Nv-centre-based current sensor with correction for fluctuations in pump-radiation intensity and detection of current direction

Abstract

The present invention relates to a device for measuring a line current. Said device comprises at least three sensor elements having paramagnetic centres, a pump-radiation source, an optical system, and at least three photodetectors which function as part of receiving systems and are each associated with a sensor element. In addition, the device contains two magnetic field sources which generate different bias magnetic fields for each sensor element. The pump-radiation source emits pump radiation into the optical system, which distributes the pump radiation to the sensor elements and thereby irradiates the paramagnetic centres with the pump radiation. These centres emit fluorescence radiation, the intensity and phase shift of which depend on the AC drive signal of the evaluation and control device and on the magnetic flux density acting upon them. The fluorescence radiation is separated from the pump radiation by the optical system and guided onto the photodetectors. The evaluation and control device generates a single-frequency sinusoidal AC drive signal having a frequency of 22 MHz (-4 MHz / +7 MHz), or within a range of 18 MHz to 29 MHz. The lock-in amplifiers are coupled to the sensor elements and evaluate the output signal of the photodetectors with respect to the AC drive signal in order to produce measurement values. Based on these measurement values, the evaluation and control device determines an estimated value for the magnitude of the line current in the line.

Description

[0001] NV-CENTER BASED CURRENT SENSOR WITH CORRECTION OF PUMP RADIATION INTENSITY FLUCTUATIONS AND CURRENT DIRECTION DETECTION

[0002] Field of invention

[0003] The invention relates to a current sensor based on NV centers with correction of pump radiation intensity fluctuations and current direction detection.

[0004] General Introduction

[0005] An electricity grid serves to transport electricity from producers to consumers. The largest electricity grid in the world is the synchronous grid of continental Europe, which encompasses 24 countries. The German electricity grid is part of this larger grid and operates at the same frequency, 50 Hz. This document will focus on the German grid, which can be divided into four categories according to typical voltages [2]:

[0006] Maximum voltage: 220kV-380kV

[0007] High voltage: 60kV - HOkV

[0008] Medium voltage: 6kV - 30kV

[0009] Low voltage: 230V - 400V

[0010] In recent decades, the share of renewable energies (especially solar and wind) in Germany's annual electricity generation has increased significantly (Fig. 1). Parallel to this increase, the total number of solar and wind power plants has also risen ([3], [4]), leading to considerable structural changes in the electricity grid [5]. This trend is expected to continue in the coming years and decades. However, adjustments are not only taking place on the generation side.Total electricity consumption is expected to increase from 595 TWh in 2018 to 658 TWh in 2030 (+11%), due to the increasing use of electric vehicles, heat pumps, and green hydrogen [7]. While the expected changes on the consumer side generally only require the expansion of transmission grids (see Figure 1) with capacity at all grid levels, developments on the generation side necessitate a transition from the current grid structure to a smart grid [5]. A smart grid ensures grid stability at all levels with low losses, high quality, security of supply, and safety. The need for a smart grid structure arises from a variety of changes in the electricity grid.One of these is the aforementioned increase in solar parks and wind power plants that feed energy into the medium- and low-voltage levels of the power grid (230 V - 30 kV) ([8, 9], Fig. 19). Private households with small solar installations also act as both consumers and producers. To date, the medium- and low-voltage grid levels in particular are unable to provide the data required for the development of such a smart grid structure [9]. Sensors for monitoring energy flow and other parameters are generally located in substations that transform the voltage from one grid level to another ([9, 10]). To meet the demand, the measurement capacities at all grid levels must be increased.

[0011] High-voltage overhead power lines

[0012] The majority of the power grid, from medium voltage to the highest voltage level, consists of overhead lines. This document describes the cross-section and key parameters of this type of power line. Figure 2a shows two cross-sections of overhead lines. The left cross-section, consisting of a steel core and an aluminum sheath, is the most common. An overview of common cables with this cross-section and their electrical and mechanical properties was taken from reference

[0012] and is shown in Figure 2b. It can be seen that maximum effective currents of several hundred amperes up to 2000 A are reached, and magnetic fields of several mT are generated near the cable.

[0013] State-of-the-art current sensors for high-voltage applications

[0014] In the extra-high voltage network level, not only are the voltages high, but the currents can also reach several thousand amperes. Current sensors for this network level must therefore have a large measuring range and high galvanic isolation. In addition, the sensors must meet internationally defined accuracy classes. A table of these classes is shown in Fig. 20. It can be seen that for each class, the tolerance band over the measuring range is strictly limited. The same applies to the phase shift. Existing sensors generally meet accuracy classes 0.2–0.5 (

[0013] ,

[0014] ). There are two main concepts for such sensors, including inductive sensors and optical sensors.

[0015] Inductive sensors

[0016] The most commonly used current sensors in the highest network level are essentially transformers that convert the high line current on the primary side into a much lower line current on the secondary side, which is then measured (Fig. 3a). This type of current sensor can measure alternating currents up to 4 kA, typically achieving a high accuracy class of 0.2–0.5

[0013] and has an expected maintenance-free lifetime of > 30 years

[0014] . However, with increasing voltage, production costs increase exponentially due to the increased insulation effort

[0015] . At voltages above 380 kV, this type of sensor has a total height of > 4.5 m

[0014] .

[0017] Optical Sensors New safety concepts and increasing digitization created a need for current sensors with higher bandwidth, very linear characteristic curve and better integrability into a digital system environment

[0015] , This led to the development of optical current sensors.

[0018] Optical current sensors have many advantages

[0015] :

[0019] • Galvanically separated,

[0020] • The sensor is not affected by electromagnetic noise,

[0021] • No direct electricity consumption (energy consumption) via the power grid,

[0022] • High bandwidth,

[0023] • Low volume and weight, and

[0024] • Cost advantages through savings in insulation material.

[0025] Existing optical sensor concepts utilize, for example, the Faraday effect, achieve an accuracy class of 0.2 s

[0013] , have a bandwidth of up to 6 kHz, and can measure short-term RMS currents of up to 63 kA (e.g.,

[0013] ). A schematic diagram of such an optical current sensor is shown in Fig. 3b. However, existing optical sensors generally have problems with harsh environments. They are more sensitive to temperature fluctuations and vibrations and have an undefined maintenance-free lifetime

[0015] .

[0026] State of the art

[0027] The intensity profile of fluorescence radiation from nitrogen-vacancy centers as a function of the magnetic field without the use of microwaves is known from the publication ARNE WICKENBROCK ET AL, "Microwave-free magnetometry with nitrogen-vacancy centers in diamond", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, June 9, 2016 (2016-06-09), XP080965843 DOI: 10.1063 / 1.4960171. DE 10 2019 128 932 A discloses a sensor based on nitrogen-vacancy centers that measures a phase shift. From DE 10 2020 119 414 Al, a microwave-free NV center-based sensor with phase shift measurement is known. From US 2017328965 Al and WO 2019 108 781 Al, magnetometers without NV centers are known. From US 2015 090 033 Al (Budker et al.), an NV center-based gyroscope is known. From US 2018 0 203 080 Al, a spectrometer based on magnetic resonance is known.From WO 2020 239 172 Al, the monolithic CMOS integration of NV centers for use in MOEMS (micro-optical-microelectronic-micromechanical systems), MOMS (micro-optical-micromechanical systems), MOS (micro-optical systems), and MOES (micro-optical-microelectronic systems) is known. From WO 2020 089 465 Al, microwave-free sensor technology based on NV centers with compensating control at a signal extreme is known. From WO 2021 089 091 Al, various applications of microwave-free sensor technology are known. From the as yet unpublished German patent application DE 102024 100474.6, the use in power distribution networks is known. From DE 11 2021000 209 Al, various uses and microsystems for such sensor devices as presented here are known.From DE 102021 132 780 Al, various aspects of the A VT of such systems and the production and properties of HDNV diamonds containing NV centers are known. From DE 10 2020 134883 Al, an LED with NV centers is known. From the still unpublished DE 10 2024 100466 Al, uses in energy technology are known. From DE 102022 121444 Al, a particularly good manufacturing process of the system used here, consisting of an optical fiber and a sensor element with a high Cpk value, is known. From DE 102022 005 094 Al, the use of such devices for motor control is known. From DE 102023 122 667 Al, the use of such devices in a sensor head is known. The use of coplanar lines and their application in spectrum analyzers is known from the still unpublished DE 10 2023 122664.9. The application in penetrating sensor heads is known from the still unpublished DE 102023 134 058.1.In the paper "ZHNG, Jixing et al. 'A pulsed lock-in method for DC ensemble nitrogen-vacancy center magnetometry' Diamond and Related Materials, 2022, Vol. 125, p. 109035," a method is presented that proposes combining the advantages of lock-in detection and pulsed techniques to increase the sensitivity of nitrogen-vacancy center magnetometry. In the paper "Chatzidrosos, Georgios, et al. 'Fiberized diamond-based vector magnetometers', Frontiers in Photonics, 2021, Vol. 2, p. 732748," two fiber-optic vector magnetic field sensors based on nitrogen-vacancy centers in diamond are proposed. According to this article, these sensors are expected to achieve a magnetic sensitivity of less than 1 nT / VHz and are characterized by a compact design, high photon yield, and minimal sensor-sample distance.All these publications and the related intellectual property rights and applications for intellectual property rights have in common that they typically use an AC control signal for modulating the pump radiation, which is PWM-modulated. The proposal therefore aims to provide a solution for the sensitivity to fluctuations in the pump radiation intensity and a method for determining the sign of the line current in a line. This problem is solved by the independent claims. Further embodiments are the subject of dependent claims.

[0028] Diamond powder magnetic field sensor

[0029] The diamond-based, all-optical magnetic field measurement technique described in this document uses diamond powder with a high density of nitrogen vacancy centers (NV centers) as the sensor element. The NV centers are illuminated with green light and then emit red photons (fluorescence). The amount of red fluorescence and its temporal dynamics depend on the strength of the local magnetic field. The fluorescence is isotropic, meaning it depends only on the strength of the magnetic field |B| and not on the direction of B*

[0017] . This is the key effect used for the optical measurement of magnetic fields, according to the technical principles presented in this document. In the QT-DMFS (Fig. 4) from Quantum Technologies GmbH, the diamond powder is attached to the tip of an optical fiber. The fiber directs the green excitation light onto the diamond powder, and the red fluorescence is collected by the fiber and directed to a photodetector.This makes the QT-DMFS from Quantum Technologies GmbH a robust, microwave-free and purely optical magnetic field sensor available on the market.

[0030] Intensity and composition of the fluorescence

[0031] The fluorescence intensity F(t) is the measured quantity from which the technical teaching presented in this document derives the magnetic field. In general, it can be expressed as the product of the following factors: with:

[0032] L: Intensity of the green excitation photons generated by a light source; a: Fraction of the light directed from L to the diamond material; rjo: Efficiency of the fluorescence of the diamond material at |B| = 0;

[0033] / ]( |B| ) The effect of |B| on the fluorescence intensity. This is the key effect used in the technical teaching of this document for measuring magnetic fields. This document refers to / ]( |β| ) as the calibration curve; b as the fraction of red fluorescence photons directed to the detector; β as the efficiency of red photon detection.

[0034] Since the overall intensity is a product of all these factors, fluctuations, drift, or general changes in any one factor directly and negatively affect the measurement. This document will discuss the stability of these factors and attenuation techniques in a subsequent section.

[0035] Calibration curve

[0036] The calibration curve / ](|β|) quantifies the influence of a magnetic field |β| on the fluorescence intensity. The calibration curve / ](|β|) can be determined by measuring F(|β|) and normalizing the curve to F(|β|=0T). This yields the normalized fluorescence intensity. which is identical to the calibration curve if L, a, b and ß do not change during the measurement:

[0037] A measured calibration curve is shown in Figure 5. By definition, the curve begins with . un( j ze jg t jann ejne njC| 1t |j neare unc j njC The fluorescence intensity is monotone. This means that it is not possible to measure a specific fluorescence intensity. to assign a unique value of |B| for all magnetic field values.

[0038] 2.1 Estimation of |β| from a measured fluorescence intensity

[0039] Our measured value is the normalized fluorescence intensity ' . Since a magnetic field is to be measured here, the magnetic flux density is estimated from this measured fluorescence intensity according to the technical principles outlined in this document. As every experimental measurement is subject to measurement errors, the method presented here takes this into account. The method presented here assumes that the measured intensity is accompanied by Gaussian noise of width O. This document defines the signal-to-noise ratio as

[0040] The document presented here further assumes that the intensity of the fluorescence radiation from the NV centers is known at a magnetic field of zero Tesla. On the left side of Figure 6a is the Gaussian distribution for three fluorescence intensity measurements ( z= [1.01, 0.9, 0.855]) with an SNR of 1000. The following document estimates the magnetic field based on the fluorescence measurement and the calibration curve. Each Gaussian distribution is projected onto the calibration curve to calculate the probability distribution p( |B| ) of the magnetic field values ​​|B| ZU that correspond to the measured intensity distribution (Fig. 6a, bottom). It can be seen that at a measured intensity of 1.01 (solid lines marked a), the calibration curve has two values ​​(=0 mT and ~4 mT). A magnetic field value |β| cannot be uniquely derived from this measured intensity. At a measured intensity of 0.9, a single peak is visible at ~25 mT, which also looks quite Gaussian, since the calibration function has only a slight curvature in this region and is bijective. At a value of 0.855, a single, broad and asymmetrical peak around 55mT is discernible.The peak is broad because the slope of the calibration curve is small, and it is asymmetric because the calibration curve is curved at this point. These probability distributions p(|B|) answer the question, "If I measured a certain intensity F with a certain SNR, what |B| was present?" In Figure 6b, the document presented here shows the probability distribution p(F), |B| for all possible intensity measurements at an SNR of 100. Since this document assumes that the magnetic field is aligned along an axis and our sensor element only measures |B|, the negative branch of the probability distribution has been added because both field directions yield the same fluorescence values.

[0041] Measurements with a bias field

[0042] The QT-DMFS is an isotropic sensor element at the end of an optical waveguide, and its response depends only on |B|. If a magnetic field with a known direction needs to be measured (e.g., B), the QT-DMFS can be used as a sensor element. x Applying a static bias field can be useful. This document distinguishes between parallel and perpendicular components of the bias field relative to the field of interest. The total magnetic field at the sensor element position is given by with the total magnetic field Bt, the magnetic field B generated by the line current ILTG in the line LTG c and the parallel magnetic field B and the perpendicular magnetic field Bi of the premagnetization fields. Figure 7 shows the response of a QT-DMFS sensor element to magnetic fields in the x-direction with different premagnetization fields. Current measurements via magnetic field measurements. Theory: The Biot-Savart Law

[0043] The Biot-Savart law describes the magnetic field generated in the line LTG by a constant electric current I LTG

[0018] . In the simplest case, the magnetic field B c generated by a conduction current ILTG in an infinitely long thin conductor LG 3 is given by: with the magnetic permeability [Jo, the line current ILTG in the line LTG and the sensor element spacing d

[0019] , This expression also applies if the line LTG has a finite cross-section, as long as the current distribution of the line current ILTG in the line LTG is rotationally symmetric about the wire axis of the line LTG and the measuring position d is not within the line LTG.

[0044] The orientation of ^ is tangential to concentric circles around the wire axis (see Fig. 8a). Figure 8b shows the magnitude of the magnetic field |^c| generated by a conduction current l cin an infinitely long wire for distances d = 1 mm... Im. Both the line current and the distance are plotted on logarithmic scales, and it can be seen that Boc I and Boc(l / d) are proportional. The range between 5 mT (solid green line) and 50 mT (dashed green line) shows the preferred operating range of the QT-DMFS sensor element.

[0045] Absolute intensity measurements using a sensor element array

[0046] Due to the non-monotonic shape of the calibration curve, measuring the line current ILTG in line LTG via magnetic field measurements requires multiple measurement points, which may, for example, have different distances from the wire. Therefore, this document considers an example configuration with three magnetic field sensor elements and examines whether this enables an unambiguous measurement of the currents. This document is based on the following assumptions:

[0047] • Each sensor element is located at a fixed, known distance from the wire axis d, ;

[0048] • The magnetic field at each sensor element consists only of the magnetic field generated by the line current I LTG in the line LTG and the premagnetization field (β|| and / or Bi );

[0049] • Each sensor element follows the same calibration curve as shown in Fig. 5;

[0050] • the intensity of F (B = 0) is known for each sensor element;

[0051] • the factors L, a, q, b and ß are constant.

[0052] Absolute intensity measurements without a sign field

[0053] This document first considers the case where no premagnetization field is present (B = Bi = 0T), so that the magnetic field at the sensor element is generated solely by the conduction current ILTG in the line LTG. To understand how currents can be measured with a QT-DMFS sensor element, this document first considers the magnetic field and the relative fluorescence intensity when a conduction current / c =[-3000 . . 3000]A flows through a conductor LTG. In Figure 9a, the magnetic field | B c The graph shows the current ILTG in the line LTG (linear scale) and the distance d = [l ... 100] mm to the wire axis of the line LTG (logarithmic scale). It is evident that the graph is symmetrical about l. cThe value is 0 because the magnitude of the magnetic field depends only on the magnitude of the conduction current ILTG in the line LTG and not on its direction. Figure 9b of this document shows the relative fluorescence intensity of the fluorescence radiation that the QT-DMFS sensor element would generate for different conduction currents ILTG in the line LTG and at different distances.

[0054] This document considers an example configuration with three sensor elements, preferably three QT-DMFS sensor elements, located at distances of 1 mm, 10 mm, and 100 mm from the wire of the LTG line. Figure 10 (top panel) shows the magnetic field that these sensor elements would measure as part of sensors when the line current is in a range of

[0055] [-3000 ... 3000]A increases. The lower field shows the relative fluorescence intensities of the fluorescence radiation that the QT-DMFS sensor elements would generate. It shows that the sensor element 0 at a distance of 1 mm reaches a magnetic field of 600 mT at the beginning and end of the ramp at ±3000A. Only at currents below 200A (corresponding to | B c Below 50 mT, the sensor element exhibits a dynamic response. At higher currents, the fluorescence signal no longer changes, and the sensor element of the QT-DMFS is saturated. Sensor element 1 shows a maximum magnetic field of 60 mT, which is well within its measuring range. Sensor element 2 experiences only a 6 mT magnetic field at maximum current; therefore, the fluorescence signal shows little contrast. Reconstructing the current signal from the fluorescence measurements of the different sensor elements is a non-trivial challenge. (The reverse problem is also relevant.)

[0056] In the previous section, the document presented here simulated the normalized fluorescence intensity (Fig. 10, bottom) for sensor elements of QT-DMFS at different distances from a current-carrying wire. This was done in two steps: First, the magnetic field around the wire of line LTG was calculated (GL 3), and then the calibration curve was used to convert the magnetic field into a fluorescence intensity F of the fluorescence radiation FL. Now, the problem is reversed: to reconstruct the conduction current in the wire of line LTG from the fluorescence intensities in Fig. 10b. This is also a two-step process: First, the magnetic field at the location of the sensor elements is estimated from the respective normalized fluorescence measurement.Secondly, the technical doctrine presented here assumes that the measurement is subject to error and uses the calibration curve (which is not bijective) together with the probability distribution p(B) (Fig. 6b). This probability distribution for the magnetic field, by compensating for the known bias fields, yields only the magnetic field contribution of the current-carrying wire of the line LTG. The technical doctrine presented here can then calculate p(B). c ) into a probability distribution p( / cThe inversion of the conduction current ILTG is converted by again using GL 3, as described in the technical teaching. For each sensor element and each measurement, the probability distribution p |B| ), where / = 0, 1, 2 indicates the index of the sensor element, is determined according to the technical teaching of this document. This document shows the distribution for the three sensor elements in Figures 11a, 11b, and 12a. Considering sensor element 0, the inversion in the range of ILTG = [-200 . . 200]A yields a good and accurate estimate of the magnetic field. For higher currents, the estimate is poor. This is because the fluorescence hardly changes outside of [-50 . . 50]mT and therefore provides no information about the true value of B in the fluorescence measurement. cis included. Sensor element 1 shows good estimation across the entire current range, and at low currents, additional branches are present in the probability distribution. These correspond to a magnetic field range of 5 mT, in which the calibration curve exhibits two possible magnetic field values ​​for one F-value (due to the dip in the calibration curve). Sensor element 2 operates only in this small magnetic field range, and it is evident that the ambiguity of the calibration curve in this range causes a complex pattern in the probability distribution of the estimated magnetic fields. With the known distances of the sensor elements 20 to the wire of the line LTG, the preferably computer-implemented and / or machine-implemented method presented here can not only |B C | estimate, but also the line current ILTG in the line LTG, the B cThis has generated. Mathematically speaking, this corresponds to a scaling of the y-axis by an estimated B. c to estimated ILTG (right axis (blue text) in Figures 11a, 11b, and 12a). Since these functions PI(ILTG, estimated, / LTG) are probability densities for each sensor element 10, the method presented here can combine them by multiplication. This leads to a combined estimate of ILTG for the entire sensor element array, shown in Figure 12b). It is evident that the ambiguities, e.g., of sensor element 2, are removed from the overall estimate. What remains is the uncertainty about the direction of the conduction current ILTG, because all sensor element signals are the same for ±ILTG. Even combining all sensor elements cannot eliminate this ambiguity. The direction of a conduction current ILTG cannot be determined with an array of QT-DMFS sensor elements without a premagnetization field.

[0057] Measurements of absolute intensity with premagnetization field

[0058] To also measure the direction of the conduction current ILTG, a magnetic bias field can be used. The method presented here assumes a parallel bias field B, and the magnitude of the magnetic field is then given by

[0059] This document considers a sensor element array with three sensor elements located at the same distance from the wire axis, but exhibiting different premagnetization fields: • Sensor element 0: d = 17mm; B / i= +15mT,

[0060] • Sensor element 1: d = 17mm; no premagnetization field,

[0061] • Sensor element 2: d = 17mm; ß / / = -15mT.

[0062] This geometry and the maximum current of the line current ILTG of 3 kA are compatible with power lines of the highest voltage level (750 kV). Figure 13 shows the magnetic field (upper field) and the fluorescence intensity of the sensor elements (lower field) when the line current ILTG in the wire of the line LTG is shifted in the range of [-3000 ... 3000] A for ILTG. Three identical curves can be seen, which are only shifted on the x-axis by ~ +900 A (corresponding to the magnetic field ±15 mT as a premagnetization field). It can be seen that the curves with the premagnetization field are no longer symmetrical about ILTG = OA because the premagnetization field shifts the operating point of the respective sensor element, which is a prerequisite for measuring the current direction of the line current ZU. The method presented here now performs the same inversion procedure as in the previous section.It can be seen that all Pi (IBI) look similar: a cross shape with an additional rhombus-like shape in its center (see Fig. 14). The method proposed here preferably recalculates the estimated line currents ILTG using computer-implemented and / or machine-implemented methods and then multiplies them all together. The combined estimate for the line current ILTG in the wire of the line LTG (see Fig. 15) yields unique values ​​that also correctly resolve the direction of the line current ILTG. Up to this point, the method proposed here has calculated the probability distribution for the combined sensor element signal using a computer-implemented and / or machine-implemented method. The following description now condenses this into an estimated line current. and the standard deviation

[0063] Figure 15b shows the results. It can be seen that the absolute error of the estimated line current ILTG is always < 10A and for currents > 1000A the relative error is < 0.5%.

[0064] Intensity Stability

[0065] In the absolute measurements, the method presented here assumed that the factors L, a, r, b, and 6 from Eq. 1 are constant. In reality, each of these factors is susceptible to changes, e.g., due to drift, aging, temperature, or mechanical motion. The technical doctrine presented here re-examines the factors of Eq. 1 and points out possible causes of undesired changes:

[0066] L (light source):

[0067] • Temperature: can change the intensity and [Severity: high]

[0068] • Aging: Light source loses efficiency over time [Severity: high]

[0069] • Electrical noise:

[0070] Power supplies and circuits cause noise. [Severity: low] (Light emission on diamond material):

[0071] • Mechanical: Movements of the optical components

[0072] (Connectors, fibers) can change a. [Severity: unknown] qo (efficiency of photon conversion at |B| = 0):

[0073] • Demonstration of the temperature dependence in

[0020] :

[0074] Ho = Ho (T). [Severity: low] q( | B | ) (Change in conversion efficiency with | B | ):

[0075] • Demonstration of the temperature dependence in

[0021] :

[0076] F| = q(|B|, T. [Difficulty level: low]

[0077] (Emission of photons to the detector):

[0078] • Mechanical: Movements of the optical components

[0079] (Connectors, fibers) can change b. [Severity: unknown] β (Efficiency of red photon detection):

[0080] • Temperature, electrical noise and aging can alter β [Severity: low],

[0081] Alternative measurement concepts

[0082] Especially when a DC measurement is desired, a fluctuation in a single factor in Eq. 1 jeopardizes the measurement. Therefore, other measurement methods should be considered that inherently provide an intensity reference or are conceptually immune to changes in some of the factors. Other measurement methods that can mitigate temporal variations in one or more of the factors are discussed below.

[0083] Sensor element arrays with relative intensity measurements

[0084] The intensity of the light source L (the pump radiation source 100) can change over time, e.g., due to aging effects, noise, or temperature changes. Therefore, it must generally be assumed that L = L(t). The technical teaching presented here assumes that all sensor elements are illuminated by the same light source (pump radiation source 100), e.g., by an LED, and that the light intensity for each sensor element can be calculated by multiplying a fixed factor by the light source intensity. Variations in the light source intensity can then be eliminated by considering not the intensity of each individual sensor element, but rather the ratio of the fluorescence intensity of any two sensor elements. In Figures 16 a)-c), the technical teaching of this document has used the ratios instead of the intensities of the three sensor elements from Figure 13.All other parameters (such as the bias fields and the SNR) are the same. Figure 17 shows the result of combining all three ratios. Due to the premagnetizing magnets, it is also possible to measure the amplitude and direction of the conduction current ILTG ZU. For this calculation, the technical teaching presented here does not simulate time-dependent intensity noise, as it uses a fixed SNR. Therefore, the relative error is higher than with pure intensity measurement. However, since this approach inherently eliminates all intensity fluctuations, the noise would be lower in a real-world application.

[0085] Non-intensity observable

[0086] Another possibility would be to use an observation quantity that does not depend on the intensity to measure the change in the magnetic field B generated by the conduction current ILTG. cto measure. One such non-intensity-related measurement quantity is, for example, the lifetime of the NV centers, which also depends on the magnetic field

[0022] . Since the lifetime of the NV centers is on the order of 10 ns, different (faster) electronics are required. This concept was tested and it was shown that this measurement scheme is insensitive to intensity fluctuations

[0023] . These include not only intensity fluctuations of the light source L, but also fluctuations of a, b and β.

[0087] Permanent magnets with premagnetization

[0088] For a current sensor capable of resolving the direction of line current (ILTG) based on QT-DMFS technology, applying a premagnetic field to at least one sensor element is required. This premagnetic field is generated by a permanent magnet. This document examines various types of commercially available permanent magnets and assesses their suitability for this application. The technical teaching presented in this document specifies the following requirements for the permanent magnet: it should withstand temperatures up to 100°C, it should be temperature-stable (minimal change in magnetization), it should be chemically stable, and it should generate a magnetic field of approximately 20 mT at a distance of 1 mm to 1 cm without significant fluctuations.

[0089] The surface area should not demagnetize in the expected magnetic field environment.

[0090] The document presented here examines the four permanent magnet materials:

[0091] 1. Neodymium-iron (NdFeB) bore

[0024] ,

[0092] 2. samarium cobalt (Sm2Col7)

[0025] ,

[0093] 3. anisotropic ferrite (ferrites)

[0026] and

[0094] 4. Iron-aluminium-nickel-cobalt (AINiCo)

[0027] ,

[0095] The technical teaching presented here now considers the temperature properties specified in the manufacturer's data for permanent magnets. Furthermore, for this purpose, the technical teaching presented here uses an exemplary disc magnet (diameter: 3 mm; thickness: 1 mm) that can be easily installed in a sensor head with a sensor element in order to test its electrical, magnetic, and mechanical properties. In developing the technical teaching of this document, FEM simulations

[0028] were performed to determine the following values:

[0096] • the magnetic field B generated by the line current ILTG c ,

[0097] • the power loss when the magnet is located near a wire of the LTG line that is carrying alternating current, and

[0098] • the magnetic force between the magnet and the wire of the LTG line.

[0099] Magnetic field strength

[0100] The technical teaching presented in this document first considers the magnetic field generated by magnets made of different materials and simulates the magnetic field of a magnetic mesh disk in a vacuum with axial magnetization. Using the material properties from the data sheets for the four magnetic materials, simulations were performed for a magnetic disk made of each of these materials. Figure 18 shows the decay of the magnetic field along the axis of rotation from the surface (z = 0). The NdFeB and Sm₂Col₇ magnets have a stronger magnetic field, which drops to 15 mT at z ~ 3.5 mm, while the fields of ferrite and AINiCo reach this value after z ~ 2.5 mm. All values ​​are listed in Table 1.

[0101] Forces on the magnets

[0102] In developing the technical teaching presented in this document, the magnetic force acting on the permanent magnet was simulated when it is located near a wire carrying a current ILTG. For this purpose, the technical teaching of this document assumed a current ILTG = 3000 A and positioned the disc magnet 17 mm from the wire axis. The magnet is oriented such that the magnetic field generated by the wire runs parallel to the magnetization. The forces acting on all magnets are less than 0.05 N. All values ​​are listed in Table 1.

[0103] Eddy currents in the magnet

[0104] Since most permanent magnets are also conductive, an oscillating magnetic field caused by an alternating current in a power line induces eddy currents in the magnet, which heat it up and could thus impair its magnetic performance. In developing the technical teaching of this document, the same geometry as in the previous case was simulated, but a 50 Hz alternating current was introduced into the wire (peak current: 3000 A), and a frequency-domain simulation was performed. From the induced currents and the resistivity of the material, the power loss in the magnet was calculated. For all magnets, the power loss is less than 0.5 pW. All values ​​are listed in Table 1.

[0105] Overview and Conclusion

[0106] Table 1 provides an overview of the relevant parameters taken from the datasheets, as well as the results of our FEM simulations. This document does not yet address the question of whether irreversible demagnetization of the magnets can occur, as this assessment depends on the precise magnet geometry, the temperature, and knowledge of the maximum external magnetic field. Due to its low temperature drift, good temperature resistance, chemical stability, and high magnetization, the technical principles outlined in this document recommend Sm₂Col₇ as the material for the permanent magnets used to generate the premagnetization field.

[0107] Table 1: Properties of various permanent magnet materials. Magnetic properties from the data sheets: B r ,nom : nominal remanent magnetic flux density; a B rreversible temperature coefficient; jU r Permeability; Q ei : specific electrical resistance; T max : Maximum operating temperature. Properties calculated using FEM simulations for a disc magnet (diameter: 3 mm; thickness: 1 mm): c / ismT : Distance from the magnet surface where the field has decayed to 15 mT; Force on the magnet at a distance of 17 mm from a mains current (DC, / = 3 kA); Pdiss '■ Power loss due to ohmic losses of eddy currents generated by a mains current (AC, 50 Hz, Ipeak = 3 kA).

[0108] Correlation of the output signals of the photodetectors with the AC control signal of the evaluation and control device (CTR)

[0109] The present invention relates, as already described, to a device for measuring a line current (ILTG) in an electrically conductive line (LTG). The measurement is optical and utilizes the interaction of magnetic fields with paramagnetic centers in at least three sensor elements (20), which are excited by pump radiation (60) from a pump radiation source (100). The respective paramagnetic centers of the respective sensor elements (20) each emit fluorescence radiation (70), which varies depending on the respective magnetic flux density to which the respective paramagnetic centers are exposed at the location of the respective sensor element (20). This fluorescence radiation (70) is directed by an optical system onto at least three respective photodetectors (150) and detected.The respective output signals of the photodetectors (150) are processed by lock-in amplifiers (210) or, alternatively, by an evaluation and control device (CTR), which is explicitly designed to perform the functions of the lock-in amplifiers (210). The following section describes various configurations of the functionally equivalent processing by the evaluation and control device (CTR). The central function, which is implemented in a functionally equivalent way using machines and / or computers, is to correlate the output signals of the photodetectors (150) with the AC drive signal generated by the evaluation and control device (CTR) in order to determine a respective measured value. This measured value reflects the proportion of the AC drive signal in the respective detector signal and forms the basis for determining an estimated value of the line current (ILTG).According to the invention, functional equivalence is achieved if the evaluation and control device (CTR) can estimate the proportion of the AC drive signal in the respective output signal of the respective photodetector. A lock-in amplifier is therefore only a highly preferred, but still very specialized, correlator device for this purpose.

[0110] 1. Basic operating principle of correlation

[0111] The correlation of the output signals with the AC drive signal is preferably based on the targeted extraction of that signal component whose temporal behavior (e.g., in frequency, phase, and / or amplitude) corresponds to that of the known AC drive signal. The AC drive signal can preferably be a monofrequency sinusoidal signal with essentially only one frequency, typically around 22 MHz (-4 MHz / +7 MHz), and / or only one frequency within an exemplary frequency range of 18 MHz to 29 MHz. Lower frequencies down to near Hz are useful if the phase is not evaluated. The correlation serves to separate the desired signal (fluorescence modulation in sync with the AC drive signal) from noise and any interfering signals.

[0112] The evaluation is performed either by dedicated lock-in amplifiers (210) or – preferably – by computer- and / or machine-implemented methods within the CTR. The choice of correlation technique depends on the requirements for precision, processing speed, flexibility, and integration depth.

[0113] 2. Classic lock-in amplifier technique

[0114] An analog lock-in amplifier (210) operates according to a method related to the heterodyne principle. It multiplies the incoming signal of the respective photodetector (150) by a reference copy of the AC drive signal. This multiplication generates a signal that, in addition to sum and difference frequencies, contains in particular a DC component proportional to the amplitude of the correlated component in the detector signal. A downstream low-pass filter suppresses the high-frequency components, so that only the DC component—the measured value—remains. This corresponds to the formation of a vector scalar product by means of an L²-M multiplication of two functions between the AC drive signal and the respective output signal of the respective photodetector. However, this is only one possibility for correlation.

[0115] This technique is particularly effective with sinusoidal reference signals and, with suitable filter design, delivers a very good signal-to-noise ratio (SNR). However, it is hardware-intensive and less flexible with regard to changes in the drive frequency or signal shape.

[0116] 3. Digital lock-in technology in the CTR

[0117] The evaluation and control device (CTR) can digitally replicate the functionality of a lock-in amplifier. For this purpose, the signals from the photodetectors (150) are first digitized and then evaluated by digital signal processing. Correlation is achieved by digital multiplication with a digital version of the AC drive signal, followed by digital filtering, e.g., with a finite impulse response (FIR) or infinite impulse response (IIR) filter.

[0118] The advantages of this solution lie in its ability to process multiple channels, adapt to different frequency ranges, and offer reproducible and diagnostic results. The CTR can therefore evaluate several channels simultaneously, independently, and flexibly.

[0119] 4. Software-based lock-in methods

[0120] Even more flexible is the purely software-based implementation on general computer architectures, such as in embedded systems or control computers. Here, correlation is performed by algorithms that can utilize, for example, discrete Fourier transforms (DFTs), fast Fourier transforms (FFTs), or Gabor analysis. A common computer and / or machine implementation is the computer- and / or machine-implemented calculation of the quadrature-modulated signal component by two orthogonal multiplications with the sine and cosine components of the reference signal—in this case, the AC drive signal.

[0121] The computer- and / or machine-implemented, for example software-based, solution also enables the computer- and / or machine implementation of adaptive filters and real-time analyses, for example to compensate for system drift or frequency changes.

[0122] 5. Mathematical methods for correlation

[0123] In addition to classical lock-in techniques, several mathematical correlation methods exist that are particularly suitable for computer- and / or machine-implemented realizations. These methods enable a precise, flexible, and automatable computer- and / or machine-implemented analysis of the respective signal components of the respective output signals of the respective photodetectors with respect to the AC drive signal. They utilize different mathematical concepts for the computer- and / or machine-implemented determination of the similarity or overlap of two signals—in this case, the respective output signal of a respective photodetector and the AC drive signal. Five suitable computer- and / or machine-implementable methods are described below:

[0124] 5.1 Cross-correlation

[0125] Computer- and / or machine-implemented cross-correlation is a standard method for determining the similarity between two signals as a function of a time shift (lag). It is defined as the integral or sum (in the discrete case) of the product of the two signals at different relative shifts. In computer- and / or machine-implemented versions, cross-correlation is often implemented using FFT-based methods to reduce the computational load and / or the number of gates. The maximum correlation provides information about the relative phase, while the magnitude of the correlation describes the strength of the correlated component. When applied to the output signal of a specific photodetector, it can precisely extract the component whose waveform corresponds to that of the AC drive signal, even at low signal strength or high noise levels.

[0126] 5.2 Autocorrelation comparison with reference modulation

[0127] A special case is the computer- and / or machine-implemented comparison of the computer- and / or machine-implemented autocorrelation of the respective output signal of the respective photodetector with that of a known modulation pattern (the AC drive signal). Here, a computer- and / or machine-implemented method is used to investigate whether and to what extent, and if so, with what phase shift, the modulation pattern is present in the respective output signal, without necessarily requiring exact synchronization. This allows for robust detection even with a disturbed or distorted signal waveform. This comparison can be performed using a computer- and / or machine-based approach with normalization of the respective autocorrelation functions to compensate for systematic distortions or signal delays.

[0128] 5.3 Modulation-demodulation methods with synchronous detection

[0129] This method is, in a sense, a software-based implementation of the lock-in principle, but it includes adaptive elements. Here, the detector signal is compared to a parametric modulation template (e.g., sinusoidal with variable offset, frequency drift, amplitude modulation). An optimization algorithm (e.g., least squares, Kalman filter) determines the parameters at which the demodulation yields the highest DC component.

[0130] This method is particularly effective under non-stationary conditions or slight frequency variations in the AC drive signal.

[0131] 5.4 Wavelet Correlation

[0132] Computer- and / or machine-implemented wavelet transformations allow for time-frequency-resolved signal analysis. By selecting suitable parent wavelets (e.g., Morlet or Gabor wavelets), signal components with similar time-frequency behavior to the AC drive signal can be extracted from the respective output signals of the individual photodetectors. The computer- and / or machine-implemented wavelet correlation calculates the respective overlap of each photodetector's output signal with a wavelet that corresponds to the AC drive signal in frequency and shape. This method is particularly suitable for analyzing short-duration signal segments or non-stationary signal components in the output signal. 5.5 Correlation methods based on synchronized Fourier coefficients

[0133] Another computer- and / or machine-implemented method involves calculating the respective Fourier coefficients of the output signal of each photodetector and comparing them with the Fourier representation of the AC drive signal. This method is particularly suitable when the AC drive signal is not exactly sinusoidal, but rather modulated, for example, in a trapezoidal or rectangular waveform. In such cases, the AC signal contains several harmonics, the respective amplitudes and phases of which can also be analyzed in the detector signal.

[0134] By implementing it on a DSP or in a software-based environment, the signal can typically be checked in real time for its harmonic matches with the control signal.

[0135] 6. Filter methods for signal processing

[0136] To determine the portion of the output signal of each photodetector (150) whose waveform corresponds to the AC drive signal of the evaluation and control device (CTR), various filters can be used. These filters serve to selectively extract, as the respective filter input signal, the portions of the output signal of each photodetector (150) that match the known AC drive signal, while suppressing other portions—especially noise and interference components. Both classic analog and modern digital and / or software-implemented and / or machine-implemented filter methods or sub-methods are employed as technical means.

[0137] The following describes suitable computer- and / or machine-implementable filters that either use the AC drive signal directly as an additional input signal or require its waveform. The selection and configuration of these computer- and / or machine-implementable filters can be performed within the evaluation and control device (CTR) as part of a computer- and / or machine-implemented evaluation procedure.

[0138] 6.1 Matched Filter

[0139] The computer- and / or machine-implemented matched filter is an optimal filter for maximizing the signal-to-noise ratio (SNR) for known signal waveforms in the noise background. It is precisely tuned to the shape of the desired signal—in this case, the AC drive signal. Mathematically, the matched filter corresponds to a convolution of the input signal with the time-inverted and conjugate version of the expected signal. The filter output reaches its maximum when the expected signal is contained within the input signal, with this maximum being proportional to the signal strength. When applied to the detector signal, the matched filter leads to the precise detection of the component that exhibits the structure of the AC drive signal.

[0140] The implementation is typically digital, based on the known pattern function. The control and evaluation unit (CTR) stores the AC control signal or generates it synchronously with the acquisition of the respective output signal of the photodetector.

[0141] 6.2 Correlation Filter

[0142] A computer- and / or machine-implemented correlation filter is based on the principle of cross-correlation, with the filter designed to generate maximum output while maintaining the highest possible match with the AC drive signal. While the Matched Filter is specifically optimized for additive white noise environments, extended versions of the correlation filter can also be adapted for other noise models. In the drive and evaluation unit (CTR), the respective output signal of each photodetector is correlated with a stored or continuously generated reference function using computer- and / or machine-implemented methods. The resulting correlation function provides information about the strength and phase of the relevant signal component. The filter can optionally also consider the harmonic components of the AC signal to evaluate more complex modulations.

[0143] 6.3 Adaptive Least Mean Squares Filter (LMS Filter)

[0144] The computer- and / or machine-implemented LMS filter is an adaptive filter capable of analyzing the input signal in real time and continuously adjusting the filter coefficients so that the output signal closely matches the AC drive signal. It minimizes the difference between the actual filter output and a desired reference signal, in this case, the AC drive signal.

[0145] In its specific application to the respective output signal of the respective photodetector, the computer- and / or machine-implemented LMS filter enables dynamic tracking even with slight frequency drift or system changes. The control and evaluation unit (CTR) uses the known AC drive signal as a reference and preferably continuously adjusts the internal filter parameters.

[0146] 6.4 Wiener filter

[0147] The computer- and / or machine-implemented Wiener filter is an optimal linear filter in terms of mean squared error reduction between an estimated signal and the actual signal. Unlike the matched filter, the Wiener filter additionally considers the spectral distribution of signal and noise. It requires knowledge or estimation of the signal and noise power spectral density spectra.

[0148] For application to the detector signal, it is assumed that the relevant signal component lies in the spectral vicinity of the AC drive signal. The Wiener filter suppresses frequency components with a high noise content relative to the expected signal strength, thereby providing an optimized estimation signal.

[0149] 6.5 Gabor filter

[0150] A Gabor filter is a special band-limited filter with a Gaussian envelope and a sinusoidal carrier function. It allows simultaneous time and frequency resolution and is therefore particularly suitable for detecting sinusoidal signal components in noisy signals.

[0151] In the control and evaluation unit (CTR), a Gabor filter can be parametrically adjusted so that its carrier frequency corresponds to the frequency of the AC drive signal. Convolution with the detector signal yields a measure of the locally time-varying signal strength relative to the AC drive signal.

[0152] 6.6 Kalman filter

[0153] The Kalman filter is a recursive estimation method developed for state-space models. It is particularly suitable for estimating time-varying signals while taking into account model uncertainties and measurement noise. The filter consists of a prediction and an update phase, in which the best estimates for the system's state are calculated. For use in the present device, the computer- and / or machine-implemented Kalman filter can be used to model and continuously estimate the respective time evolution of the correlated component of each photodetector's output signal with the AC drive signal. The model includes a known reference signal waveform (e.g., a 22 MHz sine wave) and a measurement model that describes the noisy photodetector output signal.

[0154] The control and evaluation device (CTR) can, for example, implement the Kalman filter in software and, if necessary, also use nonlinear variants (Extended Kalman Filter, Unscented Kalman Filter) when nonlinear dependencies need to be taken into account (e.g., through spectral shifts or harmonic modulation).

[0155] 6.7 Synchronization filter

[0156] A computer- and / or machine-implemented synchronization filter can be used to extract a phase-correct reference signal, in this case the AC drive signal, from the respective output signal of the respective photodetector, or to bring it into exact phase alignment with the AC drive signal. In typical implementations, synchronization is achieved by computer- and / or machine-implemented phase detectors or phase-locked loops (PLLs), which are implemented either analogously or digitally.

[0157] In the control and evaluation unit (CTR), a digital PLL can be used to precisely synchronize an internal reference signal with the AC drive signal. This is particularly useful when the external AC drive signal is not available in real time or exhibits frequency modulation. In such a case, a reconstructed "shadow" reference channel is generated, which can then be used to correlate the detector signal.

[0158] 6.8 Adaptive bandpass filters with reference adjustment

[0159] Each computer- and / or machine-implemented adaptive bandpass filter is a dynamically adjustable filter whose center frequency and bandwidth are optimized for the best possible extraction of a specific frequency component from the output signal of the respective photodetector. If the AC drive signal is not perfectly stable (e.g., due to temperature drift of the clock generator), the filter can be adjusted in real time. The control and evaluation unit (CTR) can monitor the current frequency of the AC drive signal and set corresponding parameters for the respective digital HR or FIR bandpass filters. These filters selectively amplify the specific component of the respective photodetector's output signal within a narrow frequency window around the AC drive signal and suppress components outside this window.

[0160] 6.9 Resonance filters (resonator-based detection)

[0161] A computer- and / or machine-implemented resonant filter represents another way to selectively respond to the frequency of the AC drive signal. It can be implemented as a computer- and / or machine-implemented digital mechanical model (e.g., a resonant system) sensitive to a precisely defined frequency. In the computer- and / or machine-implemented digital implementation, the control and evaluation device (CTR) can represent such a model as part of a computer- and / or machine-implemented resonant system of equations (e.g., a damped harmonic oscillator), which is excited by the detector signal. The system's response to the drive signal is maximal when the frequency matches, resulting in particularly robust frequency filtering.

[0162] 6.10 Cascaded Filter Structures

[0163] In more complex computer- and / or machine-implemented analyses, the control and evaluation device (CTR) can combine several of the aforementioned filter types in a computer- and / or machine-implemented manner. For example, a matched filter can be used for coarse detection, followed by a Wiener filter for noise suppression and a Kalman filter for smoothing the time-domain response estimation. This cascaded combination achieves a robust, multi-stage analysis that delivers reliable measurement results even under non-ideal signal conditions. The CTR is capable of automatically configuring such filter chains and adaptively responding to changing system parameters (e.g., temperature, noise behavior, modulation frequency).

[0164] 7. Software implementation of the correlation and filtering methods in the evaluation and control device (CTR)

[0165] The evaluation and control device (CTR) is preferably configured to replace or implement the functions of the lock-in amplifiers (210) not only by electronic hardware modules, but also, in particular, by software-based, computer-implemented methods, either wholly or partially. These computer- and / or machine-implemented methods enable flexible, precise, and scalable evaluation of the respective output signals of the respective photodetectors (150).

[0166] 7.1 Exemplary architecture of the signal processing unit of the evaluation and control device (CTR)

[0167] The evaluation and control device (CTR) can, for example, include a digital signal processing unit (DSP unit) based on a microcontroller, a digital signal processor (DSP), a field-programmable gate array (FPGA), an integrated computing system (e.g., a SoC with a CPU / GPU combination), or the like. This unit can be responsible for one or more of the following:

[0168] • the digitization of the analog detector signals,

[0169] • the management of synchronous clocking for the generation and referencing of the AC control signal,

[0170] • the execution of computer- and / or machine-implemented correlation analyses,

[0171] • the application of the described computer- and / or machine-implemented filtering methods,

[0172] • the continuous computer- and / or machine-implemented evaluation and aggregation of the respective signal components thus determined,

[0173] • The computer- and / or machine-implemented calculation of an estimated value for the magnitude of the line current (ILTG) using one or more respective measured values ​​of the determined measured values ​​of the respective output signals of the respective photodetectors involved. The architecture can be modular, so that individual components – such as correlation modules, filter stages, or current estimation methods – can be updated, replaced, or adapted independently.

[0174] 7.2 Software modules for digital correlation

[0175] Each signal processing unit within the control and evaluation device (CTR) preferably performs a defined processing sequence for each output signal of the respective, assigned photodetector. This includes:

[0176] 1. Pre-processing of the respective output signal of the respective photodetector: attenuation, smoothing, equalization (optional),

[0177] 2. Respective reference signal generation: either locally by a clock generator or by receiving a synchronized reference signal,

[0178] 3. respective digital correlation: respective application of one or more computer- and / or machine-implemented correlation methods according to Section 5 (e.g. cross-correlation, Fourier analysis, adaptive demodulation),

[0179] 4. Respective extraction of the respective correlated signal component: respective determination of amplitude, phase or integral measurement value as a measure of the agreement with the AC control signal,

[0180] 5. respective storage, logging, forwarding to the estimation logic.

[0181] Where applicable, the software in question can be designed to evaluate multiple channels in parallel. This enables the simultaneous computer- and / or machine-implemented correlation of at least three (preferably more) detector signals in real time. The software code in question may reside, at least temporarily, in a storage medium and / or in the memory of the control and / or evaluation device (CTR).

[0182] 7.3 Software modules for digital filtering

[0183] The filtering methods according to Section 6 “Filtering Methods” are preferably implemented as discrete, deterministic modules in the evaluation and control device (CTR) on a computer and / or machine basis. Each respective filter unit, filter module, or filter function can preferably be configured to suit the specific application, for example, with regard to:

[0184] • Filter type (e.g. FIR, 11 R, adaptive),

[0185] • Center frequency (e.g. 22 MHz),

[0186] • Bandwidth,

[0187] • Phase angle (e.g. 0° for in-phase channel, 90° for quadrature channel),

[0188] • Time constants (e.g., in Kalman filters or adaptive methods). One or more configuration modules preferably monitor the stability and precision of the respective filter outputs in real time and can preferably perform adaptive switching or fine-tuning, for example, when the operating frequency (frequency of the AC drive signal) changes or anomalies are detected.

[0189] 7.4 Error compensation and disturbance robustness

[0190] The evaluation and control device (CTR) can also be equipped with software modules designed for the detection and compensation of interference. These include, for example:

[0191] • Automatic computer- and / or machine-implemented suppression of extraneous signals (e.g., through frequency tracking),

[0192] • Computer- and / or machine-implemented detection of outliers and their computer- and / or machine-implemented masking,

[0193] • Computer- and / or machine-implemented filter reconfiguration in case of shifted signal conditions (e.g. due to temperature changes),

[0194] • Computer- and / or machine-implemented adaptive threshold adjustment to avoid misinterpretations at low signal strengths.

[0195] The implemented computer- and / or machine-implemented algorithms and procedures are preferably based partly on machine learning methods and / or computer- and / or machine-implemented methods of artificial intelligence and / or heuristic computer- and / or machine-implemented rule sets, which are, for example, hard-wired in the control and evaluation device (CTR) and / or run on a learning-capable computer- and / or machine-implemented platform.

[0196] 7.5 Real-time capability and synchronization

[0197] All signal processing steps can be designed for real-time applications. The control and evaluation unit (CTR) can synchronize the sampling of the respective output signals of the respective photodetectors with the AC control signal. Preferably (but not necessarily), the control and evaluation unit (CTR) maintains a maximum delay in the microsecond range. In cases of particularly high dynamics (e.g., in industrial or automotive applications with rapidly changing and especially pulsed currents), the control and evaluation unit (CTR) can optionally utilize optimized parallel computer- and / or machine-implemented processing structures (pipelining, multithreading). The output of the correlated measured values ​​is either continuous or event-driven and is available to other system components (e.g., a control module, a visualization unit, or a fault detection system).

[0198] 8. Overall Integration of Computer- and / or Machine-Implemented Correlation and Filtering into the Computer- and / or Machine-Implemented Current Measurement Function The computer-implementable and / or machine-implementable correlation techniques and filtering methods described in some previous sections are typically an integral part of the overall concept of the device presented here for measuring a line current (ILTG). The device uses several optical sensor elements (20) with paramagnetic centers, which, due to different bias magnetic fields, each react differently to a superimposed modulated magnetic interaction and, from this, allow conclusions to be drawn about the respective local magnetic flux density via a respective fluorescence intensity modulation.These respective fluorescence modulations are each synchronous with the preferably common AC control signal generated by the evaluation and control device (CTR).

[0199] 8.1 Parallel computer- and / or machine-implemented evaluation of multiple sensor channels

[0200] The control and evaluation device (CTR) can be configured to acquire and evaluate the respective output signals of at least three photodetectors (150) in real time. Preferably, each photodetector can be assigned a separate receiving system with an associated sensor element (20). The respective fluorescence radiation (70) can then be selectively directed to each photodetector, generating an output signal that typically comprises a signal superposition of the respective signal component of interest (with AC modulation) as well as respective noise and interference components.The control and evaluation device (CTR) preferably processes each of these respective output signals of the respective photodetectors separately and preferably performs a respective correlation with the AC control signal for each of these respective output signals. This computer- and / or machine-implemented processing preferably includes:

[0201] • the respective digital acquisition of the respective output signal of the respective photodetector (e.g. via a respective measurement channel of a multi-channel analog-to-digital converter ADC),

[0202] • the respective computer- and / or machine-implemented digital or software-based correlation, e.g., according to the preceding Section 5,

[0203] • the filtering according to section 6 to amplify and isolate the relevant component, • the extraction of a normalized measurement value that describes the amplitude and / or phase of the fluorescence modulation.

[0204] 8.2 Computer- and / or machine-implemented calculation of the estimated value for line current (ILTG)

[0205] The measured values ​​acquired from at least three channels are combined by the control and evaluation device (CTR) in a central computer- and / or machine-implemented evaluation step to obtain an estimated value of the conduction current (ILTG). This is achieved using a computer- and / or machine-implemented mathematical model that relates the spatial distribution of the magnetic flux densities of the magnetic fields at the respective sensor elements to the current (ILTG) in the conductor (LTG). Typically, an inverse model is used that derives the causative current value of the conduction current (ILTG) from the detected local fluorescence intensity values ​​(represented by the correlated signal components).

[0206] This calculation can be performed analytically (e.g., with symmetrical sensor arrangement) or numerically (e.g., by matrix solution or iterative approximation) using computer- and / or machine-implemented methods and / or algorithms. The use of at least three sensor elements allows for an unambiguous determination of the conduction current (ILTG) value even with nonlinear field distributions or asymmetrical sensor element arrangements.

[0207] 8.3 Advantages of computer- and / or machine-implemented software-based implementation

[0208] The computer- and / or machine-implemented, software-based implementation of the respective correlation and filter functions within the control and evaluation device (CTR) offers several crucial advantages over classic hardware solutions:

[0209] • Flexibility: Changes in the AC control signal (e.g., frequency, modulation shape) can be accommodated through software adjustments without physical modifications to the hardware.

[0210] • Multi-channel capability: Parallel processing of many sensor channels is easily scalable thanks to digital architectures.

[0211] • Adaptivity: The control and evaluation device (CTR) can adapt to changing environmental conditions (e.g., temperature fluctuations, system drift) through computer- and / or machine-implemented algorithms.

[0212] • Diagnostic and calibratable: All computer- and / or machine-implemented processing steps are transparent and can be recorded, analyzed, and verified. • Robustness: The combination of various computer- and / or machine-implemented filtering methods allows for highly precise and low-noise determination of the signal component, even under adverse interference conditions.

[0213] • Real-time processing: Efficient digital computer and / or machine-implemented processing allows all steps to be performed with minimal latency, enabling a fast response time to power changes.

[0214] 8.4 Summary of the section on overall integration

[0215] The present technical teaching integrates modern computer- and / or machine-implemented signal processing methods into the proposed device for the non-contact, galvanically and thermally isolated, and high-precision measurement of electrical conduction currents (ILTG). The central role of the evaluation and control device (CTR) lies in the correct and robust determination of the signal component in the output signals of the photodetectors that corresponds to the known AC drive signal. For this purpose, multiple computer- and / or machine-implemented correlation and filtering methods are used, which are specifically tailored to the signal characteristics and, in their entirety, ensure reliable current measurement.The combination of optical detection, magnetic field analysis, mathematical modeling and digital signal processing creates an innovative system that can be used in both industrial applications and sensitive scientific experiments.

[0216] Summary of the report

[0217] The preceding description investigated which currents, and thus which magnetic fields, are generated by high-voltage power lines and whether these can be detected by magnetic field sensors based on the red fluorescence of a disordered multitude of diamond crystals when irradiated with green pump radiation, and whether they can be used to measure the currents. The preceding description explained the operating principle of diamond-based magnetic field sensors and showed how the sensor element signal correlates with the magnetic fields.

[0218] The technical teaching presented here proposes a current sensor design with at least three measuring points, and the technical teaching of this document investigates the measurement possibilities for two cases: no premagnetization fields and different premagnetization fields for each measuring point. For a measuring range of ±3 kA, the technical teaching presented here estimated the relative error to be less than 0.5% for currents above 1 kA. Additionally, this document presents alternative measurement methods that are inherently immune to intensity fluctuations of the light source. The light source is the pump radiation source used to irradiate the NV centers with green pump radiation.

[0219] To confirm the feasibility of the proposed premagnetization fields for various measuring points, FEM simulations were performed for different magnetic materials. Samarium-cobalt proved to be the most suitable material. The technical teaching presented here also calculated the potential forces on the premagnetization magnets due to the magnetic fields B generated by the conduction current ILTG. c and the power loss due to eddy currents when alternating currents are present. The technical teaching presented here concluded that both effects are only minor and should not impair the measurement performance.

[0220] Thus, NV powder-based sensor elements 20 can be used for current measurements of the line current I LTG. Based on the results of the above description, this document proposes a technical implementation for a sensor based on the following application scenario:

[0221] • The current sensor is intended for measuring electrical line currents ILTG in lines LTG of high-voltage power lines, since the effort required for insulation is most costly in these applications and this application benefits most from a purely optical measurement.

[0222] • The current sensor is intended to be a retrofit or clamp solution for high-voltage cables.

[0223] • The current sensor should be able to measure alternating (~50Hz) and direct currents.

[0224] • The current sensor should be able to measure the amplitude and direction of the line current I LTG. The current sensor described below fulfills these requirements.

[0225] Advantage

[0226] In summary, NV-center-based magnetic field sensors can be used as purely optical current sensors for high-voltage applications (TRL 2). They offer good electrical isolation and are potentially more robust than current optical measurement concepts. Nevertheless, technological challenges remain, which are addressed in the following study.

[0227] The figures illustrate the proposal schematically and in a simplified manner. The disclosure of the text presented here is not limited to the figures and also includes other combinations. Figure 1

[0228] Figure 1 shows the shares of renewable energies in the electricity sector (CUR), the heat sector (HT) and the transport sector (TR) in the annual electricity generated in Germany in 2023 (Source: [State of the art: 6])

[0229] Figure 2

[0230] Figure 2a

[0231] Figure 2a shows the cross-section of an overhead line cable (diameter 20 mm). Aluminum sheath and steel core (left), aluminum cable sheath and fiber-reinforced composite core (right). (Source: [State of the art: 11])

[0232] Figure 2b

[0233] Figure 2b shows geometric parameters and performance parameters of high voltage lines of different network levels: wire diameter (bottom), maximum effective currents (middle) and expected local magnetic field near the wire (top).

[0234] Figure 3

[0235] Figure 3a

[0236] Figure 3a shows a state-of-the-art current transformer [state of the art: 14]

[0237] Figure 3b

[0238] Figure 3b (b) Schematic representation of an optical current sensor based on the Faraday effect (Source: [State of the art: 16]).

[0239] Figure 4

[0240] Figure 4 shows a commercially available diamond magnetic field sensor QT-DMFS-C2 from Quantum Technologies GmbH based on diamond powder with NV centers and an optical waveguide for pump radiation and fluorescence radiation.

[0241] Figure 5

[0242] Figure 5 shows a calibration curve of a commercially available diamond magnetic field sensor QT-DMFS-C2 from Quantum Technologies GmbH based on diamond powder with NV centers and an optical waveguide for pump radiation and fluorescence radiation.

[0243] Figure 6

[0244] Figure 6 shows estimates of the magnetic field for a given fluorescence intensity measurement, with a signal-to-noise ratio (SNR) of 100. Figure 6a

[0245] Figure 6a shows a calibration curve (center) with normalized fluorescence measurements (left field, F = 1.01 (blue), F = 0.9 (orange), F = 0.855 (green)) and corresponding probability distributions for | B | (bottom image).

[0246] Figure 6b

[0247] Figure 6b shows the overall probability distribution of p(F, Bc)

[0248] Figure 7

[0249] Figure 7 shows the response of a commercially available diamond magnetic field sensor QT-DMFS-C2 from Quantum Technologies GmbH based on diamond powder with NV centers and an optical waveguide for pump radiation and fluorescence radiation to a magnetic field B. x without a premagnetization field (black); with a premagnetization field of 15 mT parallel to B x (green); and perpendicular to B x (15mT (red); 3mT (orange)).

[0250] Figure 8

[0251] Figure 8a

[0252] Figure 8a illustrates a magnetic field generated by a current in a wire [Prior Art: 18],

[0253] Figure 8b

[0254] Figure 8b illustrates the magnetic field strength B c , which is generated by a current in a straight infinite wire at different distances from the wire axis, in a log-logarithmic representation.

[0255] Figure 9

[0256] Figure 9a

[0257] Figure 9a shows the strength of the magnetic flux density B c of the magnetic field generated by a current in an infinitely long wire. R

[0258] Figure 9b

[0259] Figure 9b shows the corresponding relative fluorescence intensity of the commercially available QT-DMFS-C2 optical diamond magnetic field sensor from Quantum Technologies GmbH, based on diamond powder with NV centers and an optical waveguide for pump radiation and fluorescence radiation, as a function of current and distance to the wire. The x-axis (current) is linear and the y-axis (distance) is logarithmic. Figure 10

[0260] Figure 10a

[0261] Figure 10a shows the magnetic field for three sensor elements at different distances compared to the line current in the line I LTG = [-3000 . . . 3000]A. Figure 10a corresponds to the horizontal section through the diagram of Figure 9a.

[0262] Figure 10b

[0263] Figure 10b shows the relative fluorescence intensity delivered by a QT-DMFS-C2 optical diamond magnetic field sensor from Quantum Technologies GmbH, based on diamond powder with NV centers and an optical waveguide for pump radiation and fluorescence radiation, compared to the conduction current in the line ILTG = [-3000 . . 3000]A. Figure 10b corresponds to the horizontal section through the diagram in Figure 9b.

[0264] Figure 11

[0265] Figure 11a

[0266] Figure 11a shows the magnetic field and current estimation for sensor element 0 with an assumed SNR = 1000.

[0267] Figure 11b shows the magnetic field and current estimation for sensor element 1 with an assumed SNR = 1000.

[0268] Figure 12

[0269] Figure 12a

[0270] Figure 12a shows the magnetic field and current estimation for sensor element 2 with an assumed SNR = 1000.

[0271] Figure 12b

[0272] Figure 12 shows the combined current estimate for all sensor elements (Sensor element O, Sensor element I, Sensor element 2).

[0273] Figure 13

[0274] Figure 13a

[0275] Figure 13a shows the magnetic field for three sensor elements at the same distance from the wire of the line LTG, but with different premagnetization fields relative to the current.

[0276] Figure 13b

[0277] Figure 13b shows the relative fluorescence intensity provided by the QT-DMFS-C2 optical diamond magnetic field sensor from Quantum Technologies GmbH, based on diamond powder with NV centers and an optical waveguide for pump radiation and fluorescence radiation. The different z-coordinates indicated in the legend are not relevant to the sensor performance. Figure 14

[0278] Figure 14a

[0279] Figure 14a shows the probability distribution for estimated magnetic fields for sensor element 0 with an SNR = 1000.

[0280] Figure 14b

[0281] Figure 14b shows the probability distribution for estimated magnetic fields for sensor element 1 with an SNR = 1000.

[0282] Figure 14c

[0283] Figure 14c shows the probability distribution for estimated magnetic fields for sensor element 2 with an SNR = 1000.

[0284] Figure 15

[0285] Figure 15a

[0286] Figure 15a shows the combined estimate of the current in the wire.

[0287] Figure 15b

[0288] Figure 15b shows calculated expected values ​​l c * and their absolute and relative errors. The mean absolute error is 5.0 A.

[0289] Figure 16

[0290] Figure 16 shows a scheme of relative measurements:

[0291] Figure 16a shows the relative intensity of the fluorescence radiation: Intensity of sensor element 0 / Intensity of sensor element 1.

[0292] Figure 16b shows the relative intensity of the fluorescence radiation: intensity of sensor element 1 / intensity of sensor element 2.

[0293] Figure 16c shows the relative intensity of the fluorescence radiation: Intensity of sensor element 2 / Intensity of sensor element 0.

[0294] Figure 16d shows the fluorescence ratio sensor element O / sensor element 1, derived from the calibration curve, taking into account the inversion of the relative signals with an SNR of 1000.

[0295] Figure 16e shows the fluorescence ratio sensor element 1 / sensor element 2, derived from the calibration curve, taking into account the inversion of the relative signals with an SNR of 1000.

[0296] Figure 16f shows the fluorescence ratio sensor element 2 / sensor element 0, derived from the calibration curve, taking into account the inversion of the relative signals with an SNR of 1000.

[0297] Figure 17

[0298] Figure 17 shows the calculated expected values, the deviation, and the absolute and relative errors for the combined inversion of the three relative signals. The current measurement shows a higher relative error than 15b because the same SNR was assumed. It should be emphasized that this relative measurement is conceptually immune to fluctuations in the light intensity of a common light source (pump radiation source). A key insight is therefore the use of ONE pump radiation source for multiple sensor elements. (Multiple optical diamond magnetic field sensors QT-DMFS-C2 from Quantum Technologies GmbH)

[0299] Figure 18

[0300] Figure 18 shows the magnetic field drop of a disc magnet (diameter: 3mm, thickness: 1mm) on the axis of symmetry from its surface (z = 0) for various permanent magnet materials.

[0301] Figure 19 a) Schematic representation of the optical setup used for lifetime measurements. b) Photograph of the ensemble of microdiamonds in a glass cuvette.

[0302] The lower disc shows a close-up that reveals individual crystals.

[0303] Figure 20 a) Single histogram of a sample of NV-rich diamond powder, recorded at B=0 (bin width lOps, 15 s acquisition time, 8 MHz repetition rate, 240 pW average power, APD count rate 81 kHz). The average count in the first 20 ns of 5.7 counts was subtracted. The lower window shows the residuals of a double exponential fit (a2,i = 0.675,T2,I = 6.13 ns, a2,2 = 0.325,T2,2 = 14.54 ns,X). 2 R= 1.0812). A duration of 40 ns was used for fitting from the tip. The IRF was acquired at the same count rate as the APD (FWHM = 0.2 ns). b) Histograms of the sample at different magnetic flux densities B, shown in the time domain used for fitting. The color bar shows the exponentials of a logarithmic scale with a power of 10. c) Single and double exponential fits for the data in (b) with fixed a2,i = 1 - a2,2 = 0.65. The three windows show the extracted lifetimes, the count rate at the APD, and X. 2 R for the respective adjustments as functions of B.

[0304] Figure 21

[0305] Figure 21a

[0306] Figure 21a shows a schematic representation of the optical and electrical setup used for the measurements in the frequency domain.

[0307] The optical and electrical setup for frequency domain measurements is shown in Figure 3a. A constant current source 260 drives a pump radiation source 100 at a DC operating point. The pump radiation source current supplied by the constant current source 260 to the pump radiation source 100 is, for example, a 520 nm laser diode (PLT5 520B, ams-OSRAM AG, Premstätten, Austria). The constant current source 260 can, for example, be based on an integrated laser driver circuit (NZN, iC-Haus GmbH, Bodenheim, Germany). The pump radiation source 100 emits the pump radiation 60. The DC intensity of the pump radiation 60 typically depends on the current value of the pump radiation source current supplied by the constant current source 260 to the pump radiation source 100 via a control line 290.A coupling capacitor 280 additionally couples the preferably mono-frequency and essentially sinusoidal AC control signal of a Vector Network Analyzer (VNA) 210 or an HF2LI Lock-in Amplifier (LIA) 210 for AC control of the pump radiation source 100 from an AC control line 250 into this control line 290 for the control signal for controlling the pump radiation source 100. The coupling capacitor 280 thus superimposes the preferably mono-frequency and essentially sinusoidal AC control signal of the Vector Network Analyzer (VNA) 210 or the HF2LI Lock-in Amplifier (LIA) 210 via a second RF amplifier 270 for additional AC control of the pump radiation source 100 with the DC control signal of the constant current source 260 to form the control signal for controlling the pump radiation source 100 in the control line 290.The coupling capacitor 100 thus imprints a small differential AC measurement signal onto the control signal in the control line 390. The second RF amplifier 270 therefore modulates the intensity of the pump radiation 60 from the pump radiation source 100 via the coupling capacitor 280, for example, at an input power of -9 dBm. The second RF amplifier 270 can, for example, be an AC-coupled high-frequency (RF) amplifier with 35 dB gain, operating range 1–700 MHz, and a power output of 3.2 W. An optical system, which typically includes optical functional elements such as mirrors (80, 90), lenses, filters, polarizers, beam splitters, and / or apertures, collimates and propagates the pump radiation 60 (the excitation light) to a dichroic mirror 110, which in this example is a dichroic beam splitter 100. This propagates the pump radiation 60 through the dichroic mirror 110, here exemplified by the dichroic beam splitter (DMPS567R, Thorlabs, Newton, NJ, USA).Further components of the optical system, which typically includes optical functional elements such as mirrors (80, 90), lenses, filters, polarizers, beam splitters, and / or apertures, etc., couple the pump radiation 100 into an optical waveguide 10 at a second end of the optical waveguide 10, preferably by means of an optical waveguide holder 50 and / or an aperture with a projection optic 50, for example, a High-NA Achromatic Collimator for Multimode Fibers from Thorlabs (preferably F950SMA-A SMA905 Multimode Collimator). Such an aperture with a projection optic 50 typically forms a spatial filter together with the optical waveguide 10. This limits the excitation of the emission of the fluorescence radiation 70 to a small spatial region in the sensor element 20.The optical waveguide 10 transports the pump radiation 60 to the second end of the optical waveguide 10, where the pump radiation 60 exits the optical waveguide 10, at least partially, and irradiates a sensor element 20 with this pump radiation 60. This sensor element 20 at the first end of the optical waveguide 10 typically comprises crystals with paramagnetic centers embedded in a substrate material of the sensor element 20. Preferably, the crystals are one or more diamonds and the paramagnetic centers are NV centers. This document refers to German patent applications DE 10 2023 122 657.6, DE 10 2023 115 906.2, DE 10 2023 121633.3, and DE 10 2023 134058.1, the technical teaching of which forms part of the technical teaching of this document. Instead of NV centers, other paramagnetic centers and / or other crystals can also be used.NV centers in diamond crystals are particularly suitable for these sensor elements. Typically, the substrate material of the sensor element 20 is essentially optically transparent to electromagnetic radiation with the pump radiation wavelength 60 of the pump radiation source 100. Typically, the substrate material of the sensor element 20 is also essentially optically transparent to electromagnetic radiation with the fluorescence radiation wavelength 70 of the paramagnetic centers of the crystals in the sensor element 20, which these paramagnetic centers emit as a result of irradiation of the sensor element 20 with the pump radiation 60. In the development of the technical teaching presented here, a 105 rm NV diamond fiber with a 105 pm core diameter (QT DMFS-C2, Quantum Technologies GmbH) was used as the optical waveguide 10, which incorporated such a sensor element 20.In this product, the sensor element 20 on the end facet at the first end of the optical fiber 10 contains NV-rich diamond powder in an optically transparent adhesive as a substrate material, with the crystal size being much smaller than the facet diameter of the end facet at the first end of the optical fiber 10. The fluorescence radiation 70 (here also sometimes referred to simply as fluorescence) is collected by the same optical fiber 10 and transported back to the optical fiber holder 50 and / or an aperture with a projection optic 50, i.e., the spatial filter. The sensor element 20 thus emits the fluorescence radiation 70 from the paramagnetic centers of the crystals in the sensor element 20 and typically reflects at least a portion of the pump radiation 60 back into the optical fiber 10.After passing through the optical fiber holder 50 and / or the aperture with a projection optic 50, the optical system guides the superposition of the reflected pump radiation 60 and the fluorescence radiation 70 through a first optical separation functional element to separate the electromagnetic radiation moving towards the optical fiber 10 with the sensor element 20 from the electromagnetic radiation moving away from the optical fiber 10 with the sensor element 20. In the example discussed here, the optical system guides the superposition of the reflected pump radiation 60 and the fluorescence radiation 70 through a second optical separation functional element to separate the electromagnetic radiation with the pump radiation wavelength of the pump radiation 60 from the electromagnetic radiation with the fluorescence radiation wavelength of the fluorescence radiation 70.In the example shown in Figure 3a, the first optical separation functional element is identical to the second optical separation functional element. In the example presented here in Figure 3a, the first optical separation functional element and the second optical separation functional element are implemented by a dichroic mirror 110, e.g., Thorlabs DMLP567-01" Longpass Dichroic Mirror, 567 nm Cut-On, which performs both functions. The function of the second optical separation functional element can also be implemented by an optical filter that suppresses electromagnetic radiation with pump radiation wavelength to suppress the pump radiation signal of the pump radiation 60.Since the separation effect of the dichroic mirror 110 is not perfect, in the example shown in Figure 3a, the optical system, after passing through the dichroic mirror 110, directs the electromagnetic radiation of the fluorescence radiation 70 together with the radiation components of the pump radiation 60 that are not separated due to the imperfection of the dichroic mirror 110 through a long-pass filter 120. For example, a Thorlabs FELH0650 - 025.0 mm long-pass filter with a cut-on wavelength of 650 nm was suitable as a long-pass filter 120 in the test setups. The long-pass filter 120 suppresses electromagnetic radiation with the pump radiation wavelength of the pump radiation 60. The optical system then directs the separated and purified fluorescence radiation 70 from the paramagnetic centers of the crystals of the sensor element 20 onto a photodetector 150. A silicon photodiode can serve as the photodetector 150.For this purpose, a focusing lens 140 in the optical system preferably focuses the fluorescence radiation 70 onto the photodetector 150. Optionally, a spatial filter can be inserted in the beam path upstream of the focusing lens 150. An aperture can be provided in the region of the focal point of the focusing lens 150 to effect spatial filtering of the fluorescence radiation 70 before it strikes the photodetector 150. The photodetector can also be optically connected to the focal point of the focusing lens 150 by means of another optical waveguide, unlike in the example shown in Figure 3a. For example, a silicon photodiode S5973 from Hamamatsu Photonics KK, Hamamatsu City, Japan, is suitable as the photodetector 150. In developing the proposal, a lens LA1951-A-ML - 01" N-BK7 Plano-Convex Lens, SMl-Threaded Mount, f = 25.4 mm, ARC: 350-700 nm from Thorlabs, Newton, NJ, USA, was used as an example focusing lens 140.The photodetector 150 generates a received signal 160 depending on the instantaneous intensity of the fluorescent radiation 70 incident upon it. A transimpedance amplifier 170, preferably optimally matched to the photodetector 150, amplifies the received signal 160, preferably with respect to power, so that subsequent amplifier stages can have a lower input impedance. The transimpedance amplifier 170 generates the first amplifier output signal 180 from the received signal 160 of the photodetector 150, which thus depends on the intensity of the fluorescent radiation 70. For example, an OPA847 operational amplifier (300 MHz, 1.2 kΩ) is suitable as the transimpedance amplifier 170. In this document, the transimpedance amplifier 170 is also referred to as a TIA photodiode. In some implementations, the photodetector 150 can be a device component of a transimpedance amplifier (TIA) 170.A low-noise first RF amplifier 190 typically amplifies the first amplifier output signal 180 to a second amplifier output signal 200 in amplitude. For example, the first RF amplifier 190 could be a low-noise RF amplifier (20 dB, 0.1–2000 MHz). The first RF amplifier 190 feeds the second amplifier output signal 200 back to the Vector Network Analyzer (VNA) 210 or the Lock-in Amplifier (LIA) 210 (HF2LI Lock-in Amplifier (LIA) 210). The Vector Network Analyzer (VNA) 210 or the Lock-in Amplifier (LIA) 210 also generate the preferably mono-frequency and essentially sinusoidal AC control signal of the Vector Network Analyzer (VNA) 210 or the Lock-in Amplifier (LIA) 210 for AC control of the pump radiation source 100 and feed it into the said AC control line 250.Suitable Vector Network Analyzers (VNA) 210 or Lock-in Amplifiers (LIA) 210 (HF2LI Lock-in Amplifier (LIA) 210) include, for example, a PicoVNA 106 Vector Network Analyzer (VNA) (Pico Technology, St Neots, United Kingdom) or an HF2LI Lock-in Amplifier (LIA) (Zürich Instruments AG, Zurich, Switzerland), which then also control the modulation of the laser diode. If the sensor system described above is to be calibrated, the device can be supplemented with auxiliary devices.The control computer 230 and / or another computer system can, for example, energize an electromagnet 30 by means of a computer-controlled power supply, which then generates a magnetic flux density that floods the sensor element 20 of the optical waveguide 10 and thus influences the intensity of the fluorescence radiation and its phase shift of the temporal course of its temporal intensity modulation compared to the temporal course of the temporal intensity modulation of the pump radiation by the AC control signal of the Vector Network Analyzer (VNA) 210 or the Lock-in Amplifier (LIA) 210.A Hall sensor 40 or another suitable magnetic field sensor can detect the actual magnetic flux density and transmit it to the control computer 230 via a sensor evaluation system (not shown) and a data bus. This allows the control computer to adjust the current to the electromagnet 30 so that a predetermined magnetic flux density flows through the sensor element 20 of the optical fiber 10. The electromagnet 30, which, as already described, is preferably connected to a PC-controlled power supply (not shown), is monitored by a Hall effect sensor, the aforementioned Hall sensor 40, and typically allows the application of magnetic flux densities from 0 to 120 mT. The modulation of the excitation light, the pump radiation 60, can be adjusted by an offset P. DC and amplitude of the AC component P ACTo describe the modulation, it was measured via the reflection of the pump beam of the pump radiation 60 after removal of the long-pass filter 120 at the DC-coupled TIA output of the transimpedance amplifier 170. Its ratio P A C / PDC ' st Figure 3(b) illustrates this. It shows a high-pass characteristic and an overall non-constant ratio resulting from the RF amplifier 270 (second RF amplifier 270) for the laser modulation of the pump radiation source 100. During the development of the technical teaching presented in this document, the output power of the laser diodes of the pump radiation source 100, measured in front of the dichroic mirror 110, fluctuated between 33 and 37 mW depending on the modulation frequency. The use of a Vector Network Analyzer (VNA) 210 requires, in particular, the calibration of the scattering parameter S determined by the Vector Network Analyzer (VNA) 210. 21i.e., the ratio of magnitude and phase of the signals at terminals one and two of the Vector Network Analyzer (VNA) 210. Therefore, the isolation and throughput of the system can be calibrated by either blocking or allowing the fluorescence path of the fluorescence radiation 70 to pass unimpeded to the photodetector 150 by means of the shutter 130, ensuring that no magnetic field is applied to the sensor element 20 of the sensor head at the first end of the optical waveguide 10. In this way, the device automatically corrects the reference frequency response.

[0308] Figure 21b

[0309] Firstly, it shows the modulation of the excitation light P / PACDC measured via reflections at a long-pass filter removed from the beam path, and secondly, the phase difference ( Bbetween the time course of the intensity of the pump radiation 60 (excitation) and the time course of the intensity of the fluorescence radiation 70 measured via the received signal 160.

[0310] Figure 22

[0311] Measurements of the complex transfer function H r = | H r |exp(jzH r The top and bottom lines show the amount | H r | or the phase at home r H r The transfer function H(ja>, B) is relative to H(ja>, B=0). (a) - (d) were determined directly by sweeping the modulation frequency with a VNA calibrated at B=0. For (e) and (f), a LIA was used at the respective modulation frequencies to capture the magnitude and phase (4th-order demodulator low-pass filter, 3.89 Hz). The data were then normalized to the data point at B=0.

[0312] Figure 23 a) Measurement quantities | H r | and at home rat f = 13 MHz (4th order demodulator, 3.89 Hz low-pass filter). The

[0313] A laser intensity disturbance with a frequency fdist = 1 Hz was switched on at 5 s. For comparison with the magnetic contrasts of the two quantities, a magnetic field of > 100 mT was applied at 12 s. b) RMS value of the AC component, including magnitude and phase, at the disturbance frequency fdist in a range of 0.1 - 10 Hz. A bias field corresponding to half the magnetic contrast was applied.

[0314] Figure 24

[0315] Noise spectral density based on the magnitude S r(a) and the phase S,|, (b). The excitation frequency was set to f = 13 MHz, and a bias field of B = 20 mT was applied. The data were recorded by the LIA using a fourth-order low-pass filter at cutoff frequencies of 88 kHz, 200 Hz, and 2 Hz, corresponding to the green, blue, and orange traces. They were subsequently converted into magnetic field values ​​using the linear matching shown in the respective insets and transformed into spectral densities using the Weich method.

[0316] Figure 25

[0317] Figure 25 shows the dependence of the lifetime of the fluorescence radiation from the NV centers of the sample on the magnetic flux density B acting on the NV centers. The measurement of the lifetime is described in the prior art, for example, in the document

[0318] DE 102019 128 932 Al is known. The novel aspect is the use of lifetime measurement to measure magnetic flux density and / or another physical quantity related to lifetime, in order to be independent of fluctuations in the intensity of the pump radiation. The lifetime measurement is insensitive to laser noise, etc. This is a significant advantage.

[0319] Figure 26

[0320] Figure 26 shows the principle of fluorescence spectroscopy.

[0321] Figure 27

[0322] Figure 27 shows the curves of Figures 22c and 22d for further magnetic flux densities passing through the sensor element 20 of the optical waveguide 10.

[0323] Figure 28

[0324] Figure 28 shows a typical sensor system for measuring the current value of the line current I LTG in the line LTG and for determining the current direction of the line current Line current ILTG in the line LTG.

[0325] The two permanent magnets PI and P2 generate a premagnetization field with opposite polarity in their respective magnetic circuits. This causes the three sensor elements 20 to be flooded with a different total magnetic flux density. The middle, zero-point sensor element 20 receives no premagnetization field. A common pump radiation source 100 generates common pump radiation 60. A pump radiation distribution system consisting of semi-transparent mirrors SM, mirrors M, and dichroic mirrors 110 distributes the pump radiation 60 equally among the three sensor elements 20. A separate optical waveguide 10 transports the respective pump radiation 60 to each sensor element 20. There, the optical waveguide irradiates the respective sensor element 20 with the pump radiation 60, specifically at the paramagnetic centers in a disordered multitude of crystals.Preferably, the paramagnetic centers in a disordered plurality of crystals are diamond crystals (diamond dust) comprising NV centers as paramagnetic centers. As a result of irradiation with pump radiation 60, the respective paramagnetic centers of the respective sensor elements 20 emit fluorescence radiation 70 with a fluorescence wavelength whose intensity depends on the respective total flux density at the respective location of the respective sensor element 20. The fluorescence wavelength is typically longer than the pump radiation wavelength. The respective optical waveguide detects the respective fluorescence radiation 70 of its respective sensor element 20 and transports the respective fluorescence radiation to a respective means for separating the respective fluorescence radiation 70 of the respective sensor element 20 from the pump radiation 60.In the example shown in Figure 27, these means comprise respective dichroic mirrors 110, which separate the respective fluorescence radiation 70 of the respective sensor element 20 from the pump radiation 60 and direct it onto a respective photodetector 150, which generates a respective fluorescence measurement signal for the respective sensor element 20. The evaluation and control device CTR acquires the respective values ​​of these respective fluorescence measurement signals for the respective intensity of the respective fluorescence radiation of the respective sensor elements 20, for example, by means of a multi-channel analog-to-digital converter ADC, which is preferably part of the evaluation and control device CTR.Preferably, at least one computer core of the evaluation and control device CTR executes a computer-implemented and / or machine-implemented evaluation program to evaluate the respective values ​​of these respective fluorescence measurement signals for the respective intensity of the respective fluorescence radiation of the respective sensor elements 20, the program code and data of which are preferably stored at least temporarily in one or more memory units (MEM) of the evaluation and control device CTR. Preferably, the computer core thus calculates the previously described ratio of these values ​​of different sensor elements 20. Since the device uses a common pump radiation source 100 for all sensor elements 20, fluctuations in the intensity of the pump radiation 60 affect these values ​​equally and can be eliminated.By using at least three sensor elements 20, the computer core of the evaluation and control device CTR can determine the current direction of the line current ILTG from the respective values ​​of these fluorescence measurement signals for the respective intensity of the respective fluorescence radiation of the respective sensor elements 20. Preferably, the evaluation and control device CTR transmits the acquired signed current measurements for the line current ILTG to a higher-level system via a data bus DB.

[0326] List of reference signs

[0327] The following abbreviations are used in this manuscript:

[0328] glossary

[0329] ZPL table

[0330] The table is only an exemplary compilation of some possible paramagnetic centers. These can be used in their respective crystals as paramagnetic centers of the sensor element (20). This document particularly recommends the use of NV centers in diamond a as paramagnetic centers of the sensor element (20). The functionally equivalent use of other paramagnetic centers in other materials or crystals is expressly possible. The pump radiation wavelengths Ä, pmp Pump radiation wavelengths 60 are also examples. Other pump radiation wavelengths % pmp are generally possible if they are shorter than the wavelength of the ZPL to be excited.

[0331] Material Fault Center ZPL exemplary of pump radiation

[0332] Crystal wavelength (E. pmp ) in sensor element 20

[0333] Diamond NV center 520nm, 532nm

[0334] Diamond SiV-Centrum 738 nm 685 nm Diamond GeV-Centrum 602 nm 532 nm

[0335] Diamond SnV-Zentrum 620 nm 532 nm

[0336] Diamond PbV-Zentrum 520 nm, 450 nm

[0337] 552 nm 715 nm 532 nm

[0338] Diamond STl-Zentrum 555 nm 532 nm

[0339] Diamond TR12-Zentrum 471 nm 410 nm

[0340] Silicon G-Zentrum 1278.38 nm 637 nm

[0341] Silicon carbide Vsi-Zentrum 862 nm(Vl) 4H, 730 nm

[0342] 858.2 nm(Vl') 4H 730 nm

[0343] 917 nm(V2) 4H, 730 nm

[0344] 865 nm(Vl) 6H, 730 nm

[0345] 887 nm(V2) 6H, 730 nm

[0346] 907 nm(V3) 6H 730 nm

[0347] Silicon Carbide DV-Zentrum 1078-1132 nm 6H 730 nm

[0348] Silicon carbide VcVsi-Zentrum 1093-1140 nm 6H 730 nm

[0349] Silicon Carbide CAV-Zentrum 648.7 nm 4H, 6H, 3C 730 nm

[0350] 651.8 nm 4H, 6H, 3C 730 nm

[0351] 665.1 nm 4H, 6H, 3C 730 nm

[0352] 668.5 nm 4H, 6H, 3C 730 nm

[0353] 671.7 nm 4H, 6H, 3C 730 nm

[0354] 673 nm 4H, 6H, 3C 730 nm

[0355] 675.2 nm 4H, 6H, 3C 730 nm

[0356] 676.5 nm 4H, 6H, 3C 730 nm

[0357] Silicon Carbide NcVsi Center 1180nm-1242nm 6H 730nm

[0358] In the document revealed here, NV centers in diamond are particularly favored. These were tested during the development of the technical doctrine of this document.

[0359] Difference between Lock-in Amplifiers (LIA) and Vector Network Analyzers (VNA)

[0360] A vector network analyzer (VNA) and a lock-in amplifier are, for the purposes of this document, two distinct instruments that can be used in the sensor technology presented here. The vector network analyzer (VNA) is a measuring device generally used primarily for characterizing and analyzing electrical networks and components in high-frequency and microwave applications. It enables the precise measurement of the amplitude and phase of electrical signals, in this case the AC drive signal in relation to the second amplifier output signal, across a wide frequency range. It provides comprehensive information about the transmission characteristics of components such as filters, amplifiers, antennas, optical systems, fiber optic cables, paramagnetic centers, lines, and other device parts involved in the signal path.In contrast, a lock-in amplifier (LIA) is a specialized device typically used for detecting and processing weak signals in a high-noise environment, particularly when using spreading codes. It typically operates with a reference signal—in this case, the AC drive signal—that is synchronized with the signal being measured, in this case, the second amplifier output signal, thus separating the desired signal from unwanted noise. This enables the measurement of extremely low-amplitude signals that could not normally be reliably detected by conventional amplifiers. While a vector network analyzer has broad applicability and provides detailed information about the characteristics of the signal path of the system presented here, a lock-in amplifier focuses on the precise detection and analysis of weak signals in highly noisy environments.The choice between these two device components therefore depends on the specific requirements of the measurement task as well as the desired accuracy and sensitivity of the measurement. However, for the purposes of the claims presented here, a VNA is encompassed by the term lock-in amplifier.

[0361] literature

[0362] [1] European Union - HORIZON Academy. TRL Assessment. https: / / horizoneuropencpportal.eu / store / trl-assessment.

[0363] [2] EnBW. Voltage levels, https: / / www.enbw.com / energie-entdecken / distribution-and-transport / electricity-network / #transport-inlet.

[0364] [3] Statista. Number of solar installations in Germany. https: / / de.statista.com / statistik / daten / studie / 1448887 / umfrage / anzahl-photovoltaikanlagen-in-deutschland / , February 2024.

[0365] [4] German Environment Agency; Agency for Renewable Energies; German Wind Energy Association (BWE); Fraunhofer Institute for Solar Energy Systems (ISE); German Wind Guard. Number of wind turbines in Germany. https: / / de.statista.com / statistik / daten / studie / 20116 / umfrage / anzahl-der-windkraftanlagen-in-deutschland-seit-1993 / , 2024.

[0366] [5] EnBW. Network control, smart grid. https: / / www.enbw.com / energie-entdecken / distribution- and-transport / network-control / , January 2015.

[0367] [6] U. Share of renewable energies in total energy demand. https: / / www.umweltbundesamt.de / themen / klima-energie / erneuerbare-energien / renewable-energies-in-zahlen, March 2024.

[0368] [7] Dr. Andreas Kemmler. Development of gross electricity consumption until 2030, https: / / www.prognos.com / de / projekt / entwicklung-des-bruttostromverbrauches-bis-2030, 2021.

[0369] [8] EnBW. Expansion towards a smart grid. https: / / www.enbw.com / energie-entdecken / verteilung- und-transport / stromnetz / ausbau.html

[0370] [9] 3M Energy. Sensor requirements for medium-voltage grids. https: / / www.3mdeutschland.de / 3M / de_DE / energie-de / stories / ganze-story / ?storyid=a496cdbb-92d6-43e2-8323-fe3330ac9357.

[0371]

[0010] Wikipedia. substation. https: / / de.wikipedia.0rg / wiki / T ransformratenstation.

[0372]

[0011] Wikipedia. Overhead power line. https: / / de.wikipedia.org / wiki / Freileitung, March 2024.

[0373]

[0012] Drahtzieh und Kunststoffaufbereitungs GmbH. Catalog of overhead power lines. https: / / bdk-gmbh.de / wp-content / uploads / 2018 / 10 / download-katalog-freileitungsseile.pdf, 2015.

[0374]

[0013] GE Grid Solutions. Optical current transformer, November 2017.

[0375]

[0014] ABB Power Technologies. Measuring transducer for outdoor installation, February 2006.

[0376]

[0015] Roland Stierlin. Fiber optic sensors: Part 2: Fiber optic current sensors. Bulletin SEV / VSE, Volume 82 (Issue 1): 21-30, December 1991.

[0377]

[0016] Michael Mendik, Klaus Bohnert, Richard Thomas. Current measurement with light, fiber optic current sensor in a high-voltage circuit breaker. Fig.

[0378]

[0017] Robert Staacke, Roger John, Ralf Wunderlich, Ludwig Horsthemke, Wolfgang Knolle, Christian Laube, Peter Gl'osek'otter, Bernd Burchard, Bernd Abel, and Jan Meijer. Isotropic scalar quantum sensing of magnetic fields for industrial applications. Advanced Quantum Technologies, 3(8):2000037, 2020.

[0379]

[0018] Wikipedia. Biot-Savart law. https: / / en.wikipedia.org / wiki / Biot%E2%80%93Savart_law, 2014.

[0380]

[0019] Wolfgang Demtröder. Experimental Physics 2: Electricity and Optics. Springer Berlin Heidelberg, 2013.

[0381]

[0020] Mohammed Attrash, Oleg Shtempluck, and Eyal Buks. High-temperature spectroscopy of ensembles of nitrogen vacancy centers in diamond. Journal of Applied Physics, 133(9), March 2023.

[0382]

[0021] Stefan Ernst, Patrick J. Scheidegger, Simon Diesch, Luca Lorenzelli, and Christian L. Degen. Temperature dependence of photoluminescence intensity and spin contrast in nitrogen vacancy centers, 2023.

[0383]

[0022] Ludwig Horsthemke and Lutz Langguth. QuantumTechnologies - Invention disclosure 'QUA65' (unpublished). 2024.

[0384]

[0023] Ludwig Horsthemke, Jens Pogorzelski, Dennis Frederik Hoffmann, Lutz Langguth, Robert Staacke, Christian Laube, Wolfgang Knolle, Markus Gregor, and Peter (fc Excited-state lifetime of nv centers for all-optical magnetic field sensing. Sensors, 24(7):2093, March 2024.

[0385]

[0024] BOMATEC. BMN-30UH / ST

[0386] https: / / www.bomatec.com / de / produkte / magnete / gesinterte-ndfeb- magnete, 2023.

[0387]

[0025] BOMATEC. BMSG-30L

[0388] https: / / www.bomatec.com / de / produkte / magnete / gesinterte-smco- magnete, 2023.

[0389]

[0026] BOMATEC. BMHFa-32 / 27

[0390] .https: / / www.bomatec.com / de / produkte / magnete / gesinterte- ferritmagnete.

[0391]

[0027] BOMATEC. BMAsa-35 / 5

[0392] .https: / / www.bomatec.com / de / produkte / magnete / sintered-Alnico magnets.

[0393]

[0028] COMSOL. Comsol Multiphysics 6.2, 2023.

[0394]

[0029] REDURGmbH. Requirements for accuracy

[0395] https: / / www.redur.de / wissen / stromwandler-lxl / anforderungen-an-die- korrektur /

Claims

1. Device for measuring a line current, comprising at least three lock-in amplifiers (210) or an evaluation and control device (CTR) configured to perform these functions of at least three lock-in amplifiers (210), a pump radiation source (100), an optical system, at least three sensor elements (20), a line (LTG) through which a line current (ILTG) flows, at least three respective receiving systems (SM, M, 150, 160, 170, 180, 190, 200) which may have common functional elements assigned to each of these respective sensor elements (20), at least three respective photodetectors (150) as a respective part of each receiving system assigned to each of the at least three sensor elements (20), and at least two magnetic field sources, in particular two permanent magnets (Pl, P2),wherein the at least three respective sensor elements (20) of the at least three, Sensor elements (20) each comprise paramagnetic centers and wherein the respective lock-in amplifiers (210) and / or a respective input channel of the evaluation and control device (CTR) are each assigned to a respective sensor element (20) and wherein the evaluation and control device (CTR) generates an AC drive signal and wherein the pump radiation source (100) pump radiation (60) with a Pump radiation wavelength (X) pmp) with a pump radiation intensity dependent on a control signal which depends on the AC control signal of the evaluation and control device (CTR), emits into the optical system and wherein the optical system distributes the pump radiation (60) of the pump radiation source (100) onto the sensor elements (20) and thus irradiates the respective paramagnetic centers of the respective sensor elements (20) with this respective pump radiation (60) for the respective sensor element (20) and wherein the respective paramagnetic centers of the respective sensor elements (20) each emit a respective fluorescence radiation (70) whose respective intensity and / or their respective temporal phase shift relative to the AC drive signal of the evaluation and control device (CTR) and depends on the respective value of a respective magnetic flux density acting on the respective paramagnetic centers of the respective sensor element (20) emitting these, and wherein the magnetic field sources set a different premagnetization field for each of the at least three sensor elements (20), and wherein the optical system separates the respective fluorescence radiation (70) of a respective sensor element (20) from the pump radiation (60) in a respective beam path of the optical system, and wherein the optical system directs the respective fluorescence radiation (70) of a respective sensor element onto a respective photodetector (150) of the receiving system (150, 160, 170, 180, 190, 200), which is assigned to exactly this respective sensor element (20).wherein the receiving system generates a respective output signal of the respective photodetector (150) depending on the respective intensity of the respective fluorescence radiation (70) of the respective sensor element (20) emitted into the photodetector (200), and wherein the respective lock-in amplifiers (210) evaluate the respective output signal of the respective photodetector (150) taking into account the AC drive signal of the evaluation and control device (CTR), and / or wherein the evaluation and control device (CTR) evaluates the respective output signal of the respective photodetector (150) with regard to an estimation of the signal component of the AC drive signal of the evaluation and control device (CTR) in the respective output signal of the respective photodetector (150), in particular with regard to magnitude and / or phase,evaluate and generate a corresponding measured value, and the evaluation and control device (CTR) determines an estimated value for the magnitude of the line current (ILTG) in the line (LTG) based on these respective measured values.

2. Device according to claim 1, wherein the AC drive signal of the evaluation and control device (CTR) is a monofrequency sinusoidal signal with substantially only one frequency, wherein substantially here is to be understood as meaning that any deviations of this monofrequency sinusoidal signal from an ideal monofrequency sinusoidal signal are so small that their effects on the device in question are negligible. The application of the claimed technical teaching can be neglected, and wherein the frequency of the AC drive signal is at a frequency of 22MHz(-4MHz / +7MHz) and / or lies in a frequency range of an operating point range (430) from 18MHz to 29MHz and 3. Device according to claim 1 or 1, wherein the respective photodetectors (150) are configured to generate a respective fluorescence measurement signal for the respective sensor element (20), and wherein the computer core of the evaluation and control device (CTR) is configured to determine the current direction of the conduction current (ILTG) from the detected respective values ​​of these respective fluorescence measurement signals for the respective intensity of the respective fluorescence radiation of the respective sensor elements (20) by using at least three sensor elements (20).

4. Device according to one of claims 1 to 3, wherein the lock-in amplifier (210) and / or a control computer (230) determines a measured value of the phase shift (A) between the AC drive signal of the evaluation and control device (CTR) and the time course of the output signal of the respective photodetector (150). r the transfer function Hr = | H r |exp(jzH r ) used.

5. Device according to one of claims 1 to 4, wherein the device consists of the detected measured value of the phase shift A between the AC control signal of the evaluation and control device (CTR) and the time course of the output signal of the respective photodetector (150) and / or from the measured value zH r the transfer function H r = | H r |exp(jzH r ) a respective measured value for a respective magnetic flux density (B(t)) at the location of the respective sensor element (20) is determined.

6. Device according to one of claims 1 to 5, wherein one or more paramagnetic centers of one or more crystals of the respective sensor element (20) are each one of the paramagnetic centers PbV, GeV, SiV, NV, STI, TRI, TR12 in a diamond crystal.

7. Device according to any one of claims 1 to 6, wherein one or more paramagnetic centers of one or more crystals of the respective sensor element (20) are NV centers in one or more diamonds and wherein the pump radiation wavelength (X pmp ) the common pump radiation (60) lies in a range of 500nm to 570nm.

8. Method for measuring a line current, comprising the steps of: in particular providing at least three lock-in amplifiers (210) or an evaluation and control device (CTR) configured to perform these functions of these lock-in amplifiers (210), in particular providing a common pump radiation source (100); in particular providing an optical system; Providing at least three sensor elements (20), wherein these sensor elements (20) comprise paramagnetic centers; In particular, providing a receiving system (150, 160, 170, 180, 190, 200) with a photodetector (150) as part of the receiving system; Generating an AC control signal, in particular by means of an evaluation and control device (CTR); Emitting and / or generating pump radiation (60) with a pump radiation wavelength (X pmp ), in particular by means of a pump radiation source (100), in particular into the optical system, with a pump radiation intensity depending on a control signal which depends on the AC control signal, in particular the evaluation and control device (CTR); Distributing the pump radiation (60) to the at least three sensor elements (20); Irradiating the respective paramagnetic centers of the respective sensor elements (20) with this respective pump radiation (60), in particular by means of the optical system, wherein the respective paramagnetic centers of the respective sensor element (20) each emit respective fluorescence radiation (70), the respective intensity and / or the respective temporal phase shift relative to the AC control signal, in particular the evaluation and control device (CTR), depends on the respective value of a respective magnetic flux density acting on these respective emitting paramagnetic centers of the respective sensor elements (20); Separate the respective fluorescence radiation (70) from the pump radiation (60), in particular in a beam path of the optical system, especially by means of the optical system; In particular, the respective irradiation of the respective fluorescence radiation (70), especially onto a respective photodetector (150) of the receiving system (150, 160, 170, 180, 190, 200) especially by means of the optical system; Generating an output signal of the respective photodetector (150) depending on the respective intensity of the respective fluorescence radiation (70) of the respective sensor element (70) that is assigned to the respective photodetector (150), in particular by means of the receiving system;wherein the method comprises the step of evaluation, in particular by means of a respective lock-in amplifier (210) and / or by means of the evaluation and control device (CTR) which is configured to perform these functions of these lock-in amplifiers (210), from the respective output signal of the respective photodetector (150) and the AC drive signal, in particular of the evaluation and control device (CTR), and the generation of a respective measured value depending on the result, and wherein the method comprises the step of estimating the magnitude of the line current (ILTG) in the line (LTG), in particular by means of the evaluation and control device (CTR) and in particular by using computer-implemented and / or machine-implemented methods, based on these respective measured values.

9. Method according to claim 8, wherein the AC drive signal is a monofrequency sinusoidal signal with substantially only one frequency, wherein substantially here is to be understood as meaning that any deviations of this monofrequency sinusoidal signal from an ideal monofrequency sinusoidal signal are so small that their effects for the relevant application of the technical teaching claimed herein can be neglected, and wherein the frequency of the AC drive signal is at a frequency of 22MHz(-4MHz / +7MHz) and / or is in a frequency range of an operating point range (430) from 18MHz to 29MHz.

10. The method of claim 8 or 9, comprising the steps of: Generating an output signal of the respective photodetector (150) depending on the respective intensity of the respective fluorescence radiation (70) of the respective sensor element (70) that is assigned to the respective photodetector (150), in particular by means of the receiving system, wherein the respective photodetectors (150) are configured to generate a respective fluorescence measurement signal for the respective sensor element (20), and Determining the current direction of the conduction current (ILTG) by using at least three sensor elements (20) in particular by the computer core of the evaluation and control device (CTR) in particular from the recorded respective values ​​of these respective fluorescence measurement signals for the respective intensity of the respective fluorescence radiation of the respective sensor elements (20).

11. Method according to one of claims 8 to 10, wherein for the respective determination of the respective measured values ​​of the respective phase shift A between the AC control signal, in particular the evaluation and control device (CTR), and the respective time course of the respective output signal of the respective photodetector (150), the respective measured value zH r the respective transfer function H r = | H r | exp(jzH r ) in particular by the respective lock-in amplifier (210) and / or the evaluation and control device (CTR).

12. Method according to one or more of claims 8 to 11, wherein the respective measured value of the respective phase shift (A ) between the AC control signal of the lock-in amplifier (210) and the respective time course of the respective output signal of the respective photodetector (150) and / or the respective measured value zH rthe respective transfer function H r = | H r | exp(jzH r ) a respective measured value for a respective magnetic flux density at the respective location of the respective sensor element (20) is determined.

13. Method according to one or more of claims 8 to 12, wherein one or more paramagnetic centers of one or more crystals of the respective sensor element (20) of one of the paramagnetic centers PbV, GeV, SiV, NV, STI, TRI, TR12 are in a diamond crystal.

14. Method according to one or more of claims 8 to 13, wherein one or more paramagnetic centers of one or more crystals of the respective sensor element (20) are NV centers in one or more diamonds and wherein the pump radiation wavelength (X pmp ) the pump radiation (60) lies in a range of 500nm to 570nm.

Images