three conductors, each of which - as in Fig.1 for a conductor - can be surrounded by a Rogowski coil). Since the values of the coupling capacitances from, for example, phase 2 to coil 1 are very often much lower than the values of the coupling capacitances from, for example, phase 1 to coil 1 due to geometry, these couplings can then be neglected. Since the values of the capacitances CK1,µ are sometimes dependent on manufacturing tolerances, it is advisable to determine the values of the coefficients for the digital correction as part of an adjustment and, in this case, especially when adjusting the voltage measurement of the PMD. During operation of the PMD, the results of the voltage measurement and the adjustment coefficients cK1,µ stored in the measuring system are then used to correct the current values obtained from the integration of the output voltage of the Rogowski coil.To facilitate a clear assignment of the individual effects, the calibration coefficients cK1,µ should be determined with no current flowing, if possible (when calibrating the measurement channels for voltage measurement, only very small currents should flow, measured against the nominal current). Accordingly, only very small voltages should be used when calibrating the measurement channels for current measurement. I→(t)=(cK11cK12cK13cK21cK22cK23cK31cK32cK33)*U→(t)+(S11S12S13S21S22S23S31S32S33)*∫0tU→ADC(τ)+I→(0)

[0040] The multiplication is marked with “*” here to indicate that it is an operation with matrices. U→(t) is the vector of the voltages of the individual phases.

[0041] It's important to note that the values of the secondary diagonal elements in both the CK matrix and the S matrix are often much smaller than the main diagonal elements and can therefore be neglected. This significantly simplifies the calculation of the capacitive coupling correction and can be easily performed on a microcontroller 6 or comparable embedded hardware.

[0042] The determination of the coupling coefficients is illustrated below using the simplest case, namely a single conductor. Equation 9 then reduces to: I(t)=cK⋅U(t)+S⋅∫0tUADC(τ)+I(0)

[0043] A predetermined, generated test current I(t) is applied. The voltage U(t) of the conductor is measured or is also known. The microcontroller 6 calculates the voltage U induced and processed in the Rogowski coil. ADC integrated, which ∫0tUADC(τ) for time t. To determine the two coefficients c K Two equations are required for S and S. For this purpose, the quantities of Eq. 10 can be determined for different times t1 and t2, or different test currents I(t) can be used.

[0044] The correction according to the invention improves measurement accuracy without requiring additional hardware. This allows measuring devices constructed with Rogowski coils to be smaller and more compact.

Claims

[1] Method for determining adjustment coefficients (C K1,1 ) a measuring system (5) for measuring current values of an alternating current flowing through a conductor (1) and correcting the measured current values with respect to capacitive interference, comprising - a Rogowski coil (2) designed to induce a voltage by alternating current, - a signal processing device (7) for detecting and processing values of the voltage induced in the Rogowski coil, - a device (6) for determining a current value by integrating the voltage processed by the signal processing device, and - a device (6) for correcting the current value with respect to capacitive interference, wherein - the device (6) for correcting the current value is designed to carry out the correction by means of an approximation (UADCkap1(t)) for a voltage drop caused by capacitive interference, - where the approximation (UADCkap1(t)) the voltage drop caused by the capacitive interference coupling as a function of at least one adjustment coefficient (C K1,1 ), wherein the at least one adjustment coefficient (C K1,1 ) can be determined by means of a calibration measurement, - where an expression of the form I→(t)=(cK11cK12cK13cK21cK22cK23cK31cK32cK33)*U→(t)+(S11S12S13S21S22S23S31S32S33)*∫0tU→ADC(τ)+I→(0) is assumed, and - the adjustment coefficients (C K1,1 ) can be determined by measurements with known currents applied to the conductor (1) and, if applicable, other phases. [2] Method according to claim 1, characterized by that the measurements are carried out at low voltages (Ü(t)) applied to the conductor (1) or the other phases.

Images