Method for determining a rectifier stage output current and / or mains-side currents of a frequency converter

Abstract

A method for determining output currents of a rectifier stage and / or mains-side currents (iu, iv, iw) of a frequency converter (1) with a passive rectifier stage (3), an inverter (4), a DC link with an DC link inductor (Ldc) and an DC link capacitor (Cdc) between the rectifier stage (3) and the inverter stage (4) is described. In a frequency converter, the current information for the mains-side currents (iu, iv, iw) should be obtained without a current sensor on the mains side (2). For this purpose, the method includes the step of calculating a current in the DC link (5) using at least one voltage value (Urec) and characteristics of the rectifier (3) in the DC link (5) and / or mains-side currents to form a corrected current using the calculated current and a measured current or currents or a fraction of a measured current or currents.

Description

[0001] The present invention relates to a method for determining the output current of a rectifier stage and / or mains-side currents of a frequency converter with a passive rectifier, an inverter, a DC link with an DC link inductor and an DC link capacitor between the rectifier and the inverter and / or an inverter on the mains side of the rectifier stage.

[0002] Such a procedure is known, for example, from DE 10 2016 224 937 A1.

[0003] DE 10 2007 016 330 A1 describes a method for determining a DC link current, DC link power, and phase current, as well as a corresponding power supply device. The method includes the step of calculating a current in the DC link by using at least one voltage value and the properties of the DC link to generate a current.

[0004] A frequency converter is typically used to convert electrical input power supplied by a mains supply at a predetermined frequency into electrical output power at a different, controllable frequency. Currents and voltages supplied from the mains side are rectified by the rectifier stage and smoothed to some extent in the DC link by the DC link capacitor, DC link inductor, and / or the AC inductor on the mains side. The DC voltage and current in the DC link are then converted into AC voltages and currents by the inverter stage.

[0005] A frequency converter with a passive rectifier stage is interesting because it is more cost-effective than a controlled PWM rectifier stage.

[0006] In many cases, the user of such a frequency converter is interested in obtaining real-time information about the rectifier-side and / or grid-side current(s). However, this requires current sensors, which would negatively impact the cost of the frequency converter and typically require significant additional space. This information can be used to assess the frequency converter's harmonic power according to a European standard, such as IEC 61000-3-12, or simply to evaluate the thermal condition of the rectifier stage for protection and lifetime monitoring purposes.

[0007] The invention is based on the objective of providing current information about the output current of the rectifier stage and / or about mains-side currents without current sensors on the mains side, the rectifier stage and the rectifier side of the intermediate circuit.

[0008] This task is solved by a method of the type described above, in that the method includes the step of calculating a current on the rectifier side of the intermediate circuit and / or mains-side currents using voltage values ​​and properties of the rectifier stage to form a corrected current using the calculated current and one or more measured currents or a fraction of one or more measured currents.

[0009] Voltage values ​​are generally easy to determine, as voltage sensors are comparatively less expensive than current sensors and require less space. Since the rectifier stage has a DC link inductor and / or an AC inductor and a DC link capacitor, it is possible to derive one or more currents from voltages across one or more of these components, in combination with a measured current or a fraction thereof. This is referred to as the "corrected current" or "corrected currents" for the following explanation. The "corrected current" and / or "corrected currents" are an estimate of the current flowing at the output of the rectifier stage and / or an estimate of the line-side currents.

[0010] In one embodiment of the invention, the calculated current and / or the absolute value of the calculated grid-side currents each has an AC component and a DC component, wherein each corrected current is determined by separating the calculated AC and DC components and subsequently adjusting the calculated DC component by a measured current or a fraction of a measured current from the inverter side of the frequency converter.

[0011] In one embodiment of the invention, the calculated current and / or currents include only an AC component; i.e., the corrected current and / or currents are obtained by adding a DC component from a measured current from the inverter side of the frequency to the calculated current and / or currents.

[0012] In one embodiment of the invention, a calculated current is obtained by using the properties of the DC inductor and measured voltages across the inductor. For example, the properties of the DC inductor could be its inductance. Another example could be the inductance of the inductor in combination with a series and / or parallel resistor.

[0013] In one embodiment of the invention, a calculated current is obtained by using the properties of the DC capacitor and the measured voltages across the capacitor. For example, the properties of the DC capacitor could be its capacitance. Another example could be the capacitance of the capacitor in combination with a series and / or parallel resistor.

[0014] In one embodiment of the invention, a calculated current is obtained by using the properties of the AC inductor and measured voltages across the inductor. For example, the properties of the AC inductor could be its inductance. Another example could be the inductance of the inductor in conjunction with a series and / or parallel resistor.

[0015] In one embodiment of the invention, a calculated current is obtained by using the properties of the DC capacitor and the properties of the DC inductor, as well as measured voltages across the components and / or the properties of another component on the rectifier side of the frequency converter. Examples of component properties are included in the three preceding embodiments of the invention. An example of the properties of another component on the rectifier side of the frequency converter is the voltage drop across a rectifier silicon device in the rectifier stage.

[0016] In one embodiment of the invention, the calculated current is obtained by using the characteristics of the DC capacitor, the DC inductor, and / or the AC inductor, as well as measured voltages across the components and / or characteristics of any other component in the rectifier stage of the frequency converter. Examples of component characteristics are included in the four preceding embodiments of the invention. An example of the characteristics of another component on the rectifier side of the frequency converter is the voltage drop of a rectifier silicon device in the rectifier stage when switched on. Another example is the characteristics of the AC inductor on the mains side.

[0017] In one embodiment of the invention, a direct current is measured and used in combination with a calculated current, the combination of the measured and calculated current forming the corrected current. In many cases, current sensors are already present on the inverter side of the frequency converter, which are used to control the inverter stage. The measured current can be derived from the information of such a current sensor or sensors. The measured current can be used, for example, to adjust the calculated current to account for inaccuracies in voltage measurements, calculation models, and component properties.

[0018] In one embodiment of the invention, the DC component or a fraction of the DC component of the measured current from the inverter side of the frequency converter is used to adjust a calculated current.

[0019] In one embodiment of the invention, a rectifier current is measured in at least one of the two inverter-side DC link rails. Alternatively, a DC current is measured by using current information from three of the legs of the inverter stage or at least two of the outputs of the inverter stage. If the current is measured in either of these two ways, information about the total current on the inverter side of the DC link can be obtained according to the prior art.

[0020] In a further embodiment of the invention, an average value or a DC component of a calculated current is determined. A fraction of this average value and a fraction of the average value or DC component of the measured current are used to adjust a calculated current so that the corrected current is achieved.

[0021] In one embodiment of the invention, an average value or a DC component or a fraction thereof of a calculated current is derived, the average value or a DC component or a fraction thereof is subtracted from the calculated current, and an average value or a DC component or a fraction thereof of the measured current is added to the difference to form a corrected current.

[0022] In one embodiment of the invention, the calculated current is derived by integrating a voltage drop across the DC link inductance. The voltage drop across the DC link inductance can be determined by measuring a voltage on the output side of the rectifier stage and a voltage downstream of the DC link inductance, for example, a voltage across the DC link capacitor. The product of the inverse inductance of the inductor and the integral of the voltage over time yields the calculated current. In some cases, this is sufficient to arrive at the corrected current. However, if model inaccuracies or errors in the voltage measurement are too large, drift can occur during integration. In this case, it is particularly advantageous to correct the calculated current using the measured current.

[0023] In one embodiment of the invention, the voltage drop is calculated from the difference between an output voltage of the rectifier stage and an intermediate circuit voltage. The intermediate circuit voltage can be measured, for example, via the intermediate circuit capacitor.

[0024] In one embodiment of the invention, the output voltage of the rectifier stage is calculated from the instantaneous maximum value minus the instantaneous minimum value of the mains voltages on the input side of the rectifier stage. As already mentioned, voltage measurement can be carried out at low cost, even on the input side of the rectifier stage.

[0025] In one embodiment of the invention, a calculated current per network phase is derived by integrating the voltage drop across the AC inductor in each network phase. The voltage drop across the AC inductor per phase can be determined by measurement. The product of the inverse inductance of the AC inductor and the integral of the voltage over time yields the calculated current. This is sufficient in some cases to obtain a corrected current. However, if model inaccuracies or errors in the voltage measurement are too large, drift can occur during integration. In this case, it is particularly advantageous to correct a calculated current with a fraction of the measured current.Since the calculated current in this embodiment of the invention is an alternating current, the absolute value of the calculated current should be used for the correction step using the measured direct current of the inverter side of the frequency converter.

[0026] Alternatively, instead of integrating the voltage drop across an AC inductor in each phase, a two-phase vector representation of the voltage drop can be performed in a manner known in the art. A calculated current can then be determined from this vector representation.

[0027] In another embodiment of the invention, the calculated current is derived by differentiating the voltage across the DC link capacitor and multiplying by the capacitor's capacitance. The calculated current and an average of the measured current are then added to form the corrected current. This is another way to obtain the DC link current ripple. This involves some lifetime considerations, as the DC capacitance varies over the frequency converter's lifetime. Furthermore, the differentiation principle is considered less robust compared to integration.

[0028] In one embodiment of the invention, the zero crossings of the mains voltages are used to derive each mains phase current from a corrected current that represents the output current of the rectifier stage. The crossings indicate which of the phases of the passive rectifier stage are conducting and which are non-conducting. This is a simple way to distribute the corrected current, which represents the output current of the rectifier stage, across the mains phases.

[0029] In a further embodiment of the invention, the phase of the measured mains voltage vector is detected in order to derive each mains phase current from a corrected current that represents the output current of the rectifier stage. The phase angle of the rotating mains voltage vector indicates which of the phases of the passive rectifier stage are conductive and which are non-conductive. This is a simple way to distribute the corrected current, which represents the output current of the rectifier stage, across the mains phases.

[0030] In one embodiment of the invention, information derived from the calculated current or from the corrected current and a mains voltage is used to calculate at least one of the following values: imbalances, voltage distortions, RMS values, and THC values. In many cases, at least one of these data points is of interest to the user of a frequency converter. In many cases, the user is interested in more than one piece of information.

[0031] In one embodiment of the invention, information derived from the corrected current, or from the corrected current and a mains voltage, is used for the protection and / or service life estimation of the frequency converter or at least one component of the frequency converter. This allows for reliable planning of maintenance and repair of the frequency converter.

[0032] An embodiment of the invention will now be described in more detail with reference to the drawing, wherein: The attached figure shows a schematic representation of a circuit diagram of a three-phase frequency converter.

[0033] A frequency converter 1 is connected to a network 6 with three phases u, v and w. The network has a network impedance 2 per phase. Alternatively, the network impedance block 2 could represent an integrated AC voltage inductor of the frequency converter block 1.

[0034] To facilitate the following explanation, the circuit diagram is divided into a mains side 6, a rectifier stage 7, an inverter stage 8 and a motor side 9.

[0035] The rectifier stage 7 includes a passive rectifier 3. The passive rectifier 3 is formed from rectifier silicon devices, such as diodes D1-D6, which are generally known in the art. Alternatively, in some frequency converters, the diodes can also be thyristors for inrush current limiting. The rectifier 3 shown is a three-phase rectifier, but it can also be a single-phase rectifier or another multi-phase rectifier. An intermediate circuit inductor Ldc is connected to one output side of the rectifier 3. Alternatively, two DC chokes could be used, one on the low-side and one on the high-side rail. Two DC chokes increase the complexity of the voltage sensor device, which is irrelevant for the present invention.A DC link capacitor Cdc is connected between a positive rail 10 and a negative rail 11 of the DC link 5 downstream of the DC link inductor Ldc. The DC capacitor can physically be used as several capacitors in series in a frequency converter, which is irrelevant for the present invention.

[0036] Inverter 4 has six controlled switches T1-T6, with each of the switches T1-T6 being assigned an antiparallel diode.

[0037] The inverter 4 has three outputs 12, 13, 14, each of which is connected to a point between two switches T1, T4; T2, T5; T3, T6, as is known in the prior art. The inverter 4 can have more or fewer than three phase outputs. Means for controlling the switches are not shown because this is irrelevant to the present invention.

[0038] The three phases u, v, w of network 2 supply a current iu, iv, iw and a voltage Vu, Vv, Vw to points between each diode pair D1, D4; D2, D5; D3, D6. Voltages and currents are rectified in rectifier 3, so that the output of rectifier 3 yields a pulsating DC output voltage Urec, as is known in the prior art. This output voltage Urec can be easily measured. It can also be calculated from the instantaneous maximum value minus the instantaneous minimum value of the voltage differences between the voltage potentials Vu, Vv, Vw on the network side 6 of rectifier 3, if these are measured. This results in a complex sampling circuit compared to measuring Urec.

[0039] Rectifier stage 7 outputs a direct current iLdc, which flows through the intermediate circuit inductor Ldc. The intermediate circuit inductor Ldc, together with an intermediate circuit capacitor Cdc, smooths the current iLdc. However, a ripple remains, which flows into the capacitor Cdc along with the ripple current induced by the inverter stage, which is associated with the current idc. The remainder of the inductor current, which is essentially its average value, is fed to inverter stage 4 via the positive rail 10 and returned to rectifier stage 3 via the negative rail 11. At any given time, the current iLdc equals the sum of the currents iCdc and idc shown in the figure, according to Kirchhoff's law. In steady state, the current iCdc averages zero.

[0040] The circuit diagram shows a current idc through the positive rail 10 of the intermediate circuit 5, currents ibu, ibv, ibw in each of the inverter legs, currents imu, imv, imw in each of the outputs of the inverter stage 4 and a current iCdc through the intermediate circuit capacitor Cdc.

[0041] Each of these currents can be measured by appropriate current sensors 15-22, whereby the current idc can be measured through the intermediate circuit in the positive rail (idcH) or in the negative rail (idcL).

[0042] However, it should be noted that not all current sensors 15-22 are required in a minimum configuration. Only one current sensor 15 or 16 in the intermediate circuit is sufficient, or current sensors 17-19 in the legs of inverter stage 4, or two of the current sensors 20-22 in the outputs 12-14 of inverter stage 4 would be sufficient.

[0043] If the voltage measurements of the voltages Urec, Udc and the DC link component model are sufficiently precise, the current sensors 15-22 can be omitted in accordance with the present invention. However, it is rather difficult to perform a voltage measurement with the required accuracy.

[0044] One way to determine the grid-side currents iu, iv, iw on the grid side 2 of the DC-DC converter 1 is to calculate the current iLdc through the DC link inductance. This can be easily achieved by calculating the difference between the voltage Urec at the output of rectifier stage 3 and the voltage Udc across the DC link capacitor. This difference is the voltage drop across the DC link inductor Ldc. By integrating this voltage drop over time and dividing it by the inductance of the DC link inductor Ldc, it is theoretically possible to calculate the current drawn by the converter stage 3. However, in many cases, the measurement or determination of the voltages Urec and Udc is not precise enough, so such integration would lead to drift, or in extreme cases, to positive or negative infinity. Furthermore, the characteristics of components may be inaccurate.To avoid this problem, an average value is calculated from the calculated current. This average value is subtracted from the calculated current, thus eliminating drift and retaining only information about the current's ripple. To obtain the true current information for rectifier stage 3, the "corrected current," the current idc across the DC link 5 is measured by one or more of the current sensors 15-22. The average of the measured current is added to the difference between the calculated current and the average of the calculated current. Alternatively, a fraction of the difference between the two averages is used to obtain the corrected current. Experts will recognize that this corresponds to a proportional controller with a gain Kp less than 1. Similarly, in the first case, the gain Kp of the proportional controller is 1.The compensation equation can thus be described as follows: ILdc_corrected = ILdc + Kp*(Idc_mean - ILdc_mean). Using a Kp value less than one introduces an error in the DC component of the corrected current if the average value of the calculated current is not sufficiently accurate. Therefore, the term "fraction" in the present invention can alternatively refer to the use of a controller with an integration function, such as a PI controller or PID controller, etc. In this case, the integration function of the controller used eliminates the introduced error when a Kp value less than one is used in the equation above.

[0045] Alternatively or additionally, the current can also be calculated using the current ICdc across the DC link capacitor Cdc. For this purpose, the voltage Udc across the DC link capacitor Cdc can be differentiated to calculate the current iCdc through the DC link capacitor Cdc. In this case, the average value of the measured DC link current idc is added to this calculated current iCdc to obtain the "true" value of the current to rectifier stage 3.

[0046] It should be noted that the inventive method can also be used if the DC link inductor Ldc has a series resistance or is connected in series with a resistor. In this case, the voltage drop across the resistor must be subtracted from the measured voltage drop across the inductor before the integration is started. Accordingly, it is possible to consider the DC link inductor Ldc as a combination of an inductor and a resistance.

[0047] Furthermore, it is assumed that the mean value of the current iCdc is zero, which is justified in steady-state operation. However, dynamic simulations show that the method can also be applied with good accuracy when the mean value of the current iCdc through the intermediate circuit capacitor Cdc is not zero during transient sequences. Therefore, it is not necessary to complicate the model, even though this is covered by the invention, since both Ldc and Cdc can be used simultaneously in a dynamic model to improve the corrected current transiently.

[0048] Based on the above-corrected current ILdc, the mains phase currents can be calculated by detecting the zero crossings or the phase of the mains voltages.

[0049] Alternatively or equivalently to the above DC inductor approach, the line currents iu, iv, and iw can be calculated directly by using the characteristics of a line-side AC inductor in the figure together with the measured voltage drops across it. Here too, drift can occur, which is compensated equivalently with the mean value of the current DC. Since each line phase occurs in 2 / 3 of the time, the mean value of the DC used in this case must be scaled accordingly.

[0050] If the calculated current and / or the absolute value of the calculated grid-side current each contain an AC component and a DC component, each corrected current is determined by separating the calculated AC and DC components and then adjusting the calculated DC component by a measured current or a fraction of a measured current from the inverter side of the frequency converter.

[0051] In some cases, the calculated current(s) contain only an AC component. This means that the corrected current(s) are obtained by adding a DC component to a measured current from the inverter side of the frequency to the calculated current(s).

[0052] It is also possible to calculate the current by using the properties of the DC inductor and the measured voltages across the inductor. One property of the DC inductor can be its inductance. Another example is the inductance of the inductor in combination with a series and / or parallel resistor.

[0053] Another possibility is to obtain the calculated current by using the properties of the DC capacitor and the measured voltages across the capacitor. One property of the DC capacitor can be its capacitance alone or in combination with a series and / or parallel resistor.

[0054] Another way to calculate the current is to use the properties of the AC inductor and the measured voltages across it. The properties of the AC inductor can be its inductance or its inductance in combination with a series and / or parallel resistor.

[0055] The calculated current can also be obtained by using the characteristics of the DC capacitor and DC inductor, as well as measured voltages across these components and / or the characteristics of another component on the rectifier side of the frequency converter. The component characteristics can be the same as described above.

[0056] The calculated current can be obtained by using the characteristics of the DC capacitor, the DC inductor, and / or the AC inductor, as well as measured voltages across these components and / or the characteristics of another component in the rectifier stage of the frequency converter. These characteristics can be the same as described above.

[0057] Furthermore, the calculated current can be adjusted by using the DC components or a fraction of the DC components of the measured current from the inverter side of the frequency converter.

[0058] The calculated current per network phase can also be derived by integrating the voltage drop across the AC inductor in each phase. A voltage drop across the AC inductor per phase can be determined by measurement. The product of the inverse inductance of the AC inductor and the integral of the voltage over time yields the calculated current. This is sufficient in some cases to obtain a corrected current. However, drift can occur during integration if model inaccuracies or errors in voltage measurement are too large. In this case, it is advantageous to correct the calculated current with a fraction of the measured current. Since the calculated current in this embodiment is an AC current, the absolute value of the calculated current should be used for the correction step, along with the measured DC current from the inverter side of the frequency converter.

[0059] Furthermore, it is possible to determine the phase of the measured mains voltage vector in order to derive each mains phase current from a corrected current representing the output current in the rectifier stage. The phase angle of the rotating mains voltage vector indicates which of the phases of the passive rectifier stage are conducting and which are non-conducting. This is a simple way to distribute the corrected current, which represents the output current of the rectifier stage, across the mains phases.

Claims

[1] Method for determining the output current of a rectifier stage and / or mains-side currents (iu, iv, iw) of a frequency converter (1) with a passive rectifier (3), an inverter (4), a DC link with an DC link inductor (Ldc) and an DC link capacitor (Cdc) between the rectifier (3) and the inverter (4) and / or an AC inductor on the mains side of the rectifier stage, characterized by, that the method comprises the step of calculating a current in the DC link (5) and / or of mains-side currents by using at least one voltage value and characteristics of the rectifier stage and / or DC link to form a corrected current using the calculated current and a measured current or currents, or a fraction of a measured current or currents, wherein information obtained from the corrected current or from the corrected current and a mains-side voltage is used to estimate the lifetime of a component of the frequency converter. [2] Method according to claim 1, characterized by, that the calculated current and / or the absolute value of the calculated grid-side currents each contain an AC component and a DC component, wherein each corrected current is determined by separating the calculated AC and DC components and then adjusting the calculated DC component by a measured current or a fraction of a measured current from the inverter side of the frequency converter. [3] Method according to claim 1 or 2, characterized by , that the calculated current and / or currents contain only an AC component, which means that the corrected current and / or currents are obtained by adding a DC component of a measured current from the inverter side of the frequency converter to the calculated current and / or currents. [4] Method according to any one of claims 1 to 3, characterized by, that the calculated current is obtained by using properties of the DC inductor and measured voltages across the inductor. [5] Method according to any one of claims 1 to 4, characterized by , that the calculated current is obtained by using the properties of the DC capacitor and the voltages measured across the capacitor. [6] Method according to any one of claims 1 to 5, characterized by , that the calculated value is obtained using properties of the AC inductor and measured voltages across the inductor. [7] Method according to any one of claims 1 to 6, characterized by , that the calculated current is obtained using the properties of the DC capacitor and the properties of the DC inductor and the measured voltages across the component and / or the properties of any other component on the rectifier side of the frequency converter. [8] Method according to any one of claims 1 to 7, characterized by , that the calculated current is obtained by using both the properties of the DC capacitor and the properties of the DC inductor and / or the properties of the AC inductor and measured voltages across the components and / or properties of any other component in the rectifier stage of the frequency converter. [9] Method according to any one of claims 1 to 8, characterized by , that a direct current (idc) is measured and used in combination with the calculated current, the combination of the measured current (idc) and the calculated current forming the corrected current. [10] Method according to claim 9, characterized by , that the DC component or a fraction of the DC component of the measured current is used by the rectifier side of the frequency converter to adjust a calculated current. [11] Method according to claim 9 or 10, characterized by , that the direct current is measured in one of the two inverter-side rails of the DC intermediate circuit (5), at least two of the legs of the inverter stage and / or at least two of the outputs (12-14) of the inverter stage (4). [12] Method according to one of claims 9 or 11, characterized by , that an average value of the DC component of a calculated current is determined. [13] Method according to any one of claims 9 to 12, characterized by , that a mean value or a DC component or a fraction thereof of a calculated current is derived, the mean value or a DC component or a fraction thereof is subtracted from the calculated value, and a mean value or a DC component or a fraction thereof of the measured DC current (idc) is added to the difference to form a corrected current. [14] Method according to any one of claims 1 to 13, characterized by , that the calculated current is derived by integrating a voltage drop across the intermediate circuit inductance (Ldc). [15] Method according to claim 14, characterized by , that the voltage drop is calculated from a difference between the output voltage (Urec) of the rectifier stage (3) and the intermediate circuit voltage (Udc). [16] Method according to claim 15, characterized by , that the output voltage (Urec) of the rectifier stage (3) is calculated from an instantaneous maximum value minus an instantaneous minimum value of the mains voltages (Vu, Vv, Vw) at the input side of the rectifier stage (3). [17] Method according to claim 16, characterized by , that a calculated current per network phase is derived by integrating the voltage drop across the AC inductor in each network phase. [18] Method according to any one of claims 2 to 17, characterized by, that the calculated current is derived by differentiating the voltage (Udc) across the intermediate circuit capacitor (Cdc), and the calculated current and an average of the measured current (idc) are added to form the corrected current. [19] Method according to any one of claims 1 to 18, characterized by , that the zero crossings of the mains voltages (Vu-Vv, VV-Vw, Vw-Vu) are used to derive each mains phase current (iu, iv, iw) from a corrected current that represents the output of the rectifier stage (3). [20] Method according to any one of claims 1 to 19, characterized by , that the phase of the measured mains voltage vector is captured in order to derive each mains phase current from a corrected current representing the output current of the rectifier stage. [21] Method according to any one of claims 1 to 20, characterized by, that information obtained from the calculated current or from the corrected current and mains voltages is used to calculate at least one of imbalances, voltage distortions, RMS values ​​and THC values.

Images