DE102018222566A1 - Method for recognizing an error state of an electrical machine - Google Patents

Abstract

The invention relates to a method for detecting a fault state (100A) in an excitation circuit (121) of an electrical machine (100), further comprising a stator (110) and a rectifier circuit (130) connected to the stator (110), a time profile of an excitation current (I) of the electrical machine (100) is detected (310), characterized in that a first statistical variable (Q1) of the excitation current (I) is determined and a further statistical quantity (Q2) of the excitation current (I) is determined , wherein an error (F) in the excitation circuit (121) is inferred (320) if the difference (D) between the first statistical variable (Q1) and the further statistical variable (Q2) exceeds a threshold value (S).

Description

The present invention relates to a method for recognizing an error state of an electrical machine and a computing unit and a computer program for carrying it out.

State of the art

Different types of power generators can be used to supply networks or load circuits. For example, a three-phase current can be generated by means of three-phase generators. To supply DC networks from such three-phase generators, converters operated as rectifiers can be used to convert a multi-phase current generated by the three-phase source into direct current. The rectification can be carried out using passive (diodes) or active (semiconductor switches) rectifier elements. In the case of an active rectifier, a corresponding control circuit is also part of the generator controller in addition to the field controller. Three-phase generators can often be implemented as electrical machines, which can be operated as a generator to generate electrical energy or as a motor to convert electrical energy into mechanical energy.

For example, power generators of this type can be used in motor vehicles to supply a motor vehicle electrical system. A corresponding electrical machine can, for example, be operated as a generator in order to supply the motor vehicle electrical system or to charge a motor vehicle battery. For this purpose, the electrical machine can be connected to or disconnected from the vehicle electrical system via so-called output stages or output stage switching.

Disclosure of the invention

According to the invention, a method for recognizing an error state in an excitation circuit, preferably a rotor of an electrical machine, and a computing unit and a computer program for carrying it out are proposed with the features of the independent claims. Advantageous embodiments are the subject of the subclaims and the following description.

The electrical machine can in particular be designed as a generator, e.g. as a claw pool or leg pool generator. Furthermore, the electrical machine can also be designed as a motor and / or generator-operated electrical machine with the geometries mentioned above. The electrical machine has, in particular, a rotor and a stator and a rectifier circuit connected to the stator for rectifying an AC voltage applied to the stator. The rectifier circuit can in particular have bridge circuits composed of passive switching elements, in particular diodes, or else active switching elements, in particular semiconductor switches such as metal oxide semiconductor field effect transistors (MOSFET).

The invention provides a possibility of easily recognizing an error state in the excitation circuit of the rotor. A classification according to the scope and progress of the aforementioned error is also given.

In order to achieve a generator voltage, the electrical machine is controlled in a clocked manner by a computing unit by means of a control signal.

As part of the method for detecting an error state in an excitation circuit of the rotor, in particular an impending short circuit in the excitation circuit, a time profile of an excitation current of the electrical machine is recorded. Following this, a first statistical quantity of the excitation current is determined and a further statistical quantity of the excitation current is determined, an error in the excitation circuit being concluded if the difference between the first statistical quantity and the further statistical quantity exceeds a threshold value. In the present case, the first statistical quantity is preferably a first quantile of the excitation current and the further statistical quantity is preferably a further quantile of the excitation current.

Such a statistical evaluation of the excitation current values in order to check whether there is a fault, in particular a short circuit in the excitation circuit, can be implemented particularly simply and robustly in the sense of a safety of an error determination. By selecting a suitable quantile as the lower limit and a further suitable quantile as the upper limit, the noise contained in the excitation current signal can be eliminated in particular. In addition, the deficiency of the aforementioned quantile provides a good measure of the aforementioned error.

The error can be determined particularly easily by comparing the difference between the first quantile and the further quantile with a threshold value. The threshold value, on the other hand, results from a reference curve of the excitation current in an electrical machine in the defect-free state. Depending on the operating point of the electrical machine, the respective threshold values can be stored in a control device for the corresponding use. In a further preferred embodiment According to the invention, the first quantile lies in the interval between 1% and 10%, preferably 2%, and the further quantile lies in the interval between 90% and 99%, preferably 98%.

Using this preferred embodiment of the invention, the quantiles can be specified specifically for a corresponding electrical machine in such a way that the noise component in the excitation current signal can be reliably excluded. On the basis of such a choice of the quantiles, which is preferably 2% as the lower limit and 98% as the upper limit, the method according to the invention can diagnose a short circuit in the excitation circuit particularly accurately. The 2% quantile is given by the fact that 2% of the measured values determined, which were determined in particular within a defined time interval, lie below or correspond to the value of the quantile. As a result, 98% of the determined values, in this case the excitation current, are above this value.

In a further preferred embodiment of the invention, the excitation current or the first quantile and / or the further quantile is determined within a time interval, preferably less than or equal to 100 ms, further preferably less than or equal to 10 ms.

To determine the excitation current and from this the first quantile and / or the further quantile, only very small time intervals are required for reliable fault detection in the excitation circuit in order to carry out a corresponding diagnosis. Only a few characteristic features in the excitation current, which are given by the essentially periodic voltage fluctuations, are sufficient for this.

In a further preferred embodiment of the invention, the threshold value is determined on the basis of a step current in the excitation circuit. The determination of the threshold value on the basis of the step current of an excitation circuit within an electrical machine is advantageous since this step current can easily be determined using an analytical formula.

In a further preferred embodiment of the invention, the magnitude of the difference between the first statistical variable and / or the further statistical variable, preferably the first quantile and / or the further quantile, determines the extent and / or progress of the error, in particular a short circuit in the excitation circuit, closed. Through a suitable choice of the first quantile and the further quantile, a defect within the excitation circuit can be inferred particularly reliably on the basis of their difference and comparison thereof with a threshold value. The severity and / or the progress of the error within the excitation circuit can be derived from the amount of the difference. The larger the amount, the more serious or advanced the error. This can also be classified with corresponding amount limits, whereby a distinction is made between moderate, severe and very serious errors and appropriate measures, such as a warning message or an emergency shutdown can occur.

A computing unit according to the invention, e.g. a control unit of a motor vehicle is, in particular set up in terms of programming, to carry out a method according to the invention.

The implementation of the method in the form of a computer program is also advantageous, since this causes particularly low costs, in particular when an executing control device is still used for further tasks and is therefore already present anyway. Suitable data carriers for providing the computer program are in particular magnetic, optical and electrical memories, such as Hard disks, flash memory, EPROMs, DVDs etc. can also be downloaded from a program via computer networks (Internet, intranet, etc.).

Further advantages and refinements of the invention result from the description and the accompanying drawings.

The invention is illustrated schematically in the drawing using exemplary embodiments and is described below with reference to these drawings.

Figure list

• 1 schematically shows an electrical machine with a preferred embodiment of a computing unit according to the invention, which is set up to carry out a preferred embodiment of a method according to the invention;
• 2nd schematically shows a voltage-time diagram of a time profile of a current value, which can be determined in the course of a preferred embodiment of a method according to the invention;
• 3rd schematically shows an electrical machine in a fault state;
• 4a-b show the course of the excitation current without short circuit (a) and with short circuit (b) in the excitation circuit; and
• 5 shows schematically a flow diagram of a method for error detection.

In 1 is an electrical machine in the form of a generator shown schematically and designated 100.

The electrical machine 100 is designed in this example as a three-phase electrical machine, with stator inductances (phases) of a stator 110 are connected to a delta connection. A rotor 120 has an excitation winding 121 with a diode connected in parallel. An excitation transistor can also be used in an excitation circuit 122 be provided. By switching the excitation transistor on and off 122 , usually by means of PWM operation, is applied to the excitation winding 121 a voltage (here the rectified generator voltage) is applied intermittently, whereupon an excitation current is established. By changing the duty cycle, also called the control rate, of the PWM operation, in particular the level of the excitation current and thus the level of the generator voltage can be changed.

The electrical machine 100 still has one with the stator 110 connected rectifier circuit 130 with three half bridges for rectifying one on the stator 110 applied three-phase AC voltage. Each half bridge has a center tap between its two rectifying elements, which are designed here as diodes, via which the respective half bridge with a phase connection of the stator 110 connected is.

Between two DC voltage connections 140 the rectifier circuit 130 a DC voltage UB + is provided as a rectified generator voltage. The electrical machine 100 can be used, for example, in a motor vehicle to supply a motor vehicle electrical system that is connected to the DC voltage connections 140 connected is.

A computing unit 150 is to control the electrical machine 100 intended. For example, the computing unit 150 be designed as a control unit of the corresponding motor vehicle. The computing unit 150 is set up to detect error states of the electrical machine. For this purpose, the computing unit 150 , in particular in terms of programming, set up to carry out a preferred embodiment of a method according to the invention.

In 2nd is the time course in a voltage-time diagram 200 the DC voltage UB + when the electrical machine is in perfect condition 100 shown schematically.

In 3rd is the in 1 electrical machine already described, but with an error F , in the form of a short circuit 100A in the excitation circle 121 , shown. The short circuit 100A or errors F is the shape that the leader in the excitation circle 121 with a resistance R ₁ from a few mΩ another conductor with resistance R _K is connected in parallel, the resistance of which, depending on the type of short circuit, approaches 0 Ω.

When the excitation circuit is short-circuited 121 , additional current flows on the short circuit branch with resistance R _K when the exciter transistor 122 from the controller 150 is switched on. Will the transistor 122 switched off, a current flows back through the diode due to the inductance, but only a small current flows through the short-circuit branch. That is why there is a periodic jump in current I _Sp in the excitation current course I _Err open if there is an excitation circuit short circuit. This behavior is in the excitation current curve I _Err Particularly easy to observe when the generator is not fully loaded. The theoretical jump in current in this case can be calculated using equation (1): $${\text{I.}}_{\text{Sp}}\approx {\text{U}}_{\text{Err}}/{\text{R}}_{\text{K}}$$

U _Err is the excitation voltage and R _k is the resistance in the short circuit branch. This can cause the excitation circuit short circuit with the jump in current I _Sp be localized. The strength, i.e. the progress and / or the extent of the error, is shown in the magnitude of the current jump I _Sp .

The corresponding courses of the excitation current I _Err in the excitation circle 121 are for the short circuit 100A in 4b shown. 4a shows a typical course of the excitation current I _ErrRef without short circuit in the excitation circuit 121 .

In addition, the threshold is S , which here corresponds to approximately 1 A, and the first quantile Q1 as a minimum and the further quanile Q2 given as maximum. The first quantile Q1 corresponds approximately to the 2% quantile and the further quantile in the present case Q2 corresponds approximately to the 98% quantile in the present case. Now is the difference D from the quantile Q2 and the quantile Q1 in terms of amount greater than the threshold S , there is an error. Is the difference in amount of the first quantile Q1 and the further quantile Q2 ≤ the threshold value (cf. 4a) this is the pathogen 121 the electrical machine is intact and there is no fault. Even very small time intervals Δt of approx. 10 ms are sufficient to use the difference D the quantile Q1 and Q2 certainly on a fault in the excitation circuit 121 close. The corresponding procedure for determining the error is in 5 explained in more detail.

In 5 is a schematic flow diagram of a method according to the invention for detecting the in the 3rd and 4th shown fault condition 100A represented in the form of a short circuit. The process starts with an initialization step 200 . In step 202 the excitation current is determined, preferably within the time interval Δt. In step 204 becomes the first quantile Q1 or the first statistical variable is determined. In step 206 becomes the second statistical quantity, preferably the further quantile Q2 determined. In step 208 we the absolute difference of the first statistical quantity or the first quantile Q1 and the further statistical quantity, preferably the further quantile Q2 determined. In step 210 is the absolute difference of the quantiles | Q2-Q1 | with a threshold S compared. If the difference in amount of the quantiles is greater than or equal to the threshold value (shown in 5 with a +) in step 214 detected an error.

Optionally, in step 216 a corresponding error message in the form of a flag is stored in the control unit of the electrical machine or a warning message or further measures for an emergency operation of the electrical machine 100 be initiated. The process then follows in step 218 completed. Now in step 210 found that the difference in amount of the first quantile Q1 and the further quantile Q2 not greater than or equal to the threshold S (shown with a -) is in step 212 recognized that the electrical machine, especially the excitation circuit 120 the electrical machine is intact. The process then follows in step 218 also ended.

Claims (9)

Method for recognizing an error state (100A) in an excitation circuit (121) of an electrical machine (100), further comprising a stator (110) and a rectifier circuit (130) connected to the stator (110), wherein an excitation current (I _Err ) of the electrical machine (100) is recorded (202), characterized in that a first statistical quantity (Q1) of the excitation current (I _Err ) is determined (204) and a further statistical quantity (Q2) of the excitation current (I _Err ) is determined (206), an error (F) in the excitation circuit (121) being concluded (210) if the difference (D) between the first statistical variable (Q1) and the further statistical variable (Q2) has a threshold value (S) exceeds. Procedure according to Claim 1 , wherein the first statistical quantity is a first quantile (Q1) of the excitation current (I _Err ) and the further statistical quantity is a further quantile (Q2) of the excitation current (I _Err ). Procedure according to Claim 2 , wherein the first quantile (Q1) in the interval between 1% and 10%, preferably 2%, and the further quantile (Q2) in the interval between 90% and 99%, preferably 98%. Method according to one of the preceding claims, wherein the first statistical variable (Q1) and the further statistical variable (Q2) are determined within a time interval (Δt), preferably less than or equal to 100 ms, further preferably less than or equal to 10 ms. Method according to one of the preceding claims, wherein the threshold value (S) is the step current (I _S ) in the excitation circuit (121). Method according to one of the preceding claims, wherein the amount of the difference between the first statistical variable (Q1) and the further statistical variable (Q2) indicates the scope and / or progress of the error (F), in particular a short circuit in the excitation circuit (121) becomes. Computing unit (150) which is set up by hardware and / or by software to carry out a method according to one of the preceding claims. Computer program which causes a computing unit (150) to carry out a method according to one of the Claims 1 to 7 to be carried out when it is executed on the computing unit (150). Machine-readable storage medium with a computer program stored on it Claim 9 .

Images