Device and method for condition monitoring of an electric motor pump

Abstract

Device (1) for monitoring the condition of an electric motor pump (2), preferably arranged in an aircraft, comprising: at least a first sensor (3) for detecting an actuating current (IRMS) of the electric motor pump (2), at least a second sensor (4) for detecting a rotational speed (n) of the electric motor pump (2), a computer unit for processing and evaluating the data of the electric motor pump (2) detected by the first sensor (3) and the second sensor (4), and a storage unit for storing the values ​​for the actuating current (IRMS) and the rotational speed (n) of the electric motor pump (2) detected by the first sensor (3) and the second sensor (4), wherein the computer unit is designed to analyze the values ​​for the actuating current (IRMS) and the rotational speed (n) stored in the storage unit and, on the basis thereof, to detect or predict a fault of the electric motor pump (2), characterized in thatthat the electric motor pump (2) is based on the principle of variable motor control with a constant displacement pump (8), in which a motor control unit (10) is provided to control a motor (9), in particular a permanent magnet synchronous motor, connected to the constant displacement pump (8), and the computer unit is further designed to assign a detected or predicted fault to the motor (9) or the constant displacement pump (8), wherein for this purpose the computer unit is designed to: determine the output power of the motor control unit (10), determine the input power, i.e. the shaft power of the motor (9), determine the output power of the displacement pump (8), calculate the efficiency of the subcomponents motor (9) and displacement pump (8) from the power ratios in order to assign a fault to the motor (9) or the displacement pump (8).

Description

The present invention relates to a device for monitoring the condition of an electric motor pump and a corresponding method. Hydraulic pumps are used in a wide variety of technical fields, but it is always advantageous to identify any malfunctions or impending problems as early as possible. For example, when using a hydraulic pump in an aircraft, it is desirable to carry out any upcoming maintenance work to repair or replace components of the hydraulic pump well in advance. This prevents costly, unplanned malfunctions that could affect flight operations or even lead to the aircraft being out of service for an extended period. Conventional electric motor pumps typically use an asynchronous motor that operates at a constant speed, dependent on the mains frequency, and drives a variable displacement pump. The pumping volume is then simply varied using the variable displacement pump. A disadvantage of this design is that condition monitoring of such an electric motor pump is not possible, as it is connected directly to the power source, e.g., the electrical system of an aircraft, without an intermediate control unit. A device for monitoring the condition of an electric motor pump is already known from EP 1 564 411 A1. Further prior art can also be found in WO 2018 / 197 033 A1 and DE 10 2014 224 261 A1. The object of the present invention is to provide a device for monitoring the condition of an electric motor pump in order to detect or predict a fault. Advantageously, this should make it possible to obtain a forecast of when a necessary maintenance intervention is required, so that it can be planned and carried out as early as possible. This is achieved with a device having all the features of claim 1 or with a method according to claim 11. Further advantageous embodiments of the present invention are specified in the independent claims. The device according to the invention for monitoring the condition of an electric motor pump, which is preferably arranged in an aircraft, comprises at least one first sensor for detecting a control current of the electric motor pump, at least one speed detection unit for detecting a speed of the electric motor pump, a computer unit for processing and evaluating the data of the electric motor pump detected by the first sensor and the speed detection unit, and a storage unit for storing the values ​​for the control current and the speed of the electric motor pump detected by the first sensor and the speed detection unit, wherein the computer unit is designed to analyze the values ​​for the control current and the speed stored in the storage unit and, on the basis thereof, to detect or predict a fault of the electric motor pump. A correlation of motor parameters with system or control variables enables condition monitoring of the electric motor pump, which is easy to implement. Over time, both the actuator current and the speed of the electric motor pump are recorded and compared with historical data to determine whether the required actuator current varies to achieve a predetermined speed. If the actuator current required to reach a specific speed increases over time, this is a reliable indicator of increasing mechanical friction losses, caused, for example, by faulty motor bearings or mechanical defects in the pump bearings. This enables a mechanical wear monitor or condition assessment that provides an indication of a potentially imminent maintenance operation or an immediate repair. According to an optional modification of the present invention, the computing unit can be designed to analyze the values ​​for the actuating current and the speed stored in the memory unit and to detect a fault of the electric motor pump if an increase in the actuating current over time at a constant speed of the electric motor pump exceeds or falls below a threshold value, preferably by detecting the increase or decrease based on the operating points considered over time, which consist of a respective data pair of actuating current and associated speed of the electric motor pump at a specific time. Advantageously, according to the invention, the computing unit can also be designed to analyze the values ​​for the actuating current and the speed stored in the memory unit and to predict a fault of the electric motor pump if a tendency of an increase or decrease of the actuating current over time at a constant speed of the electric motor pump indicates that a threshold value is being exceeded or fallen below, preferably by predicting the tendency by means of an extrapolation of the operating points considered over time, which consist of a respective data pair of actuating current and associated speed of the electric motor pump at a specific time. If a trend is discernible in the stored data pairs of control current and speed, indicating that an increasingly higher / lower control current is required to achieve a certain speed, an extrapolation can be used to determine when the threshold value at which maintenance or repair of the electric motor pump is deemed necessary will actually be exceeded / fall below. This involves using several time-spaced data pairs of control current and speed from the past (which together represent an operating point) and projecting when the threshold value is expected to be exceeded / fall below. According to a further development of the present invention, it can be provided that the computer unit is designed to use only those values ​​of the control current and the speed of the electric motor pump for fault detection or prognosis that have been generated in a quasi-stationary operation of the electric motor pump, preferably wherein a quasi-stationary operation is defined by keeping a speed range or a speed of the electric motor pump constant over a predetermined time in which no dynamic load change takes place. It is advantageous to store only those related data pairs of control current and speed measured in a quasi-steady-state condition of the electric motor pump in the storage unit. In quasi-steady-state operation, there are no dynamic load changes, so any increase in the control current at a specific torque value cannot be attributed to a sudden increase in the pump's load. Therefore, in quasi-steady-state operation, an increase in the control current at a specific speed can only be attributed to increased wear of the electric motor pump. According to a further optional modification of the present invention, the state device may further comprise at least one third sensor for detecting the temperature of the fluid pumped by the electric motor pump, wherein the storage unit is further configured to store the values ​​for the temperature of the pumped fluid detected by the third sensor, and the computing unit is further configured to take the temperature of the pumped fluid into account during fault detection or prediction, wherein preferably the threshold value used in fault detection or prediction and / or the operating point determined by the actuating current and speed is raised or lowered depending on the detected temperature of the pumped fluid if it deviates from a regular temperature. As explained above, since an operating point—that is, the required control current for a specific rotational speed—depends on the temperature of the fluid being pumped (usually hydraulic oil), it is advantageous to also take this effect into account. For this purpose, the temperature of the fluid being pumped is determined, and the measured value is considered in the fault detection or fault prediction. If the temperature of a fluid being pumped, especially a hydraulic oil, is very low, the required control current to achieve a predetermined torque increases because a cool hydraulic oil is typically more viscous. Conversely, at an increased temperature, the viscosity of the hydraulic oil improves, so that less force and therefore a lower control current are required to achieve a predetermined torque value.Since no conclusions can be drawn from this regarding any potential wear-related deterioration of the electric motor pump, it is advantageous to factor this effect out. This is done by adjusting the operating point, which is based solely on the rotational speed and the control current, according to the temperature of the fluid being pumped. Advantageously, according to the invention, the electric motor pump can be based on the principle of variable motor control with a constant displacement pump, in which a motor control unit is provided to control a motor connected to the constant displacement pump, in particular a permanent magnet synchronous motor. Unlike previous pumping systems, the rotational speed of the pump shaft is no longer kept constant and the volume of swallowed is varied; instead, the volume of swallowed is kept constant and the rotational speed of the pump shaft is varied. It may be provided that the first sensor and the speed detection unit are integrated into a motor control unit that knows the speed and the actuating current of the electric motor. This is advantageous because no additional sensors are required to detect the speed and current of the electric motor, which is connected to the pump shaft via its output shaft. According to an optional modification of the present invention, the computing unit may further be designed to assign a detected or predicted fault to the motor or the constant displacement pump, wherein for this purpose the computing unit is designed to determine the output power of the motor control unit, to determine the input power, i.e. the shaft power of the motor, to determine the output power of the displacement pump, and to calculate the efficiency of the subcomponents motor and displacement pump from the power quotients in order to assign a fault to the motor or the displacement pump. The resulting ability to identify a detected or predicted fault facilitates the planning of maintenance operations, as appropriate spare parts for the motor or the displacement pump can already be kept on hand. Preferably, the output power of the motor control unit is determined by calculating: 3 / 2*(Vq * Iq + Vd * Id), the shaft power is determined by calculating: Mth* n, and the output power of the positive displacement pump is determined by calculating: P* ΔP *k, where Vq represents the voltage vector in the d / q rotor coordinate system for the q-axis, Vd represents the voltage vector in the d / q rotor coordinate system for the d-axis, Iq represents the current vector in the d / q rotor coordinate system for the q-axis, Id represents the current vector in the d / q rotor coordinate system for the d-axis, Mth represents the theoretical hydraulic pump torque, n represents the rotational speed, P represents the power at the pump shaft, and ΔP represents the hydraulic differential pressure. According to a further development of the present invention, the computer unit can also be designed to infer a fault in a demagnetized motor by determining the torque constant kT of the motor from the hydraulic data on the one hand and from the voltage and speed data on the other, and comparing them, preferably in the case of a deviation of the torque constant kT determined in different ways exceeding a threshold value, a fault in the demagnetized motor is inferred. It can be provided that the torque constant kT of the motor is calculated from: M = 3 / 2 * kT* Iq, and the torque constant kT is calculated from the voltage and speed data using: Vq = kT * n, where M stands for the torque, kT for the torque constant, Iq for the current phasor in the d / q rotor coordinate system for the q-axis, Vq for the voltage phasor in the d / q rotor coordinate system for the q-axis, and n for the speed. The invention further relates to a system with an electric motor pump and a device for condition monitoring according to one of the preceding variants, preferably wherein the condition monitoring is carried out by a correlation of motor parameters in order to predict or detect a fault case on the basis of system or control variables of the electric motor pump. The invention further relates to a method for monitoring the condition of an electric motor pump, preferably an electric motor pump based on the principle of variable motor control with a constant displacement pump, in which a motor control unit is provided to control a motor connected to the constant displacement pump, in particular a permanent magnet synchronous motor, wherein the method comprises the steps of: detecting a control current of the electric motor pump, detecting a speed of the electric motor pump, analyzing the detected values ​​for the control current and the speed of the electric motor pump in order to detect or predict a fault of the electric motor pump on the basis thereof. In this process, a deviation in the control current over time at a given speed is interpreted as being solely attributable to wear in the electric motor pump. If a clear trend emerges in this regard, or if a corresponding threshold is exceeded, a fault can be predicted or detected. In a further development of the method according to the invention, the method may further comprise the following steps: detecting a fault in the electric motor pump if a rise or fall in the control current over time at a constant speed of the electric motor pump exceeds or falls below a threshold value, preferably by detecting the rise or fall based on the operating points considered over time, which consist of a respective data pair of control current and associated speed of the electric motor pump at a specific time, and / or predicting a fault in the electric motor pump if a tendency of a rise or fall in the control current over time at a constant speed of the electric motor pump indicates that a threshold value will be exceeded or fallen below.preferably by predicting the trend using an extrapolation of the operating points considered over time, which consist of a respective data pair of control current and associated speed of the electric motor pump at a specific time. It can advantageously be provided that, when detecting or predicting a fault, only those values ​​of the actuating current and the speed of the electric motor pump are used for fault detection or prediction that have been generated in a quasi-stationary operation of the electric motor pump, preferably wherein a quasi-stationary operation is defined by keeping a speed range or a speed of the electric motor pump constant over a predetermined time in which no dynamic load change occurs. The present invention will now be explained in more detail with reference to exemplary embodiments illustrated in the figures. The figures show: Fig. 1: a schematic diagram of the inventive device for condition monitoring with an electric motor pump, Fig. 2: a system architecture of a motor control unit for controlling the electric motor pump, Fig. 3: a graph of the current characteristic curve versus the speed of the electric motor pump, which depends on the temperature of the fluid to be pumped, and Fig. 4: an idealized representation of the streamline characteristic curve to illustrate the operating principle of the inventive device. Fig. 1 shows a schematic diagram of the inventive device 1 for condition monitoring, also showing an electric motor pump 2, which in this case is implemented by a fixed displacement pump 8. The fixed displacement pump 8 is connected to the output shaft of the motor 9, which can vary its speed. In this case, the motor 9 is implemented as a permanent magnet synchronous motor, which can vary the speed of the output shaft depending on its control signal. This makes it possible to operate the fixed displacement pump 8 at different speeds and thus adjust the corresponding delivery volume of the pump. A motor control unit 10 is provided for controlling the motor 9, which outputs corresponding control signals to the motor 9 depending on the desired speed of the motor 9 or the pump 8. Compared to conventional electric motor pumps, which use a constant-speed asynchronous motor to drive a variable displacement pump, this results in a significantly improved overall efficiency. Another advantage of the architecture shown in Fig. 1 lies in the motor control unit 10, which is required anyway and always knows the speed of the output shaft (and thus the rotational speed of the displacement pump 8) and the actuating current of the electric motor 9. Based precisely on these two parameters, namely control current and speed, the invention proposes a condition monitoring system for the electric motor pump in order to predict or detect a fault. These two parameters are continuously monitored, and the recorded operating points are analyzed to determine whether a fault exists or is imminent. Fig. 2 reveals a system architecture of a motor control unit 10 for controlling the electric motor pump 2, illustrating the control with a triple cascaded control loop. As explained above, it can be seen that the motor control unit 10 has, among other things, knowledge of the actuating current and the speed of the output shaft, since the respective parameters have been tapped at the corresponding points using a first sensor 3 and a second sensor 4. These parameters are already used for controlling the permanent magnet synchronous motor, so they are already available in the implementation of the present invention. Fig. 3 shows several current characteristics RT, HT, LT from a motor, which are slightly dependent on the hydraulic temperature. In general, the relationship can be seen that the current characteristic increases only slightly with the rotational speed. It is important to note in this diagram that the rotational speed increases from right to left. In quasi-steady-state operation, where there is no dynamic load change, the control system sets a current proportional to the torque, resulting in the specific characteristic curve that increases only slightly with the rotational speed. The reason for this is that the motor current is proportional to the effective torque, which is dominated by the theoretical hydraulic pump torque "Mth". This, in turn, depends only on the hydraulic differential pressure "Δp" and the pump's displacement volume "V", both of which are constant and completely independent of the rotational speed. For the sake of simplicity, the relevant relationships are shown below: The slight increase in the characteristic curve with increasing rotational speed is mainly due to the speed-dependent increase in friction losses, as can be the case with bearings and the shaft seal. Should the mechanical friction losses in the motor-pump unit increase, this leads to an increase in the control current for the same speed. In fact, the increase in mechanical friction can be caused not only by defects in the motor bearings, but also by mechanical defects in the pump bearings. Therefore, by monitoring the required control current at a certain speed, a wear monitor for the mechanical motor-pump unit is obtained. Figure 3 also shows that low-temperature fluid, represented by the characteristic curve LT (low temperature), essentially requires a higher control current to achieve a predetermined speed, as would be the case with higher-temperature fluid, represented by the characteristic curves RT (regular temperature) and HT (high temperature). Fig. 4 shows a schematic representation illustrating the operating principle of the device according to the invention. The abscissa represents the rotational speed of the motor 8 or the pump 9, and the ordinate represents the actuating current for the motor 8 in amperes. The three essentially parallel lines (6, 7) illustrate the typical characteristic of the actuating current, which increases only slightly with the rotational speed. If one considers a constant operating point 7 over the service life of the electric motor pump 2, it can be observed that with increasing wear, the required actuating current to achieve a predetermined rotational speed increases over time. This is illustrated by the multiple operating points 7 arranged one above the other, where, for example, the operating point requiring the lowest actuating current is the oldest, and the operating point requiring the highest actuating current is the current one.Lines 5 and 6 define a corridor within which operating point 7 can be located without triggering a fault. However, if operating point 7 exceeds the threshold defined by line 5, a fault is assumed. Furthermore, it is possible to predict when the threshold (represented by line 5) will be exceeded by estimating a time based on past operating point 7 values. This can be achieved, for example, through extrapolation that estimates the progressive deterioration and arrives at a future point in time when the threshold, represented by line 5, is expected to be exceeded. Furthermore, when defining the monitoring limits, i.e., the threshold values ​​represented by lines 5 and 6, it is advisable to consider the measurement tolerances of the motor current, as these can also have a range of ±10%. The rotational speed, on the other hand, can be determined much more precisely, so it does not require any further consideration in this regard. Since changes in the determined operating point can also depend on the temperature of the fluid being pumped, the fluid temperature can be measured and used to correct the operating point. To account for this, Fig. 4 shows that the control current IRMS can be represented by a function that depends on the temperature of the fluid being pumped by the electric motor pump and the motor speed. Furthermore, the output power of the motor control unit 10 can also be determined and, using corresponding recorded parameters, all of which are known to the motor control unit, a predicted or detected fault can be assigned to either the motor or the pump. The output power of the motor control unit, the shaft power (motor output power), and the pump output power can be calculated in the computer unit of the condition monitoring device, thus determining the efficiency of the motor and the pump. If the efficiency of either component drops unusually sharply, the detected or predicted fault is attributed to that component. Specifically, the output power of the motor control unit 10 can be determined using the formulas shown below, all of which are known to those skilled in the art. The key feature of the invention lies in the evaluation and the ability to derive benefits from it: Output power of the motor control unit = 3 / 2 * (Vq * Iq + Vd * Id). The shaft power of the shaft connecting the motor and the pump = Mth * n. The output power of the pump = P * ΔP * k. Furthermore, the torque constant kT can be determined from both the motor's torque equation and the hydraulic data supplied to the motor control unit 10. This constant is identical for a fault-free electric motor pump, regardless of the method used for its determination. Any deviation can also be used to attribute a detected or predicted fault to this constant. Firstly, a torque constant can be calculated using the torque equation of the motor; M = 3 / 2 * kT * Iq. On the other hand, the torque constant can be calculated if the voltage and rotational speed are known, namely by the equation: Vq = kT * n One example of a possible motor fault is demagnetization. In this case, the motor can be monitored by correlating the torque (Mthaus) with the hydraulic data for determining kT, and comparing the kT from the voltage and speed data, so that a demagnetization fault can be detected.

Claims

Device (1) for monitoring the condition of an electric motor pump (2), preferably arranged in an aircraft, comprising: at least a first sensor (3) for detecting an actuating current (IRMS) of the electric motor pump (2), at least a second sensor (4) for detecting a rotational speed (n) of the electric motor pump (2), a computer unit for processing and evaluating the data of the electric motor pump (2) detected by the first sensor (3) and the second sensor (4), and a storage unit for storing the values ​​for the actuating current (IRMS) and the rotational speed (n) of the electric motor pump (2) detected by the first sensor (3) and the second sensor (4), wherein the computer unit is designed to analyze the values ​​for the actuating current (IRMS) and the rotational speed (n) stored in the storage unit and, on the basis thereof, to detect or predict a fault of the electric motor pump (2), characterized in thatthat the electric motor pump (2) is based on the principle of variable motor control with a constant displacement pump (8), in which a motor control unit (10) is provided to control a motor (9), in particular a permanent magnet synchronous motor, connected to the constant displacement pump (8), and the computer unit is further designed to assign a detected or predicted fault to the motor (9) or the constant displacement pump (8), wherein for this purpose the computer unit is designed to: determine the output power of the motor control unit (10), determine the input power, i.e. the shaft power of the motor (9), determine the output power of the displacement pump (8), calculate the efficiency of the subcomponents motor (9) and displacement pump (8) from the power ratios in order to assign a fault to the motor (9) or the displacement pump (8). Device (1) according to claim 1, wherein the computing unit is designed to analyze the values ​​for the actuating current (IRMS) and the rotational speed (n) stored in the memory unit and to detect a fault of the electric motor pump (2) if an increase in the actuating current (IRMS) over time at a constant rotational speed (n) of the electric motor pump (2) exceeds or falls below a threshold value (5, 6). Device (1) according to one of the preceding claims, wherein the computer unit is designed to predict a fault of the electric motor pump (2) if a tendency of a time-dependent increase in the actuating current (IRMS) at constant speed (n) of the electric motor pump (2) indicates that a threshold value (5, 6) is being exceeded. Device (1) according to one of the preceding claims, wherein the computer unit is designed to use only those values ​​of the control current (IRMS) and the rotational speed (n) of the electric motor pump (2) for fault detection or prognosis that have been generated in a quasi-stationary operation of the electric motor pump (2). Device (1) according to one of the preceding claims, further comprising at least one third sensor for detecting the temperature (TFluid) of the fluid pumped by the electric motor pump (2), wherein the computer unit is further designed to take the temperature (TFluid) of the pumped fluid into account in fault detection or prognosis. Device (1) according to one of the preceding claims, wherein the first sensor (3) and the second sensor (4) are integrated in the motor control unit (10). Device (1) according to one of the preceding claims, wherein the output power of the motor control unit (10) is determined by calculating: 3 / 2*(Vq * Iq + Vd * Id), the shaft power is determined by calculating: Mth* n, and the output power of the positive displacement pump (8) is determined by calculating: P* ΔP *k. Device (1) according to one of the preceding claims, wherein the computer unit is further designed to infer a fault case of a demagnetized motor (9) by determining the torque constant kT of the motor (9) on the one hand from the hydraulic data and on the other hand from the voltage and speed data and comparing them together. Device (1) according to the preceding claim 8, wherein the torque constant kT of the motor (9) is calculated from: M = 3 / 2 * kT* Iq, and the torque constant kT is calculated from the voltage and speed data from: Vq = kT * n. System comprising an electric motor pump (2) and a device (1) for condition monitoring according to one of the preceding claims. Method for condition monitoring of an electric motor pump (2), wherein the electric motor pump (2) is based on the principle of variable motor control with a constant displacement pump (8) and a motor control unit (10) controls a motor (9) connected to the constant displacement pump (8), the method comprising the steps: acquiring a control current (IRMS) of the electric motor pump (2), acquiring a rotational speed (n) of the electric motor pump (2), analyzing the acquired values ​​for the control current (IRMS) and the rotational speed (n) of the electric motor pump (2) in order to detect or predict a fault of the electric motor pump (2) on the basis thereof, attributing a detected or predicted fault to the motor (9) or the constant displacement pump (8), comprising the steps: determining the output power of the motor control unit (10), determining the input power, i.e.the shaft power of the motor (9), determining the output power of the positive displacement pump (8), and calculating the efficiency of the subcomponents motor (9) and positive displacement pump (8) from the quotients of the power, in order to assign a fault to the motor (9) or the positive displacement pump (8). Method according to the preceding claim 11, further comprising the steps: detecting a fault of the electric motor pump (2) if a time-dependent increase in the control current (IRMS) at constant speed (n) of the electric motor pump (2) exceeds or falls below a threshold value (5, 6), and / or predicting a fault of the electric motor pump (2) if a tendency of a time-dependent increase in the control current (IRMS) at constant speed (n) of the electric motor pump (2) indicates that a threshold value (5, 6) is being exceeded. Method according to claim 11 or 12, wherein, when detecting or predicting a fault, only those values ​​of the control current (IRMS) and the rotational speed (n) of the electric motor pump (2) are used for fault detection or prediction that have been generated in a quasi-stationary operation of the electric motor pump (2).

Images