A motor controller, a propulsion system for an electric or hybrid aircraft and a method for operating a motor

Abstract

A propulsion system for an electric or hybrid aircraft (100), comprising: - a motor (94) comprising a rotor and a stator; - a motor controller (93) connected to the motor (94), wherein the motor controller (93) is arranged for supplying driving signals (is, vs) to the motor (94), wherein the motor controller (93) comprises a control structure (950) configured to implement a field-oriented control for controlling the driving signals (is, vs) in dependency of a given rotor position (Θdet), wherein the control structure (950) is adapted to determine a rotor position static deviation (ΔΘdet) between the given rotor position (Θdet) and an effective rotor position (Θact) of the rotor, wherein the control structure (950) is further arranged to correct the given rotor position (Θdet) based on the determined rotor position static deviation (ΔΘdet).

Description

A motor controller, a propulsion system for an electric or hybrid aircraft and a method for operating a motor Technical domain

[0001] The present disclosure concerns propulsion systems and methods for an electric or hybrid aircraft. Related art

[0002] Electric and hybrid vehicles have become increasingly significant for the transportation of people and goods. Such vehicles can desirably provide energy efficiency advantages over combustion-powered vehicles and may cause less air pollution than combustion-powered vehicles during operation.

[0003] Although the technology for electric and hybrid automobiles has significantly developed in recent years, many of the innovations that enabled a transition from combustion-powered to electric-powered automobiles unfortunately do not directly apply to the development of electric or hybrid aircraft. The functionality of automobiles and the functionality of aircraft are sufficiently different in many aspects so that many of the design elements for electric and hybrid aircraft must be uniquely developed separate from those of electric and hybrid automobiles.

[0004] Moreover, any changes to an aircraft's design, such as to enable electric or hybrid operation, also require careful development and testing to ensure safety and reliability. If an aircraft experiences a serious failure during flight, the potential loss and safety risk from the failure may be very high as the failure could cause a crash of the aircraft and pose a safety or property damage risk to passengers or cargo, as well as individuals or property on the ground.H55-18-PCT1

[0005] The certification standards for electric or hybrid aircraft are further extremely stringent because of the risks posed by new aircraft designs. Designers of aircraft have struggled to find ways to meet the certification standards and bring new electric or hybrid aircraft designs to market.

[0006] In view of these challenges, attempts to make electric and hybrid aircraft commercially viable have been largely unsuccessful. New approaches for making and operating electric and hybrid aircraft thus continue to be desired.

[0007] Flying a manned or unmanned aircraft such as an airplane can be dangerous. Problems with the aircraft may result in injury or loss of life for passengers in the aircraft or individuals on the ground, as well as damage to goods being transported by the aircraft or other items around the aircraft.

[0008] The reliability of systems can be improved with redundant subsystems. Various designs have been suggested in order to replace a faulty subsystem with a backup subsystem. For example, in the context of electric powered object or vehicles, US20171210229 A1 and US20111254502A1 both describe a fault-tolerant battery management system in which the state of battery cells is monitored and / or controlled by redundant battery management systems (BMS), such that a default in one BMS does not prevent the battery from functioning as long as the redundant BMS performs properly. However, if the two BMS are identical, they are more likely to present the same defaults or conception problems, and are also more likely to have failure simultaneously or at short interval. Moreover, those solutions have not been designed with the aim of certification for aircraft; adding additional components increase the complexity of the system and makes the certification even more difficult.

[0009] In order to attempt to mitigate potential problems associated with an aircraft, numerous organizations have developed certificationH55-18-PCT1standards for ensuring that aircraft designs and operations satisfy threshold safety requirements. The certification standards may be stringent and onerous when the degree of safety risk is high, and the certification standards may be easier and more flexible when the degree of safety risk is low.

[0010] As an example, the FAA advisory circular AC 25.1309-1 describes acceptable means for showing compliance with the airworthiness requirements of US Federal Aviation Regulations defines different levels of failure conditions according to their severity: - Failure Conditions with No Safety Effect. - Minor Failure Conditions. - Major Failure Conditions. - Hazardous Failure Conditions must be no more frequent than Extremely Remote. - Catastrophic Failure Conditions must be Extremely Improbable.

[0011] While airplanes must be designed so that hazardous and catastrophic failure conditions are extremely remote or even extremely improbable, those severe failure conditions must nevertheless be monitored, so that warning signals are sent to the pilot and driver who may attempt to remedy to the condition or try to land the aircraft. The monitoring and warning systems must be reliable and also requires certification.

[0012] Such certification standards have unfortunately had the effect of slowing commercial adoption and production of electric or hybrid aircraft. Electrical hybrid aircraft may, for example, utilize new aircraft designs relative to traditional aircraft designs to account for differences in operations of electric or hybrid aircraft versus traditional aircraft. The new designs however may be significantly different from the traditional aircraft designs. These differences may subject the new designs to extensive testingH55-18-PCT1prior to certification. The need for extensive testing can take many resources, time and significantly drive up the ultimate cost of the aircraft.

[0013] Compliance of a monitoring and warning subsystem with the certification standard depends on the severity of the monitored failure condition. Therefore, a hazardous or catastrophic failure condition requires a strict level of certification of the corresponding monitoring and warning system, while a minor failure condition or a condition without any safety effect have lower safety requirements and requires a monitoring and warning system that is easier to certify, or requires no certification.

[0014] There is therefore a need for simplified, yet robust, components and systems for an electric powered aircraft that simplify and streamline certifications requirements and reduce the cost and time required to produce a commercially viable electric aircraft.

[0015] Subsystems for propelling the electric aircraft, such as motor control systems, are one of the most critical subsystems in an electric or hybrid aircraft, as a loss of propulsion might lead to catastrophic scenarios. At the same time, there is a necessity to operate these subsystems efficiently, as the range of aircrafts equipped with alternative electric propulsion concepts is still limited. In fact, concepts known from the prior art allow the subsystems to be operated more efficiently, thereby providing improved performance. However, this is often accompanied by additional components, weight, and / or complexity. All of this is undesirable, as additional weight reduces the aircraft's range, and additional components or complexity may come at further costs for certification or make the certification even more difficult.

[0016] According to one aspect, the disclosure is related to the control of the motor based on the rotor position.

[0017] Since the field and torque stator current components are defined with respect to the rotor flux position, a precise knowledge of the effectiveH55-18-PCT1or current rotor flux position is a prerequisite for efficient control of the motor's torque output. The rotor position may not be equivalent to the rotor flux position. For synchronous machines, permanent magnets or an external excitation produce the flux that has a constant position with respect to the rotor. On the other hand, in induction machines, the flux produced by the rotor does not have a constant position with respect to the single point on the rotor.

[0018] The rotor flux position, also known as the rotor angle (in synchronous machines) or the rotor flux angle, can refer to a parameter in electrical engineering that describes the position of the rotor's magnetic field in a rotating machine, such as an AC synchronous motor. It can typically be measured in electrical degrees and can vary continuously with time as the rotor rotates. Various methods, such as sensor-based and sensorless techniques, can be used to determine or estimate the rotor position in different types of machines.

[0019] In sensor-based techniques, the rotor flux position is conventionally determined using a rotor position sensor in mechanical connection with the rotor, whereby the sensor signals can be fed into the control structure. The control structure, in return, determines the given rotor position in dependency on the sensor signals, which can be used for controlling the field and torque component stator currents.

[0020] The rotor position static deviation between the given rotor position (for example the measured or otherwise determined position) and the effective rotor position can be caused by misalignment or mispositioning of the rotor position sensor, due to manufacturing tolerances or vibrations and temperature variations, by delays during reading the sensor signals, by delays due to the signal processing, or an incorrect initial rotor position information.

[0021] Static deviations can also occur in sensorless techniques, in which motor models and / or observers are used to determine and track theH55-18-PCT1effective rotor position. Inconsistencies in the used motor model may cause such static deviations. Dynamic deviations, such as those caused by disturbances in the signal path of the encoder signal, may not be considered as static deviations. However, the control structure can be enabled to correct any deviation of the given rotor position from the effective rotor position by means of the determined static deviation, which can function as a permanent correction term in the control structure.

[0022] The motor can comprise a rotor position sensor configured to output rotor position sensor signals for indicating a rotor position, wherein the motor controller can be arranged to receive the rotor position sensor signals, and the control structure can be adapted to determine the given rotor position based on the rotor position sensor indicating the rotor position. The rotor position sensor can be provided as a rotary encoder, incremental encoder, or the like. As outlined before, rotor position sensors are typically used in sensor-based vector control applications.

[0023] The motor controller can alternatively comprise a measurement circuit for measuring voltages and currents of the driving signals, wherein the control structure can comprise a rotor position estimator configured to determine the given rotor position based on measured voltages and currents. As outlined before, rotor position estimators are typically used in sensorless vector control applications. Typical rotor position estimators can be based on signal-injection methods, motor model-based methods, or observer-based methods, whereas all have in common that the measurement of the stator voltages and currents is a prerequisite for position estimation.

[0024] The deviations between the given (determined) position of the rotor and its effective position result in inefficient motor control.

[0025] A precise knowledge of the effective rotor position is also important if a field oriented control scheme is used. A field-oriented control is used nowadays in many motor controllers to control theH55-18-PCT1operation efficiently. The advantage of the field-oriented control over other motor control approaches is that an AC motor can be controlled under all operating conditions like a separately excited DC motor. This means that the AC motor acts like a DC motor, with the field flux linkage and armature flux linkage created by their respective field and armature (or torque component) currents aligned orthogonally. As a result, when controlling the torque in DC machines by controlling the armature flux linkage, the field flux linkage is not impacted, allowing for a dynamic torque response.

[0026] Similarly in AC machines, stator currents can be controlled by separately controlling their field and torque current components. The control of the former enables the equivalent to the control of the field flux linkage in DC machines, while the control of the latter is equivalent to the armature flux linkage control in DC machines.

[0027] In summary, the given rotor position that is determined with a sensor or sensorless system is often different from the effective rotor position. This results in inefficient control of the motor, in particular in higher losses, reduced torque or speed, and excessive heat generation. A precise alignment of the position sensors is therefore of paramount importance in an electric or hybrid airplane. Short disclosure

[0028] An aim of the present disclosure is to overcome or at least mitigate at least some of the shortcomings and limitations of the state of the art.

[0029] In particular, it is an aim of the disclosure to propose a propulsion system for an electric or hybrid aircraft that allows for a more efficient control of the motor.H55-18-PCT1

[0030] According to one aspect, this problem is solved with a propulsion system comprising: - a motor having a rotor and a stator; - a motor controller connected to the motor, wherein the motor controller is arranged for supplying driving signals to the motor, wherein the motor controller comprises a control structure configured to implement a field- oriented control for controlling the driving signals in dependency of a given rotor position, wherein the control structure is adapted to determine a rotor position static deviation between the given rotor position and an effective rotor position of the rotor, wherein the control structure is further arranged to correct the given rotor position with the use of the determined rotor position static deviation for controlling the driving signals based on the corrected given rotor position.

[0031] The motor controller can thus compensate the difference between a given (i.e., expected) rotor position and an effective rotor position, which can result in an optimal control of the motor's torque output.

[0032] The term “given rotor position” can refer to a calculated rotor position if the control structure is implemented in a digital controller or can alternatively refer to an electrical value outputted by an analog circuit if the control structure is implemented using analog techniques.

[0033] The driving signals can be supplied by the motor controller to the motor in the form of stator voltages and stator currents, wherein the control structure comprises a d-axis current controller (flux controller) configured to control a d-axis voltage component of the stator voltages in dependency of a d-axis current component setpoint of the stator currents and a q-axis current controller (torque controller) configured to control a q- axis voltage component of the stator voltages in dependency of a q-axis current component setpoint of the stator currents, wherein said voltage components and said current component setpoints can be referenced in a dq-reference frame. The control structure can be further arranged with aH55-18-PCT1feed-forward decoupling structure configured to decouple the d-axis current controller and the q-axis current controller from each other, in particular the cross-coupled voltage terms thereof.

[0034] The d-axis current controller can be used to maintain a constant or desired level of magnetic flux in the machine, even when the electrical and mechanical conditions of the system change. To achieve this, the d-axis current controller can adjust the amount of current flowing through the stator windings, which in turn affects the amount of magnetic flux produced. The q-axis current controller can be used to maintain a constant or desired level of torque output. This can be achieved by adjusting the amount of current flowing through the stator windings, thereby affecting the torque produced. The stator currents, however, depend on the stator voltages (and other parameters, such as the rotor position). Therefore, the stator voltages can be set and controlled in dependency on the rotor position and the flux linkage and / or torque setpoint using the d-axis and q- axis current controller. The controllers can be implemented using conventional PI-controllers.

[0035] Both the stator voltages and currents can be present in the control structure in the form of d-axis and q-axis components in the dq- reference frame. They can be obtained by measurement of the stator currents, whereby the stator currents can be transformed into the dq- reference frame using the Clark and Park transformation. The d-axis current controller controls the d-axis component of the stator voltages in dependency on the setpoint for the d-axis component of the stator currents. The q-axis current controller controls the q-axis component of the stator voltages in dependency on the setpoint for the q-axis component of the stator currents. The values outputted by the said controllers can be transformed using the inverse Clarke and Park transformation into stator voltages that the motor controller outputs or supplies to the motor.H55-18-PCT1

[0036] The q- and d-axis components in the dq-reference frame of the stator voltages can be referred to as the d-axis voltage component and the q-axis voltage component, respectively.

[0037] The control structure can be further adapted to set the said current component setpoints to zero and can be configured to detect a steady state at outputs of the d-axis and the q-axis current controller, wherein the control structure can be provided to determine the rotor position static deviation based on output values provided by the d-axis and q-axis current controllers when the steady state of the outputs of the d-axis current controller and q-axis current controller is detected. Instead of detecting the steady state, the control structure can wait a predetermined amount of time, in which the outputs conventionally settle.

[0038] The said d- and q-axis components can be controlled to zero in dependency of the respective setpoint for the d-axis and q-axis stator current component of the d-axis and q-axis current controller, respectively.

[0039] The term steady state can refer to the state where the d-axis or the q-axis current controller settles its output, and the respective controller's output value remains constant or stable over time.

[0040] The feed-forward decoupling structure can decouple the cross- coupled voltage terms that are inherently present when consulting the motor equations. The feed-forward decoupling structure will be, however, discussed in more detail in the course of the present disclosure. The feed- forward decoupling structure can be useful to compensate for unmodelled dynamics but may or may not be essential for the present aspect.

[0041] The control structure can also be adapted to determine the rotor position static deviation based on a ratio of the output value of the d-axis current controller and the output value of the q-axis current controller. H55-18-PCT1

[0042] In addition, the control structure can be adapted to determine an effective rotor flux linkage, wherein the control structure can be further adapted to correct a parametrized rotor flux linkage with the use of the determined effective rotor flux linkage for controlling the driving signals based on the corrected given rotor position and the corrected parametrized rotor flux linkage.

[0043] The effective rotor flux linkage can refer to the average value of the magnetic field that can be effectively generated by the rotor and linked to the stator windings over a single electrical cycle or complete turn of the rotor. The term parametrized can refer to a variable in a digital control structure or a reference value in an analog control structure.

[0044] The control structure can also be provided to determine the effective rotor flux linkage based on the output value provided by the q- axis current controller and an actual rotor speed when the steady state of the d-axis current controller and q-axis current controller outputs are detected. The basic principle can refer to the effect that the rotor induces a voltage into the stator windings. This so-called back-EMF can be determined in the control structure, in particular at the output of the q-axis current controller; suppose the rotor speed is constant, and the torque and the d-axis current controllers are settled. The rotor speed can be constant over a relatively large amount of time thanks to the mass moment of inertia of the rotor and a propeller mechanically connected to the shaft of the motor. The actual rotor speed can refer to the speed obtained using the rotor position sensor providing the control structure with sensor signals.

[0045] The motor can be configured as a synchronous motor, preferably as a three-phase permanent magnet synchronous motor. However, the control structure can also be used for motors differently configurated, subjected to a necessary adaptation of the control structure. H55-18-PCT1

[0046] The propulsion system can also comprise a control means configured to accommodate the control structure. The control structure can be implemented in the control means on a digital controller or in the form of an analog circuit. However, the term accommodate can refer to the circumstance that the control means comprises the control structure, wherein the control structure can be exemplarily implemented on a microcontroller, digital processor, or in the form of an analog circuit.

[0047] Another aspect relates to a method for controlling a propulsion system in an electric or hybrid aircraft. The method can be used in a propulsion system as previously disclosed to correct a given rotor position which can be used for motor control in case of a deviation with respect to an effective rotor position.

[0048] The control quality of the motor can be improved, which can result in optimal control of the motor's torque output. The method can also be efficiently implemented in and executed by a control device implementing a vector control.

[0049] The method can comprise the steps of: - supplying the motor with driving signals; - rotating the rotor of the motor, preferably at a constant speed; - setting the current component setpoints of the d-axis current controller and the q-axis current controller to zero; - detecting the steady state at the outputs of the d-axis current controller and the q-axis current controller or waiting a predetermined amount of time to let the controllers to reach the steady state; - determining the rotor position static deviation between the given rotor position and the effective rotor position of the rotor when the steady state is detected or reached; - correcting the given rotor position using the determined rotor position static deviation; - controlling the driving signals using the corrected given rotor position. H55-18-PCT1

[0050] The method can further comprise the step of: - calculating the ratio of the output value of the d-axis current controller and the output value of the q-axis current controller, wherein the rotor position static deviation can be determined using the said calculated ratio. Calculating can mean that a calculated result is outputted if the method is implemented on a digital controller or that the result may be provided in the form of an electrical value, suppose the method is implemented using analog techniques.

[0051] The method can also comprise the steps of: - determining the effective rotor flux linkage; - correcting the parametrized rotor flux linkage using the determined effective rotor flux linkage; - controlling the driving signals using the corrected given rotor position and the corrected parametrized rotor flux linkage.

[0052] The effective rotor flux linkage being determined using the output value provided by the q-axis current controller and the actual rotor speed.

[0053] The steps for determining the static deviation between the given rotor position and the effective rotor position is preferably performed while the electric or hybrid aircraft is grounded; the measure typically takes less than 10ms and only needs to be repeated before each flight, or less often, since the alignment of the position encoder is unlikely to change fast. A determination of this static deviation can also be done while the aircraft is flying, if needed.

[0054] The steps for determining the effective rotor flux linkage can be performed while the electric or hybrid aircraft is grounded or flying, preferably at a constant speed. A constant speed is not a prerequisite: the detection period may be quite small (several milliseconds), so that the motor speed can be considered constant within this short detection period. H55-18-PCT1This measure can be repeated several time during each flight, as the flux is more likely to change with temperature variations for example.

[0055] Correcting the given rotor position can be useful to control the torque with more accuracy and thereby provide an improved torque output, wherein correcting the parametrized rotor flux linkage can be useful to better control the motor in field-weakening operation and thereby improving the motor power output in the field-weakening region. It can therefore be desirable to correct the given rotor position when the aircraft is grounded, such that the optimum performance of the motor and the motor control is set from the beginning, at least before take-off. Correcting the parametrized rotor flux linkage while flying can also be useful, and it does not affect the operation of the electric drive, as the duration for determining the effective rotor flux linkage procedure is significantly shorter than the mechanical time constant of the propeller shaft.

[0056] Another aspect of the disclosure relates to a propulsion system for an electric or hybrid aircraft which can be configured to operate the motor in two different modes. Wherein the first mode, the motor can be controlled to provide a constant torque output, and in the second mode the motor can be controlled with constant voltage, in dependency on an actual DC link voltage. The second mode conventionally can be referred to as field-weakening, or constant power, whereby the rotor flux is weakened, and the motor's torque output can be controlled in dependency of the actual DC voltage of the motor controller. The present aspect can contribute to the overall performance of the motor control, in particular, to optimally control the motor in field-weakening and thereby optimally utilizing the motor capabilities or performance.

[0057] The propulsion system can comprise: - a motor comprising a rotor and a stator; - a motor controller can be connected to the motor, wherein the motor controller can be arranged for supplying driving signals in form of stator H55-18-PCT1voltages and stator currents to the motor, wherein the motor controller can comprise a control structure configured to implement a field-oriented control for controlling the driving signals, and can be arranged to control the driving signals for controlling the motor in at least two different modes, wherein in a first mode the control structure can be adapted to operate the motor with constant torque and in a second mode the control structure can be provided to operate the motor with constant voltage in which the control structure can be further configured to weaken an amount of rotor flux of the rotor in dependency of an actual DC link voltage of the motor controller and a rotor speed.

[0058] The term "mode" can refer to a control mode or an operational mode in which the motor controller or the control structure remains and can be considered as the state in which the control structure controls the diving signals. The amount of rotor flux can refer to the strength of the magnetic field generated by the rotor. In dependency can refer to the circumstance that, dependent on the level of the DC link voltage and the rotor speed, the amount of rotor flux can be reduced less, more, or even not at all. The motor controller can be configured with a DC voltage link, and the actual DC link voltage can be measured by means of a voltage sensor.

[0059] The motor can comprise a rotor position sensor configured to output rotor position sensor signals for indicating a rotor position, wherein the motor controller can be arranged to receive the rotor position sensor signals, and the control structure can be adapted to determine a rotor speed based on the rotor position sensor signals, or the motor controller can comprise a measurement circuit for measuring voltages and currents of the driving signals, wherein the control structure can comprise a rotor position estimator configured to determine the rotor speed based on measured voltages and currents.

[0060] The control structure can be arranged to control and limit the torque of the motor in the second mode in dependency on the amount of H55-18-PCT1the weakened rotor flux and in dependency of a predetermined limit of the stator currents.

[0061] The motor controller for an electrically or hybrid driven aircraft, may comprise: - an input end configured to receive electrical energy from an electrical source; - an output end configured to supply electrical energy to and / or receive electrical energy from a motor comprising a stator and a rotor; - a signal input configured to receive signals from a speed or position sensor, wherein the electrical source, the motor, and the speed sensor are external to the motor controller. The motor controller may be arranged to provide electrical energy at the output end in the form of a set of driving signals with variable voltage and variable frequency with, wherein the motor controller is configured to control the set of driving signals in different control modes, wherein in a sensor-based control mode the motor controller is configured to control the set of driving signals based on a first control scheme, wherein the first control scheme is provided to utilize the signals receivable at the signal input, and in a sensorless control mode, the motor controller is configured to control the set of driving signals based on a second control scheme configured to disregard the signals receivable at the signal input.

[0062] The sensor-based control mode can be termed first control mode, the sensorless control mode can be termed second control mode, and the initialization control mode can be termed third control mode. The corresponding terms can be used interchangeably during the present disclosure.

[0063] The motor controller provides the advantage that it can provide the set of driving signals to the motor with the use of two independent control schemes, in which the first control scheme makes use of the speed sensor signals, and in which the second control scheme disregards the speed sensor signals. The functioning of the motor controller and the H55-18-PCT1corresponding operation of the motor is ensured independently of the presence of a speed sensor. Suppose the speed sensor fails during the operation of the motor controller, i.e. during flight, the motor controller can advantageously unperturbed continue its operation.

[0064] The electrical source can be any kind of electrical source capable of providing electric energy to the motor controller. For instance, batteries, or fuel electric generators can be used as an electrical source. The output end can be configured to supply electrical energy to or receive electrical energy from the motor. However, the motor can be external to the motor controller, and thus can be connectable to the motor controller. The output end can also be referred to input-output end for better describe the functioning.

[0065] The speed or position sensor can be a synonym for every device that is suitable to provide the motor controller with the position or the speed (angular frequency) of the rotor of the motor. The position provided by the sensor can be relative or absolute. The speed or position sensor can be part of the motor controller or can preferably belong to the motor. In this case, the sensor can be considered as being external to the motor controller. However, in any case the motor controller can be configured to receive sensor signals. The signal input can be configurated as a dedicated interface to connect a speed sensor. The interface can be provided as a hard-wired connection to the speed or position sensor or as an interface connected to a communication bus. The term "receivable" with respect to the sensor signals at the signal input can imply that the sensor definitely provide the related signal, in case the rotor or the motor rotates, at least in the absence of a defect.

[0066] Some or all previously embodiments can be combined when it is useful and feasible from a technical standpoint.

[0067] According to another aspect of the disclosure, a propulsion system for an electric or hybrid aircraft can comprise: H55-18-PCT1- an electrical source; - a first motor comprising a stator and a rotor; - a speed or position sensor operably coupled to the rotor for measuring a speed and / or a position of the said rotor; - a motor controller as previously described (including any embodiments or any combination thereof) connected to the electrical source at the input end, connected to the speed sensor at the signal input, and connected to the first motor at the output end.

[0068] In an embodiment, the first motor can be configured as a permanent-magnet synchronous motor comprising a propeller coupled to the rotor. The propeller can be configured with a plurality of propeller blades, wherein the pitch or attack angle of said the blades can be varied to adjust the thrust. The propeller is conventionally mechanically connected to the motor shaft.

[0069] In another embodiment, the propulsion system can comprise a second motor mechanically coupled to the rotor of the first motor. The second motor can be configured with a smaller power rating compared to the first motor. The second motor can be energized if a failure at the signal input is detected, which may be caused by the loss or the failure of the speed sensor. The second motor can be configured as an induction machine, synchronous machine or a permanent-magnet synchronous motor.

[0070] The second motor can be used to keep the first motor rotating, such that the first motor supplies energy to the output end of the motor controller, for determining the rotor position of the first motor. For instance, if the second control scheme is entered, the second motor can be de-energized. The second motor can be useful if the aircraft is grounded and there is no airstream surrounding the propeller for turning the rotor of the first motor to supply electrical energy to the motor controller. H55-18-PCT1

[0071] All aspects and sub-aspects as set out herein can be combined in whole or in part when technically feasible and useful. Otherwise they remain independent from one another. Short description of the drawings

[0072] Exemplar embodiments or aspects are disclosed in the description and illustrated by the drawings in which: Fig.1A illustrates an aircraft, such as an electric or hybrid aircraft; Fig.1B illustrates a simplified block diagram of an aircraft; Fig.2 illustrates management systems for operating an aircraft; Fig.3 illustrates a battery monitoring system for an aircraft; Figs.4 and 5 illustrate implementations of battery monitoring circuits; Figs.6 and 7 illustrate implementations of master circuits for monitoring battery monitoring circuits; Figs.8, 9, 10, 11, 12, and 13 illustrate schematic views of implementations of a power management system; Figs.14A and 14B illustrate a battery module usable in an aircraft; Figs.15A and 15B illustrate a power source formed of multiple battery modules; Fig.16 illustrates multiple power sources arranged and connected for powering an aircraft; H55-18-PCT1Figs.17A and 17B illustrate multiple power sources positioned in a nose of an aircraft for powering the aircraft; Figs.18A and 18B illustrate multiple power sources positioned in a wing of an aircraft for powering the aircraft; Fig.19 illustrates a motor with multiple field coils; Fig.20, 21, 22 illustrate motors connected to a motor controller in different arrangements; Fig.23 illustrates a simplified control diagram for a motor controller connected to a motor; Fig.24 illustrates a torque diagram of a motor that is controlled by a motor controller using the control diagram of Fig.23; Fig.25 illustrates a torque diagram and further motor parameters controlled by a motor controller using the control diagram of Fig.23; Fig.26 illustrates a rotating dq-reference frame along with a flux and a stator current space vectors; Fig.27 illustrates a variant of the control diagram of Fig.23, which includes a feed-forward compensation structure; Fig.28 illustrates two rotating dq-reference frames; Fig.29 illustrates a flowchart of a process for determining a phase angle error and a rotor flux error; Fig.30 illustrates an influence of the rotor position estimation error in the presence of a back-EMF in the d- and q- axis in the rotating dq- reference frame; H55-18-PCT1Fig.31 represents a diagram of a field-weakening block using the motor control diagram from Fig.27; Fig.32 illustrates a comparison between a conventional feed-forward based method for controlling a motor in field-weakening operation and the method of Fig.31 for controlling a motor in field-weakening operation in presence of motor parameters deviation; Fig.33 illustrates a propulsion system with a braking chopper; Fig.34 illustrates a method for decelerating a propeller using the propulsion system of Fig.33; Fig.35 illustrates different control parameters and the influence on a DC link voltage while decelerating a rotor with the use of the method for decelerating; Fig.36 illustrates a simplified control diagram for a motor controller connected to a motor. Examples of embodiments System Overview

[0073] Fig.1A illustrates an aircraft 100, such as an electric or hybrid aircraft, and Fig.1B illustrates a simplified block diagram of the aircraft 100. The aircraft 100 includes a motor 110, a management system 120, and a power source 130. The motor 110 can be used to propel the aircraft 100 and cause the aircraft 100 to fly and navigate. The management system 120 can control and monitor the components (equipment) of the aircraft 100, such as the motor 110 and the power source 130. The power source 130 can power the motor 110 to drive the aircraft 100 and power the management system 120 to enable operations of the management system 120. The management system 120 can include one or more motor controllers as well H55-18-PCT1as other electronic circuitry for controlling and monitoring various components of the aircraft 100.

[0074] Fig.2 illustrates components 200 of an aircraft, such as the aircraft 100 of Figs.1A and 1B. The components 200 can include a power management system 210, a motor management system 220, and a recorder 230, as well as a first battery pack 212A, a second battery pack 212B, a warning panel 214, a fuse and relay 216, a converter 217, a cockpit battery pack 218, a motor controller 222, one or more motors 224, and a throttle 226.

[0075] The power management system 210, the motor management system 220, and the recorder 230 can monitor communications on a communication bus, such as a controller area network (CAN) bus, and communicate via the communication bus. The first battery pack 212A and the second battery pack 212B can, for instance, communicate on the communication bus enabling the power management system 210 to monitor and control the first battery pack 212A and the second battery pack 212B. As another example, the motor controller 222 can communicate on the communication bus enabling the motor management system 220 to monitor and control the motor controller 222.

[0076] The recorder 230 can store some or all data communicated (such as component status, temperature, or over / undervoltage information from the components or other sensors) on the communication bus to a memory device for later reference, such as for reference by the power management system 210 or the motor management system 220 or for use in troubleshooting or debugging by a maintenance worker. The power management system 210 and the motor management system 220 can each output or include a user interface that presents status information and permits system configurations. The power management system 210 can control a charging process (for instance, a charge timing, current level, or voltage level) for the aircraft when the aircraft is coupled to an external power source to charge a power source of the aircraft, such as the first battery pack 212A or the second battery pack 212B. H55-18-PCT1

[0077] The warning panel 214 can be a panel that alerts a pilot or another individual or computer to an issue, such as a problem associated with a power source like the first battery pack 212A. The fuse and relay 216 can be associated with the first battery pack 212A and the second battery pack 212B and usable to transfer power through a converter 217 (for example, a DC-DC converter) to a cockpit battery pack 218. The fuse and relay 216 can protect one or more battery poles of the first battery pack 212A and the second battery pack 212B from a short or overcurrent. The cockpit battery pack 218 may supply power for the communication bus.

[0078] The motor management system 220 can provide control commands to the motor controller 222, which can in turn be used to operate the one or more motors 224. The motor controller can include an inverter for generating AC currents that are needed for operating the one or more motors. The motor controller 222 may further operate according to instructions from the throttle 226 that may be controlled by a pilot of the aircraft. The one or more motors can include an electric brushless motor.

[0079] The power management system 210 and the motor management system 220 can execute the same or similar software instructions and may perform the same or similar functions as one another. The power management system 210, however, may be primarily responsible for power management functions while the motor management system 220 may be secondarily responsible for the power management functions. Similarly, the motor management system 220 may be primarily responsible for motor management functions while the power management system 210 may be secondarily responsible for the motor management functions. The power management system 210 and the motor management system 220 can be assigned respective functions, for example, according to system configurations, such as one or more memory flags in memory that indicate a desired functionality. The power management system 210 and the motor management system 220 may include the same or similar computer hardware. H55-18-PCT1

[0080] The power management system 210 can automatically perform the motor management functions when the motor management system 220 is not operational (such as in the event of a rebooting or failure of the motor management system 220), and the motor management system 220 can automatically perform the power management functions when the power management system 210 is not operational (such as in the event of rebooting or failure of the power management system 210). Moreover, the power management system 210 and the motor management system 220 can take over the functions from one another without communicating operation data, such as data about one or more of the components being controlled or monitored by the power management system 210 and the motor management system 220. This can be because both the power management system 210 and the motor management system 220 may be consistently monitoring communications on the communication bus to generate control information, but the control information may be used if the power management system 210 and the motor management system 220 has primary responsibility but not if the power management system 210 and the motor management system 220 does not have primary responsibility. Additionally or alternatively, the power management system 210 and the motor management system 220 may access data stored by the recorder 230 to obtain information usable to take over primary responsibility. System Architecture

[0081] Electric and hybrid aircraft (rather than aircraft powered during operation by combustion) have been designed and manufactured for decades. However, electric and hybrid aircraft have still not yet become widely used for most transport applications like carrying passengers or goods.

[0082] This failure to adopt may be in large part because designing an aircraft that is sufficiently safe to be certified by certification authorities may be very difficult. The certification of prototypes may moreover not be H55-18-PCT1sufficient to certify for commercial applications. Instead, a certification of each individual aircraft and its components may be required.

[0083] This disclosure provides at least some approaches for constructing electric powered aircraft from components and systems that have been designed to pass certification requirements so that the aircraft itself may pass certification requirements and proceed to active commercial use.

[0084] Certification requirements can be related to a safety risk analysis. A condition that may occur with an aircraft or its components can be assigned to one of multiple safety risk assessments, which may in turn be associated with a particular certification standard. The condition can, for example, be catastrophic, hazardous, major, minor, or no safety effect. A catastrophic condition may be one that likely results in multiple fatalities or loss of the aircraft. A hazardous condition may reduce the capability of the aircraft or the operator ability to cope with adverse conditions to the extent that there would be a large reduction in safety margin or functional capability crew physical distress / excessive workload such that operators cannot be relied upon to perform required tasks accurately or completely or serious or fatal injury to small number of occupants of aircraft (except operators) or fatal injury to ground personnel or general public. A major condition can reduce the capability of the aircraft or the operators to cope with adverse operating condition to the extent that there would be a significant reduction in safety margin or functional capability, significant increase in operator workload, conditions impairing operator efficiency or creating significant discomfort physical distress to occupants of aircraft (except operator), which can include injuries, major occupational illness, major environmental damage, or major property damage. A minor condition may not significantly reduce system safety such that actions required by operators are well within their capabilities and may include a slight reduction in safety margin or functional capabilities, slight increase in workload such as routine flight plan changes, some physical discomfort to occupants or aircraft (except operators), minor occupational illness, minor environmental damage, or minor property damage. A no safety effect condition may be one that has not effect on safety. H55-18-PCT1

[0085] An aircraft can be designed so that different monitoring and warning subsystems, such as battery monitoring circuits, of the aircraft are constructed to have a robustness corresponding to their responsibilities and any related certification standards, as well as potentially any subsystem redundancies.

[0086] Where a potential failure of the responsibilities of a monitoring and warning subsystem would likely be catastrophic, the subsystem can be designed to be simple and robust and thus may be able to satisfy difficult certification standards. The subsystem, for instance a battery, motor or motor controller monitoring circuit, can be composed of non- programmable, non-stateful components (for example, analog or non- programmable combinational logic electronic components) rather than programmable components (for example, a processor, a field programmable gate array (FPGA), or a complex programmable logic device (CPLD)) or stateful components (for example, sequential logic electronic components) and activate indicators such as lights rather than more sophisticated displays.

[0087] On the other hand, where either (i) a monitoring and warning subsystem (such as a battery monitoring circuit, a motor monitoring circuit or a motor controller monitoring circuit) of an aircraft monitors a parameter redundantly with another subsystem of the aircraft that is composed of non-programmable, non-stateful components or (ii) a potential failure of the responsibilities of such a monitoring and warning subsystem would likely be less than catastrophic, or less than hazardous, the subsystem can be at least partly digital and designed to be complicated, feature-rich, and easier to update and yet able to satisfy associated certification standards. Such a subsystem can, for instance, include a processor or other programmable components that outputs information to a sophisticated display for presentation.

[0088] In some implementations, some or all catastrophic conditions monitored for by an aircraft can be monitored for with at least one monitoring and warning subsystem that does not include a programmable H55-18-PCT1component or a stateful component because certifications for programmable components or stateful components may demand statistical analysis of the responsible subsystems, which can be very expensive and complicated to certify. Such implementations can moreover be counterintuitive at least because an electric or hybrid aircraft may include one or more relatively advanced programmable or stateful components to enable operation of the electric or hybrid aircraft, so the inclusion of one or more subsystems in the aircraft that does not include any programmable components or any stateful components may be unexpected because the one or more relatively advanced programmable or stateful components may be readily and easily able to implement the functionality of the one or more subsystems that does not include any programmable components or any stateful components.

[0089] An aircraft monitoring system can include a first monitoring and warning subsystem and a second monitoring and warning subsystem. The second subsystem, such as a second battery monitoring circuit, can be supported by an aircraft housing and include non-programmable, non- stateful components, such as analog or non-programmable combinational logic electronic components. The non-programmable, non-stateful components can monitor a component (such as battery cells in a battery pack) supported by the aircraft housing and output a second alert to notify of a catastrophic condition associated with the component. The non- programmable, non-stateful components can, for instance, activate an indicator or an audible alarm for a passenger aboard the housing to output the first alert. The indicator or audible alarm may remain inactive unless the indicator is outputting the first alert. Additionally or alternatively, the non-programmable, non-stateful components can output the second alert to a computer aboard or remote from the aircraft (for example, to automatically trigger actions to attempt to respond to or address the catastrophic condition, such as to stop charging or activate a fire extinguisher, a parachute, or an emergency landing procedure or other emergency response feature) or an operator of the aircraft via a telemetry system. The non-programmable, non-stateful components may, moreover, not be able to control the component or at least control certain H55-18-PCT1functionality of the component, such as to control a mode or trigger an operation of the component.

[0090] The first subsystem, such as a first battery monitoring circuit, can be supported by the aircraft housing and include a processor (or another programmable or stateful component), as well as a communication bus. The processor can monitor the component from communications on the communication bus and output a first alert to notify of a catastrophic condition or a less than catastrophic condition associated with the component. The processor can, for instance, activate an indicator or audible alarm for a passenger aboard the housing to output the first alert. Additionally or alternatively, the processor can output the first alert to a computer aboard or remote from the aircraft (for example, to automatically trigger actions to attempt to address the catastrophic condition, such as to activate a fire extinguisher, a parachute, or an emergency landing procedure) or an operator of the aircraft via a telemetry system. The processor may control the component.

[0091] The non-programmable, non-stateful components of the second subsystem additionally may not be able to communicate via the communication bus. It may not include any programmable communication circuit for allowing communication via such a bus.

[0092] An example of such a design and its benefits are next described in the context of battery management systems. Notably, the design can be additionally or alternatively applied to other systems of a vehicle that perform functions other than battery management, such as motor and motor control.

[0093] Battery packs including multiple battery cells, such as lithium-ion cells, can be used in electric cars, electric aircraft, and other electric self- powered vehicles. The battery cells may be connected in series or in parallel to deliver an appropriate voltage and current. H55-18-PCT1

[0094] Battery cells in battery packs can be managed and controlled by battery management systems (BMS). A BMS can be a circuit that manages a rechargeable battery cell by controlling its charging and discharge cycles, preventing it from operating outside its safe operating area, balancing the charge between cells, or the like. BMS can also monitor battery parameters, such as the temperature, voltage, current, internal resistance, or pressure of the battery cell, and report anomalies. BMS can be provided by various manufacturers as discreet electronic components.

[0095] Damage to battery cells can be very serious incidents that may cause fire, explosions, or interruption of the powered circuit. Therefore, any damage to a battery in a vehicle, such as an electric aircraft, may desirably be reported immediately and reliably to the pilot or driver of the vehicle. A reliable monitoring of battery cells by BMS can be critical for the safety of electric airplanes.

[0096] However, BMS can have failings in rare occurrences that cause problems with battery cells which may not be reported correctly. For example, an overvoltage or overtemperature condition can, in some situations, affect not only a battery cell, but also its BMS, so that the failure of the battery cell is either not detected or not reported correctly. Even if the BMS functions correctly, a connecting bus between the BMS and the Cockpit might be defective and prevent warning signals from being transmitted.

[0097] In order to prevent this risk, battery cells can be monitored with a second, redundant BMS. If both BMS are of the same type, a defect or conception flaw that affects one BMS may also affect the redundant BMS as well, so that the gain in reliability can be limited. The present disclosure provides at least approaches to increase the reliability of the detection of malfunctions of battery cells in an electric vehicle, such as an electric aircraft. Redundant monitoring of parameters of each battery cell can be performed with two different circuits. Because a second, redundant monitoring circuit may include non-programmable, non-stateful components rather than processors, sequential logic electronic components, H55-18-PCT1or programmable combinational logic electronic components, its certification can be easier, and its reliability may be increased. For example, because the second, redundant circuit may be processorless, may not include any sequential or programmable combinational logic electronic components, and may not rely on any software (for example, executable program code that is executed by a processor), its certification is made easier than if the second, redundant circuit relied on processors, sequential or programmable combinational logic electronic components, or software.

[0098] The second, redundant monitoring circuit can provide for a redundant monitoring of battery parameters and for a redundant transmission of those parameters, or warning signals depending on those parameters. The second battery monitoring system may transmit analog or binary signals but not multivalued digital signals. The second battery monitoring circuit may not manage the charge and discharge of battery cells, but instead provide for monitoring of battery parameters, and transmission of parameters or warning signals. Therefore, the second, redundant battery monitoring circuit can be made simple, easy to certify, and reliable.

[0099] Fig.3 illustrates a battery monitoring system. This system can be used in an electric vehicle, such as an electric aircraft, a large size drone or unmanned aerial vehicle, an electric car, or the like, to monitor the state of battery cells 1 in one of multiple battery packs and report this state or generate warning signals in case of dysfunctions.

[0100] The battery cells 1 can be connected in series or in parallel to deliver a desired voltage and current. Fig.3 shows serially connected battery cells. The total number of battery cells 1 may exceed 100 cells in an electric aircraft. Each of the battery cells 1 can be made up of multiple elementary battery cells in parallel.

[0101] A first battery monitoring circuit can control and monitor the state of each battery cell 1. The first battery management circuit can include multiple BMSs 2, each of the BMSs 2 managing and controlling one H55-18-PCT1of the battery cells 1. The BMSs 2 can each be made up of an integrated circuit (for instance, a dedicated integrated circuit) mounted on one printed circuit board (PCB) of the PCBs 20. One of the PCBs 20 can be used for each of the battery cells 1 or for a group of battery cells. Fig.4 illustrates example components of one of the BMSs 2.

[0102] The control of a battery cell can include control of its charging and discharge cycles, preventing a battery cell from operating outside its safe operating area, or balancing the charge between different cells.

[0103] The monitoring of one of the battery cells 1 by one of the BMSs 2 can include measuring parameters of the one of the battery cells 1, to detect and report its condition and possible dysfunctions. The measurement of the parameters can be performed with battery cell parameter sensors, which can be integrated in the one of the BMSs 2 or connected to the one of the BMSs 2. Examples of such parameter sensors can include a temperature sensor 21, a voltage sensor 22, or a current sensor. An analog- to-digital converter 23 can convert the analog values measured by one or more of the parameter sensors into multivalued digital values, for example, 8 or 16 bits digital parameter values. A microcontroller 24, which can be part of each of the BMSs 2, can compare the values with thresholds to detect when a battery cell temperature, battery cell voltage, or battery cell current is outside a range.

[0104] The BMSs 2 as slaves can be controlled by one of multiple first master circuits 5. In the example of Fig.3, each of the first master circuits 5 can control four of the BMSs 2. Each of the first master circuits 5 can control eight of the BMSs 2, or more than eight of the BMSs 2. The first master circuits 5 can control more BMS and more battery cells in yet other implementations. The first master circuits 5 can be connected and communicate over a digital communication bus 55.

[0105] The first master circuits 5 can also be connected to a computer 9 that collects the various digital signals and data sent by the first master circuits 5, and may display information related to the battery state and H55-18-PCT1warning signals on a display 13, such as a matrix display. The display 13 may be mounted in the vehicle’s cockpit to be visible by the vehicle’s driver or pilot. Additionally or alternatively, the computer 9 can output the information to a computer remote from the aircraft or to control operations of one or more components of the aircraft as described herein.

[0106] The BMSs 2 can be connected to the first master circuits 5 over a digital communication bus, such as a CAN bus. A bus driver 25 can interface the microcontroller 24 with the digital communication bus and provide a first galvanic isolation 59 between the PCBs 20 and the first master circuits 5. In one example, the bus drivers of adjacent BMSs 2 can be daisy chained. For example, as shown in Fig.4, the bus driver 25 is connected to the bus driver 27 of the previous BMS and to the bus driver 28 of the next BMS.

[0107] Each of the BMSs 2 and their associated microcontrollers can be rebooted by switching its power voltage Vcc. The interruption of Vcc can be controlled by the first master circuits 5 over the digital communication bus and a power source 26.

[0108] Fig.6 illustrates example components of one of the first master circuits 5. The one of the first master circuits 5 can include a first driver 51 for connecting the one of the first master circuits 5 with one of the BMSs 2 over the digital communication bus, a microcontroller 50, and a second driver 52 for connecting the first master circuits 5 between themselves and with the computer 9 over a second digital communication bus 55, such as a second CAN bus. A second galvanic isolation 58 can be provided between the first and second master circuits 5, 7 and the computer 9.32ulfild galvanic isolation 58 can, for example, be 1500 VDC, 2500 Vrms, 3750 Vrms, or another magnitude of isolation. The microcontroller 50, the first driver 51, and the second driver 52 can be powered by a powering circuit 53 and may be mounted on a PCB 54, one such PCB can be provided for each of the first master circuits 5.

[0109] Fig.3 further illustrates a second battery monitoring circuit, which can be redundant of the first battery monitoring circuit. This second H55-18-PCT1battery monitoring circuit may not manage the battery cells 1; for example, the second battery monitoring circuit may not control charge or discharge cycles of the battery cells 1. The function of the second battery monitoring circuit can instead be to provide a separate, redundant monitoring of each of the battery cells 1 in the battery packs, and to transmit those parameters or warning signals related to those parameters, such as to the pilot or driver or a computer aboard or remote from the aircraft as described herein. The second battery monitoring circuit can monitor the state of each of the battery cells 1 independently from the first battery monitoring circuit. The second battery monitoring circuit can include one of multiple cell monitoring circuits 3 for each of the battery cells. The parameters or warning signals may moreover, for example, be used by the second battery monitoring circuit to stop charging (for instance, by opening a relay to disconnect supply of power) of one or more battery cells when the one or more battery cells may be full of energy and a computer of the aircraft continues to charge the one or more battery cells.

[0110] Fig.5 illustrates example components of one of the cell monitoring circuits 3. Each of the cell monitoring circuits 3 can include multiple cell parameter sensors 30, 31, 32, 33 for measuring various parameters of one of the battery cells 1. The sensor 30 can measure a first temperature at a first location in one battery cell and detect an overtemperature condition; the sensor 31 can measure a second temperature at a second location in the same battery cell and detect an overtemperature condition; the sensor 32 can detect an undervoltage condition in the same battery cell; and the sensor 33 can detect an overvoltage condition on the same battery cell. The undervoltage condition can be detected, for example, when the voltage at the output of one battery cell is under 3.1 Volts or another threshold. The overvoltage condition might be detected, for example, when the voltage at the output of one battery cell is above 4.2 Volts or yet another threshold. The thresholds used can depend, for instance, on the type of battery cell 1 or a number of elementary cells in the cell. Therefore, each or some of the sensors 30-33 can include a sensor as such and an analog comparator for comparing the value delivered by the sensor with one or two thresholds, H55-18-PCT1and outputting a binary value depending on the result of the comparison. Other battery cell parameter sensors, such as an overcurrent detecting sensor, can be used in other implementations.

[0111] Various parameters related to one of the battery cells 1 can be combined using a combinational logic circuit 35, such as an AND gate. The combinational logic circuit 35 may not include programmable logic. In the example of Fig.5, binary signals output by the sensors 30, 31, and 32 are combined by a AND gate into a single warning signal, which can have a positive value (warning signal) if and only if the temperature measured by the two temperature sensors exceeds a temperature threshold and if the voltage of the cell is under a voltage threshold. The detection of an overvoltage condition by the sensor 33, in the example of Fig.5, may not combined with any other measure and can be directly used as a warning signal.

[0112] The warning signals delivered by the combinational logic 35 or directly by the parameter sensors 30-33 can be transmitted to a second master circuit 7 over lines 76, which can be dedicated and different from the digital communication bus used by the first battery monitoring circuit. Optocouplers 36, 37, 38 provide a third galvanic isolation 60 between the components 30-38 and the second master circuit 7. The third galvanic isolation 60 can provide the same isolation as the first galvanic isolation 59, such as 30V isolation, or the third galvanic isolation 60 may provide a different isolation form the first galvanic isolation 59.

[0113] The sensors 30-33 and the combinational logic element 35 can be powered by a powering circuit 34 that delivers a power voltage Vcc2. This powering circuit 34 can be reset from the second master circuit 7 using an ON / OFF signal transmitted over the optocoupler 38.

[0114] The sensors 30-33 and the combinational logic element 35 can be mounted on a PCB. One such PCB can be provided for each of the battery cells 1. The sensors 30-33 and the combinational logic element 35 can be H55-18-PCT1mounted on the same PCB 20 as one of the BMSs 2 of the first battery monitoring circuit.

[0115] Fig.7 illustrates example components of one of the second master circuits 7. In the example of Fig.5, the one of the second master circuits 7 can include a combinational logic element 72, which may not include programmable logic, for combining warning signals, such as overtemperature / under- voltage warning signals uv1, uv2, ... or overvoltage signals ov1, ov2, ... from different battery cells into combined warning signals, such as a general uv (undervoltage condition in case of overtemperature) warning signal and a separate overvoltage warning signal ov. Those warning signals uv, ov can be active when any of the battery cells 1 monitored by the one of the second master circuits 7 has a failure. They can be transmitted over optocouplers 70, 71 and lines 76 to the next and previous second master circuits 74, 75, and to a warning display panel 11 in the cockpit of the vehicle for displaying warning signals to the driver or pilot. The warning display panel 11 can include lights, such as light emitting diodes (LEDs), for displaying warning signals.

[0116] With the disclosed design of the cell monitoring circuits 3 and the second master circuits 7, no dormant alarms may remain undetected. For example, if a cable may be broken or a power supply is inactive, the warning panel 11 can correctly show an alarm despite the broken cable or the inactive power supply. This can be accomplished, for instance, by using an inverted logic so that if the warning panel 11 does not receive a voltage or a current on an alarm line, an indicator may activate, but if the warning panel 11 does receive the voltage or the current on the alarm line, the indicator can deactivate.

[0117] The one of the second master circuits 7 can be mounted on a PCB. One such PCB can be provided for each of the second master circuits 7. One of the second master circuits 7 can be mounted on the same PCB 54 as one of the first master circuits 5 of the first battery monitoring circuit. H55-18-PCT1

[0118] As can be seen, the second battery monitoring circuit can include exclusively non-programmable, non-stateful components (such as, analog components or non-programmable combinational logic components). The second battery monitoring circuit can be processorless, and may not include any sequential or programmable combinational logic. The second battery monitoring circuit may not run any computer code or be programmable. This simplicity can provide for a very reliable second monitoring circuit, and for a simple certification of the second battery monitoring circuit and an entire system that include the second battery monitoring circuit.

[0119] The second battery monitoring circuit can be built so that any faulty line, components, or power source triggers an alarm. In one example, an “0” on a line, which may be caused by the detection of a problem in a cell or by a defective sensor, line, or electronic component, can be signalled as an alarm on the warning panel; the alarm may only be removed when all the monitored cells and all the monitoring components are functioning properly. For example, if the voltage comparator or temperature sensor is broken, an alarm can be triggered.

[0120] The computer 9, the display 13, and the warning display panel 11 in the cockpit can be powered by a power source 15 in the cockpit, which may be a cockpit battery and can be independent of other power sources used to power one or more other components. Monitoring and warning about failure conditions in motors or motor controllers for electric and hybrid aircrafts

[0121] As indicated, the aspects, blocks and circuits that have been described so far in the context of battery monitoring systems could apply to monitoring and warning for different components of an electric or hybrid aircraft. For example, a first monitoring and warning subsystem could be used for detecting an warning catastrophic, or hazardous, failure conditions of a motor or motor controller, while a second subsystem could be used for redundant monitoring of those catastrophic or hazardous failure conditions, and / or for monitoring and warning about less serious H55-18-PCT1failure conditions, such as major, minor or no safety risk conditions of an electric motor or motor controller. The first monitoring and warning subsystem can be composed of non-programmable, non-stateful components and thus easier to certify, while the second monitoring and warning subsystem can comprise a processor or other programmable components, and output information to a sophisticated display 13, via a computer 9, for presentation. Motor and Battery System

[0122] Battery packs including multiple battery cells, such as lithium-ion cells, can be used in electric cars, electric aircraft, and other electric self- powered vehicles. The battery cells can be connected in series or in parallel to deliver an appropriate voltage and current.

[0123] In electrically driven aircraft, the battery packs can be chosen to fulfil the electrical requirements for various flight modes. During short time periods like take off, the electrical motor can utilize a relatively high power. During most of the time, such as in the standard flight mode, the electrical motor can utilize a relatively lower power, but may consume a high energy for achieving long distances of travel. It can be difficult for a single battery to achieve these two power utilizations.

[0124] The use of two battery packs with different power or energy characteristics can optimize the use of the stored energy for different flight conditions. For example, a first battery pack can be used for standard flight situations, where high power output may not be demanded, but a high energy output may be demanded. A second battery pack can be used, alone or in addition to the first battery pack, for flight situations with high power output demands, such as take-off manoeuvring.

[0125] An electrical powering system can charge the second battery pack from the first battery pack. This can allow recharging of the second battery pack during the flight, subsequent to the second battery pack being used in a high power output demanding flight situation. Therefore, H55-18-PCT1the second battery pack can be small, which can save space and weight. In addition, this can allow different battery packs for different flight situations that optimize the use of the battery packs.

[0126] The electrical powering system can also charge the second battery pack by at least one motor which works as generator (the motor may also accordingly be referred to as a transducer). This can allow recharging of the second battery pack during the flight or after the second battery pack has been used in a high power output demanding flight situation. Therefore, the second battery pack can be small, which can save space and weight. In addition, the different battery packs can allow the recovery of braking energy. Braking energy during landing or sinking recovered by a generator motor can create high currents which may not be recovered by battery packs used for traveling long distances. By using a second battery pack suitable for receiving high power output in a short time, more braking energy can be recovered via the second battery pack than the first battery pack, for example.

[0127] The electrical powering system can also include a third battery pack, which includes a supercapacitor. Because supercapacitors can receive and output large instantaneous power or high energy in a short duration of time, the third battery pack can further improve the electrical powering system in some instances. A supercapacitor may, for example, have a capacitance of 0.1 F, 0.5 F, 1 F, 5 F, 10 F, 50 F, 100 F, or greater or within a range defined by one of the preceding capacitance values.

[0128] Figs.8 to 13 illustrate multiple electrical power systems.

[0129] Fig.8 shows an electrical powering system that includes a first battery pack 91, a second battery pack 92, a circuit 90, and at least one motor 94.

[0130] The first battery pack 91 and the second battery pack 92 can each store electrical energy for driving the at least one motor 94. The first battery pack 91 and the second battery pack 92 can have different electrical H55-18-PCT1characteristics. The first battery pack 91 can have a higher energy capacity per kilogram than the second battery pack 92, and the first battery pack 91 can have a higher power capacity (watt hours) than the second battery pack 92. Moreover, the first battery pack 91 can have a lower maximum, nominal, or peak power than the second battery pack 92; the first battery pack 91 can have a lower maximum, nominal, or peak current than the second battery pack 92; or, the first battery pack 91 can have a lower maximum, nominal, or peak voltage than the second battery pack 92. More than one or even all of the mentioned electrical characteristics of the first battery pack 91 and the second battery pack 92 can be different. However, only one of the mentioned electrical characteristics may be different or at least one other characteristic than the mentioned electrical characteristics may be different. The first battery pack 91 and the second battery pack 92 can have the same electrical characteristics.

[0131] The type or the material composition of the battery cells of the first battery pack 91 and the second battery pack 92 can be different. The type or the material composition of the battery cells of the first battery pack 91 and the second battery pack 92 can be the same, but an amount of copper or an arrangement of conductors can be different. In one example, the first battery pack 91 or the second battery pack 92 can be a lithium-ion (Li-ion) battery or a lithium-ion polymer (Li-Po) battery. The second battery pack 92 may include a supercapacitor (sometimes referred to as a supercap, ultracapacitor, or Goldcap).

[0132] The first battery pack 91 can include relatively high energy- density battery cells that can store a high amount of watt-hours per kilogram. The first battery pack 91 can include low power battery cells. The first battery pack 91 can provide a DC voltage / current / power or can be connected by a (two phase or DC) power line with the circuit 90.

[0133] The second battery pack 92 can include relatively low energy- density battery cells. The second battery pack 92 can include relatively high power battery cells. The second battery pack 92 can provide a DC H55-18-PCT1voltage / current / power or is connected by a (two phase or DC) power line with the circuit 90.

[0134] The first battery pack 91 can form an integrated unit of mechanically coupled battery modules or the first battery pack 91 may be an electrically connected first set of battery modules. Similarly, the second battery pack 92 can form an integrated unit of mechanically coupled battery modules or the second battery pack 92 may be an electrically connected second set of battery modules. Some or all of the battery modules of each of first battery pack 91 or the second battery pack 92 can be stored in one or more areas of a housing of an aircraft, such as a within a wing or nose of the aircraft.

[0135] The first battery pack 91 can have a total energy capacity that exceeds a total energy capacity of the second battery pack 92. For example, a ratio of the total energy capacity of the first battery pack 91 and the total energy capacity of the second battery pack 92 can be 2:1, 3:1, 4:1, 5:1, 10:1, 20:1, 40:1, or 100:1 or within a range defined by two of the foregoing ratios.

[0136] The electrical powering system can include an external charging interface for charging the first battery pack 91 or the second battery pack 92 when the aircraft is on the ground and connected to a charging station outside of the aircraft.

[0137] Each, some, or one of the at least one motor can be an electrical motor. The at least one motor 94 can be connected to the circuit 90. The at least one motor 94 can receive over the circuit 90 electrical energy / power from the first battery pack 91 or the second battery pack 92 to drive the at least one motor 94. For example, the at least one motor 94 can be a three phase motor, such as a brushless motor, which is connected via a three phase AC power line with the circuit 90. However, the at least one motor 94 can instead be a different type of motor, such as any type of DC motor or a one phase AC motor. The at least one motor 94 can move a vehicle, such as an airborne vehicle like an aircraft. The at least one motor 94 can H55-18-PCT1drive a (thrust-generating) propeller or a (lift-generating) rotor. In addition, the at least one motor 94 can also function as a generator. The electrical powering system or the at least one motor 94 can include two or more electrical motors as described further herein.

[0138] The different motors of the at least one motor 94 can have the same or different characteristics. The at least one motor 94 can be a motor with a first set of windings connected with a first controller 96 and with a second set of windings connected with a second controller 97, as shown for example in Fig.12. This can allow use of the at least one motor 94 as generator and motor at the same time or to power the at least one motor 94 from the first controller 96 and the second controller 97. The at least one motor 94 can include a first motor 98 and a second motor 99 as shown for example in Figs.11 and 13. The first and the second motor 98 and 99 can be mechanically connected such that the rotors of the first and second motor 98 and 99 are mechanically coupled, for instance for powering both the same propeller or rotor (as shown in Figs.11 and 13). The first and the second motor 98 and 99 can, for example, drive the same axis which rotates the propeller or rotor. However, the first and second motor 98 and 99 may not be mechanically coupled and may drive two distinct propellers or rotors. The at least one motor 94 can include more than two motors M1, M2, … Mi which are mutually connected, or multiple mutually connected motors.

[0139] The circuit 90 can be connected with the first battery pack 91, the second battery pack 92, and the at least one motor 94.

[0140] The circuit 90 can include a controller 93 connected with the first battery pack 91, the second battery pack 92, and the at least one motor 94. The controller 93 can, for example, be connected over a two phase or DC power line with the first battery pack 91 and the second battery pack 92 or connected over a three phase power line with the at least one motor 94. The controller 93 can transform, convert, or control the power received from the first battery pack 91 or the second battery pack 92 into motor driving signals for driving the at least one motor 94. The controller 93 can H55-18-PCT1include a power converter for converting the DC current of the first battery pack 91 or the second battery pack 92 into a (three phase) (AC) current for the at least one motor 94 (power converter working as inverter). The power converter can treat different input DC voltages (if the first battery pack 91 and the second battery pack 92 have different DC voltages). If the at least one motor 94 acts as generator, the power converter can convert the current generated from each phase of the at least one motor 94 into a DC current for loading the first battery pack 91 or the battery pack 92 (power converter working as rectifier). The controller 93 can create the motor driving signals for the at least one motor 94 based on user input.

[0141] The controller 93 can include more than one controller. The controller 93 can include, for instance, a first controller 96 for powering the at least one motor 94 from at least one of the first battery pack 91 and the second battery pack 92 and a second controller 97 for powering the at least one motor 94 from at least one of the first battery pack 91 or the second battery pack 92. The features described for the controller 93 can apply to the first controller 96 or the second controller 97. Examples of such a circuit are shown in the Figs.10 to 13. In Figs.10 to 12, the first controller 96 powers the at least one motor 94 from the first battery pack 91 and the second controller 97 powers the at least one motor 94 from the second battery pack 92. The first controller 96 and the second controller 97 can power the at least one motor 94 as shown in Fig.10 or the at least one motor 94 with different driving windings (or poles) as shown in Fig.12.

[0142] As shown in Figs.11 and 13, the first controller 96 can drive a first motor 98 and the second controller 97 can drive a second motor 99. The first controller 96 and the second controller 97 can be flexible and drive the first motor 98 or the second motor 99 depending on a switching state of a switch 101 as shown in Fig.13. The first controller 96 and the second controller 97 can be different. For example, the input DC voltage of the first controller 96 and the second controller 97 from the first battery pack 91 and the second battery pack 92 can be different. However, the first controller 96 and the second controller 97 can instead be identical. H55-18-PCT1

[0143] The circuit 90 can select from at least two of the following connection modes. In a first connection mode, the first battery pack 91 can be electrically connected over the controller 93 with the at least one motor 94, while the second battery pack 92 may be electrically disconnected from the at least one motor 94. In the first connection mode, power can flow between the at least one motor 94 and the first battery pack 91, but may not flow between the at least one motor 94 and the second battery pack 92. In a second connection mode, the second battery pack 92 can be electrically connected over the controller 93 with the at least one motor 94, while the first battery pack 91 may be electrically disconnected from the at least one motor 94. In the second connection mode, power can flow between the at least one motor 94 and the second battery pack 92, but may not between the at least one motor 94 and the first battery pack 91. In a third connection mode, the first battery pack 91 and the second battery pack 92 can be electrically connected over the controller 93 with the at least one motor 94. In the third connection mode, power can flow between the at least one motor 94 and the first battery pack 91 and the second battery pack 92. Electrical switches can be used to perform this selection between different connection modes, and the electrical switches can be between the controller 93 and first battery pack 91 and the second battery pack 92, in the controller 93, or between the controller 93 and the at least one motor 94. If the at least one motor 94 has more than one motor, there can be further connection modes. The first battery pack 91 can be connected with the first motor 98 and not the second motor 99 (fourth connection mode) or with the second motor 99 and not the first motor 98 (fifth connection mode) or with the first motor 98 and the second motor 99 (sixth connection mode). The second battery pack 92 can be connected with the first motor 98 and not the second motor 99 (seventh connection mode) or with the second motor 99 and not the first motor 98 (eighth connection mode) or with the first motor 98 and the second motor 99 (ninth connection state). The first battery pack 91 and the second battery pack 92 can be connected with the first motor 98 and not the second motor 99 (tenth connection mode) or with the second motor 99 and not the first motor 98 (eleventh connection mode) or with the first motor 98 and the second motor 99 (twelfth connection state). The numbering of the H55-18-PCT1connection modes can be arbitrarily chosen. If there may additionally be a third battery pack, there can be correspondingly more potential connection modes between the at least one motor and the three battery packs.

[0144] The circuit 90 can select from at least two of the following drive modes. In a first drive mode, the at least one motor 94 can be driven by the first battery pack 91 (without using the power of the second battery pack 92). In this first drive mode (which may be referred to as a standard drive mode), the circuit 90 can be in the first connection mode. Alternatively, in the first drive mode, the circuit 90 can also be in the third connection mode, while no power flows from the second battery pack 92 to the at least one motor 94. This standard drive mode can be used when the power consumption of the least one motor 94 may be low, such as during steady flight conditions, gliding flight, or landing of an aircraft. In a second drive mode (which may be referred to as a high energy drive mode), the at least one motor 94 can be driven by the second battery pack 92 (without using the power of the first battery pack 91). In this second drive mode, the circuit 90 can be in the second connection mode. Alternatively, in the second drive mode, the circuit 90 can also be in the third motor connection mode, while no power flows from the first battery pack 91 to the at least one motor 94. This second drive mode can be used when the power consumption of the at least one motor 94 may be high, such as during manoeuvring, climb flight, or take off. In a third drive mode (which may be referred to as a very high energy drive mode), the at least one motor 94 can be simultaneously driven by the first battery pack 91 and the second battery pack 92. In this third drive mode, the circuit 90 can be in the third connection mode. This third drive mode can be used when the power consumption of the least one motor 94 may be high, such as during manoeuvring, climb flight, or take off.

[0145] The circuit 90 can include a detector for detecting the power requirements of a present flight mode. The detection can be performed from user input or sensor measurements, such as by measuring the current in the motor input line. The circuit 90 can select the drive mode or the connection mode based at least on the detection result of this detector. H55-18-PCT1

[0146] The selection between connection modes can depend at least on the charging level of the different battery packs. For example, a high- power battery pack can be used instead, or in addition to, a high energy- density battery pack when the charge of the high energy-density battery pack is low.

[0147] The electrical powering systems of Figs.8 to 13 can be configured such that the second battery pack 92 can be charged from the first battery pack 91, such as via the circuit 90. Moreover, the electrical powering systems can be configured such that the second battery pack 92 can be charged from the first battery pack 91 while the first battery pack 91 powers or drives the at least one motor 94.

[0148] In Figs.9 to 11, the circuit 90 can electrically connect the first battery pack 91 and the second battery pack 92 for charging. The connection can be steady or realized by a switch which switches between a first battery connection mode in which the first battery pack 91 and the second battery pack 92 are electrically connected and a second battery connection mode in which the first battery pack 91 and the second battery pack 92 are electrically disconnected. As explained further herein, the first battery connection mode can be realized by connecting the first battery pack 91 and the second battery pack 92 over a charging circuit 95 or over the controller 93 or over one or more other controllers.

[0149] In Fig.9, the circuit 90 the charging circuit 95 for charging the second battery pack 92 from the first battery pack 91. The charging circuit 95 can control energy flow from the first battery pack 91 to the second battery pack 92 and may transfer the energy without transferring the energy through the controller 93. The charging circuit 95 can include a switch (not shown) for connecting the first battery pack 91 with the second battery pack 92 for charging. Such a switch may have the advantage that the charging process can be controlled by a user or by a microprocessor. For example, if the full power of the first battery pack 91 is desired to power the at least one motor 94, the process of charging the second battery pack 92 may automatically be interrupted. However, the charging circuit 95 can H55-18-PCT1instead work switchless so that the process of charging automatically starts when a certain electrical parameter, like the voltage or capacitance of the second battery pack 92, falls below a certain threshold.

[0150] If the voltage of the first battery pack 91 and the second battery pack 92 may be different, the charging circuit 95 can include a DC / DC converter for converting the DC voltage of the first battery pack 91 into the DC voltage of the second battery pack 92. The second battery pack 92 can be charged from the first battery pack 91 at the same time that the at least one motor 94 is driven by the first battery pack 91 or at a time that the at least one motor 94 is not powered, such as by the first battery pack 91.

[0151] In Fig.10, the second battery pack 92 can be charged over the first controller 96 and the second controller 97. The first battery pack 91 can provide energy and power for the first controller 96, which can convert this energy and power into the electrical driving signals for the at least one motor 94. For charging the second battery pack 92, the electrical driving signals from the first controller 96 can be converted by the second controller 97 into the charging signal (DC voltage) for the second battery pack 92. The electrical driving signals for the at least one motor 94 from the first controller 96 can be used for charging the second battery pack 92 and for driving the at least one motor 94 at the same time. This can allow the second battery pack 92 to charge from the first battery pack 91 at the same time that the at least one motor 94 may be driven by the electrical driving signals from the first controller 96. The second battery pack 92 can however instead be charged by the electrical drive signals without powering the motor at the same time.

[0152] Instead of or in addition to electrically connecting the first battery pack 91 with the second battery pack 92 for transferring electrical energy from the first battery pack 91 to the second battery pack 92, the first battery pack 91 can be mechanically connected with the second battery pack 92 for transferring mechanical energy to charge the second battery pack 92 from the first battery pack 91. H55-18-PCT1

[0153] In Fig.11, mechanical charging can be realized by driving the first motor 98 from the first battery pack 91 (over the first controller 96) and generating energy from the second motor 99 which is mechanically connected to the first motor 98 and working as generator. The energy generated by the second motor 99 can be used to charge the second battery pack 92 (by converting the generated motor signals of the second motor 99 via the second controller 97 into the charging signal (DC voltage) of the second battery pack 92). This can allow the second battery pack 92 to charge from the first battery pack 91 at the same time that the at least one motor 94 is driven by the energy from the first battery pack 91.

[0154] In Fig.12, mechanical charging can be realized by driving the at least one motor 94 from the first battery pack 91 (such as over the first controller 96) with the first set of windings of the at least one motor 94 and generating energy from the at least one motor 94 over the second set of windings of the at least one motor 94 which can function as a generator. The energy generated by the second set of windings can be used to charge the second battery pack 92 by converting the generated motor signals of the at least one motor 94 via the second controller 97 into the charging signal (DC voltage) of the second battery pack 92. This can allow the second battery pack 92 to charge from the first battery pack 91 at the same time that the at least one motor 94 is driven by the energy from the first battery pack 91. Moreover, this can enable the second battery pack 92 to charge from the first battery pack 91 without utilizing separate circuitry, such as a DC / DC converter, which would increase a weight of the aircraft.

[0155] Fig.13 shows a switch 101 which can select from different battery packs or connection modes as described herein. This can allow the first battery pack 91 to connect with the second battery pack 92 (first battery connection mode) to charge the second battery pack 92 directly from the first battery pack 91. This can allow the first battery pack 91 to connect with (i) one of the first controller 96 or the second controller 97, (ii) one of the first motor 98 or second motor 99 and the second battery pack 92 with the other of the first controller 96 or the second controller 97, or (iii) the first motor 98 and the second motor 99 to charge the second battery pack H55-18-PCT192 mechanically. This can allow for selection of the first motor 98 or the second motor 99 to be driven by the first battery pack 91 or the second battery pack 92.

[0156] The design of Fig.13 can give the flexibility to choose among electrical charging or mechanical charging.

[0157] The second battery pack 92 can be charged by the at least one motor 94 which can work as a generator. When the at least one motor 94 may work as a generator, the generation can be driven by braking energy, such as during descent or landing of the aircraft. The second battery pack 92 can as a result recover energy without affecting the functioning of the first battery pack 91 for long distances. When the at least one motor 94 may work as a generator, the generation can be driven from the first battery pack 91 to charge the second battery pack 92. The second battery pack 92 can be charged by the at least one motor 94 working as a generator while the same motor or another motor of the at least one motor 94 can be driven by the energy from the first battery pack 91, such as for instance described with respect to Figs.11, 12, and 13.

[0158] The electrical powering system can include a third battery pack (not shown). The second battery pack 92 and the third battery pack can have different electrical characteristics. The second battery pack 92 can, for instance, have a higher energy capacity than the third battery pack. The second battery pack 92 can have a higher energy density than the third battery pack. The second battery pack 92 can have a lower maximum, nominal, or peak power than the third battery pack. The second battery pack 92 can have a lower maximum, nominal, or peak current than the third battery pack. The second battery pack 92 can have a lower maximum, nominal, or peak voltage than the third battery pack. The type or the material composition of the battery cells of the second battery pack 92 and the third battery pack can be different or the same. The third battery pack can include a supercapacitor. The third battery pack can increase a maximum power that may be delivered or recovered by the electrical powering system. The power recovered by the at least one motor 94 acting H55-18-PCT1as a generator from a braking action can, for example, immediately be recovered in the third battery pack up to a high recover power level. The third battery pack can be charged from the first battery pack 91 or the second battery pack 92, such as even while the at least one motor 94 may be driven from the power of the first battery pack 91 or the second battery pack 92. Modular Battery System

[0159] The power sources in an electric or hybrid aircraft can be modular and distributed to optimize a weight distribution or select a center of gravity for the electric or hybrid aircraft, as well as maximize a use of space in the aircraft. Moreover, the batteries in an electric or hybrid aircraft can desirably be designed to be positioned in place of a combustion engine so that the aircraft can retain a similar shape or structure to a traditional combustion powered aircraft and yet may be powered by batteries. In such designs, the weight of the batteries can be distributed to match that of a combustion engine to enable the electric or hybrid aircraft to fly similarly to the traditional combustion powered aircraft.

[0160] Fig.14A illustrates a battery module 1400 usable in an aircraft, such as the aircraft 100 of Figs.1A and 1B. The battery module 1400 can include a lower battery module housing 1410, a middle battery module housing 1420, an upper battery module housing 1430, and a multiple battery cells 1440. The multiple battery cells 1440 can together provide output power for the battery module 1400. The lower battery module housing 1410, the middle battery module housing 1420, or the upper battery module housing 1430 can include slots, such as slots 1422, that are usable to mechanically couple the lower battery module housing 1410, the middle battery module housing 1420, or the upper battery module housing 1430 to one another or to another battery module. Supports, such as supports 1424 (for example, pins or locks), can be placed in the slots to lock the lower battery module housing 1410, the middle battery module housing 1420, or the upper battery module housing 1430 to one another or to another battery module. H55-18-PCT1

[0161] The battery module 1400 can be constructed so that the battery module 1400 is evenly cooled by air. The multiple battery cells 1440 can include 16 total battery cells where the battery cells are each substantially shaped as a cylinder. The lower battery module housing 1410, the middle battery module housing 1420, or the upper battery module housing 1430 can be formed of or include plastic and, when coupled together, have an outer shape substantially shaped as a rectangular prism. The lower battery module housing 1410, the middle battery module housing 1420, or the upper battery module housing 1430 can together be designed to prevent a fire in the multiple battery cells 1440 from spreading outside of the battery module 1400.

[0162] The battery module 1400 can have a length of L1, a width of W, and a height of H1. The length of L1, the width of W, or the height of H1 can each be 50 mm, 65 mm, 80 mm, 100 mm, 120 mm, 150 mm, 200 mm, 250 mm or within a range defined by two of the foregoing values or another value greater or less than the foregoing values.

[0163] Fig.14B illustrates an exploded view of the battery module 1400 of Fig.14A. In the exploded view, a plate 1450 and a circuit board assembly 1460 of the battery module 1400 is shown. The plate 1450 can be copper and may electrically connect the multiple battery cells 1440 in parallel with one another. The plate 1450 may also distribute heat evenly across the multiple battery cells 1440 so that the multiple battery cells 1440 age at the same rate. The circuit board assembly 1460 may transfer power from or to the multiple battery cells 1440, as well as include one or more sensors for monitoring a voltage or a temperature of one or more battery cells of the multiple battery cells 1440. The circuit board assembly 1460 may or may not provide galvanic isolation to the battery module 1400 with respect to any components that may be electrically connected to the battery module 1400. Each of the multiple battery cells 1440 can have a height of H2, such as 30 mm, 50 mm, 65 mm, 80 mm, 100 mm, 120 mm, 150 mm or within a range defined by two of the foregoing values or another value greater or less than the foregoing values. H55-18-PCT1

[0164] Fig.15A illustrate a power source 1500A formed of multiple battery modules 1400 of Figs.14A and 14B. The multiple battery modules 1400 of the power source 1500A can be mechanically coupled to one another. A first side of one battery module 1400 can be mechanically coupled to a first side of another battery module 1400, and a second side of the one battery module 1400 that is opposite the first side can be mechanically coupled to a first side of yet another battery module 1400. The multiple battery modules 1400 of the power source 1500A can be electrically connected in series with one another. As illustrated in Fig.15A, the power source 1500A can include seven of the battery modules 1400 connected to one another. The power source 1500A may, for example, have a maximum power output between 1 kW and 60 kW during operation, a maximum voltage output between 10 V and 120 V during operation, or a maximum current output between 100 A and 500 A during operation.

[0165] The power source 1500A can include a power source housing 1510 mechanically coupled to at least one of the battery modules. The power source housing 1510 can include an end cover 1512 that covers a side of the power source housing 1510. The power source housing 1510 can have a length of L2, such as 3 mm, 5 mm, 10 mm, 15 mm, 20 mm, 25 mm, 30 mm, 40 mm, 50 mm or within a range defined by two of the foregoing values or another value greater or less than the foregoing values. The width and the height of the power source housing 1510 can match the length of L1 and the width of W of the battery module 1400.

[0166] The power source 1500A can include power source connectors 1520. The power source connectors 1520 can be used to electrically connect the power source 1500A to another power source, such as another of the power source 1500A.

[0167] Fig.15B illustrates a power source 1500B that is similar to the power source 1500A of Fig.15A but with the end cover 1512 and the upper battery module housings 1430 of the battery modules 1400 removed. Because the end cover 1512 has been removed, a circuit board assemblyH55-18-PCT11514 can be electrically coupled to the battery modules 1400. The circuit board assembly 1514 can additionally provide galvanic isolation (for instance, 2500 Vrms) for the power source 1500B with respect to any components that may be electrically connected to the power source 1500B. The inclusion of galvanic isolation in this manner may, for instance, enable grouping of the battery modules 1400 together so that isolation may be provided to the grouping of the battery modules 1400 rather than individual modules of the battery modules 1400 or a subset of the battery modules 1400. Such an approach may reduce the costs of construction because isolation can be expensive, and a single isolation may be used for multiple of the battery modules 1400.

[0168] Fig.16 illustrates a group 1600 of multiple power sources 1500A of Fig.15A arranged and connected for powering an aircraft, such as the aircraft 100 of Figs.1A and 1B. The multiple power sources 1500A of the group 1600 can be mechanically coupled to or stacked on one another. The multiple power sources 1500A of the group 1600 can be electrically connected in series or parallel with one another, such as by a first connector 1610 or a second connector 1620 that electrically connects the power source connectors 1520 of two of the multiple power sources 1500A. As illustrated in Fig.16, the group 1600 can include 10 power sources (for instance, arranged in a 5 row by 2 column configuration). In other examples, a group may include a fewer or greater number of power sources, such as 2, 3, 5, 7, 8, 12, 15, 17, 20, 25, 30, 35, or 40 power sources.

[0169] The grouping of the multiple power sources 1500A to form the group 1600 or another different group may allow for flexible configurations of the multiple power sources 1500A to satisfy various space or power requirements. Moreover, the grouping of the multiple power sources 1500A to form the group 1600 or another different group may permit relatively easy or inexpensive replacement of one or more of the multiple power sources 1500A in the event of a failure or other issue.

[0170] Fig.17A illustrates a perspective view of a nose 1700 of an aircraft, such as the aircraft 100 of Figs.1A and 1B, that includes multiple H55-18-PCT1power sources 1710, such as multiple of the power source 1500A, for powering a motor 1720 that operates a propeller 1730 of the aircraft. The multiple power sources 1710 can be used to additionally or alternatively power other components of the aircraft. The multiple power sources 1710 can be sized and arranged to optimize a weight distribution and use of space around the nose 1700. The motor 1720 and the propeller 1730 can be attached to and supported by a frame of the aircraft by supports, which can be steel tubes, and connected by multiple fasteners, which be bolts with rubber shock absorbers. A firewall 1740 can provide barrier between the multiple power sources 1710 and the frame of the aircraft in the event of a first at the multiple power sources 1710. An enclosure composed of glass fiber, metal, or mineral composite can be around the multiple power sources 1710 to protect from water, coolant, or fire.

[0171] Fig 17B illustrates a side view of the nose 1700 of Fig.17A.

[0172] Fig.18A illustrates a top view of a wing 1800 of an aircraft that includes multiple power sources 1810, such as multiple of the power source 1500A, for powering one or more components of the aircraft. The multiple power sources 1810 can be sized and arranged to optimize a weight distribution and use of space around the wing 1800. For example, the multiple power sources 1810 can be positioned within, between, or around horizontal support beams 1820 or vertical support beams 1830 of the wing 1800. A relay 1840 can further be positioned in the wing 1800 as illustrated and housed in a sealed enclosure. The relay 1840 may open if there is not a threshold voltage on a breaker panel or if a pilot opens breakers to shut down the multiple power sources 1810.

[0173] Fig 18B illustrates a perspective view of the wing 1800 of Fig. 18A. H55-18-PCT1Multi-Coil Motor Control

[0174] An electric or hybrid aircraft can be powered by a multi-coil motor, such as an electric motor, in which different coils of the motor power different phases of a modulation cycle for the motor.

[0175] As can be seen from Fig.19, a motor 1910 can include four different field coils (sometimes also referred to as coils) for generating a torque on a rotor of the motor 1910. The different field coils can include a first field coil 1902, a second field coil 1904, a third field coil 1906, and a fourth field coil 1908. Each of the different field coils can be independently powered by one or more controllers. The first field coil 1902, the second field coil 1904, the third field coil 1906, and the fourth field coil 1908 can be respectively powered by a first controller 1912, a second controller 1914, a third controller 1916, and a fourth controller 1918. One or more of the first controller 1912, the second controller 1914, the third controller 1916, and the fourth controller 1918 may be the same controller.

[0176] The first controller 1912, the second controller 1914, the third controller 1916, and the fourth controller 1918 can vary a current provided to individual coils of the first field coil 1902, the second field coil 1904, the third field coil 1906, and the fourth field coil 1908 to compensate for a failure of one or more (such as, one, two, or three) of the field coils. The first controller 1912, the second controller 1914, the third controller 1916, and the fourth controller 1918 may, for example, no longer provide current to a coil that has failed and provide additional current to one or more coils that have not yet failed. The first controller 1912, the second controller 1914, the third controller 1916, and the fourth controller 1918 can attempt to maintain a power output of the motor (for example, above a threshold) despite the failure of the one or more of the field coils.

[0177] The first controller 1912, the second controller 1914, the third controller 1916, or the fourth controller 1918 can determine the failure of one or more of the field coils from one or more sensors monitoring the motor or one or more individual field coils, such as proximate to the motor H55-18-PCT1or one or more individual field coils. The one or more sensors can include a temperature sensor, a current sensor, or a magnetic field sensor, among other types of sensors. For example, where the one or more sensors includes at least one temperature sensor, the first controller 1912, the second controller 1914, the third controller 1916, or the fourth controller 1918 can determine the failure of one or more of the field coils from a change in the temperature sensed by the temperature sensor (for instance, a temperature drop over time or proximate different field coils may correspond to a failure of a particular field coil or a number of field coils in the motor 1910). The first controller 1912, the second controller 1914, the third controller 1916, or the fourth controller 1918 may moreover attempt to operate the motor so that the temperature sensed remains constant within a tolerance. As another example, where the one or more sensors includes at least one voltage sensor, the first controller 1912, the second controller 1914, the third controller 1916, or the fourth controller 1918 can determine the failure of one or more of the field coils from a change in the voltage sensed by the voltage sensor (for instance, a voltage spike may correspond to a failure of a particular field coil or a number of field coils in the motor 1910). In another example, where the one or more sensors includes at least one magnetic field sensor, the first controller 1912, the second controller 1914, the third controller 1916, or the fourth controller 1918 can determine the failure of one or more of the field coils from a change in the resonance sensed by the magnetic field sensor.of the motor controller, theand the method for controlling the motor in an electric oraircraft in case of a failure in at least one of the field coils

[0178] Figs.20, 21 and 22 illustrate a motor 94 connected to a motor controller 93 in different arrangements.

[0179] Fig.20 illustrates a first battery pack 91 and a second battery pack 92 connected in series for providing a DC voltage VDC at the input side H55-18-PCT1of the motor controller 93. The motor controller 93 supplies the motor 94 on the output side with electrical energy.

[0180] The motor 94 in the following examples is configured as a three- phase permanent-magnet synchronous machine. Other motor types, such as synchronous machines, permanent-magnet, DC machines or induction machines, might be used instead. The motor 94 can propel the aircraft.

[0181] The motor controller 93 provides the motor 94 with three-phase alternating currents and voltages with varying amplitudes. The motor controller 93 is also configured to vary the fundamental frequency of the alternating currents and voltages provided to the motor 94.

[0182] Fig.21 illustrates a motor controller 93 with a DC voltage VDC at the input side and a motor 94 connected at the output side. The motor controller comprises multiple switches 931-933 configured to interrupt or arrange the connection between electrical circuits (not shown) comprised in the motor controller 93 and each motor phase individually. The electrical circuits can comprise power semiconductors for converting the DC voltage VDC at the input side into alternating quantities, such as alternating voltages and / or currents.

[0183] The switches 931-933 may be mechanical contactors. Other sorts of switches, such as semiconductor switches, might be used instead.

[0184] A control instance (not shown) is connected to the switches 931- 933 and configured to control said switches from a non-conductive state into a conductive state and vice versa. The control circuit may include digital components, including for example a processor, a FPGA circuit, and / or any combination of digital and / or analog components for controlling the switches. H55-18-PCT1

[0185] Fig.22 illustrates a motor controller 93 with the DC voltage Vdc at the input side and a motor 94 connected at the output side. The motor controller comprises multiple switches 931-933 configured to interrupt or arrange the connection between electrical circuits (not shown) comprised in the motor controller 93 and each motor phase individually.

[0186] In this example, the connection between the first phase and the motor 94 is interrupted by the first switch 931, whereas the remaining phases are connected using the corresponding switches 932, 933 in the phase lines.

[0187] A coil failure detection circuit or software (not illustrated) is arranged for detecting a failure in any of the field coils, and for isolating that defective field coil after such a detection. In the illustrated example, the coil failure detection circuit has detected a failure in the field coil of the first phase and has controlled the first switch 931 into a non-conductive state.

[0188] The remaining field coils in the remaining phases are functional. Thus, the motor controller 93 can provide those remaining phases with electrical energy.

[0189] In all illustrated examples of Figs.20 to 22, the propeller 970 is configured as a variable-pitch propeller that is mechanically connected to the rotor shaft. The blade pitch can be variably adjusted to generate more or less thrust.

[0190] Fig.23 illustrates a simplified control diagram for a motor controller 93 connected to the motor 94.

[0191] The motor controller includes an inverter circuit 945 configured to convert the DC voltage vdc at the input side into three-phase alternating currents and voltages at the output side. Three phase lines interconnect the H55-18-PCT1inverter circuit 945 of the motor controller 93 and the motor 94. Each of the three phase lines includes a switch 931-933 connected in series with each phase line. Each switch 931-933 is connected to the control circuit 950 and can be controlled by the control circuit 950 between a conductive and non-conductive state and vice versa. In this example, each switch 931-933 is in a conductive state as no failure in one of the field coils of the motor 94 has been detected.

[0192] In this example, each phase line further includes a plurality of current sensors 961-963 between the inverter circuit 945 and the switches 931-933, for measuring an electrical current flowing in each of the phase lines during operation of the motor 94. Each current sensor 961-963 is connected to the control circuit 950. The current sensors 961-963 may be Hall effect current transducers. Other types of current sensors, such as current transformers, Rogowski coils or shunt resistors, might be used instead.

[0193] The motor 94 may include an encoder 960 connected to the shaft of the motor for determining the rotor speed ωM*and the rotor position Θ. The encoder 960 is connected to the control circuit 950. The rotor speed ωM and the rotor position Θ can be alternatively provided and / or estimated by the control circuit 950 using a Position and Speed estimator 943. The Position and Speed estimator 943 calculates and / or estimates the rotor speed ωM and the rotor position Θ based on measured phase currents and a machine model known from prior art.

[0194] The control diagram illustrates key elements only for a field- oriented control of the motor 94. A field-oriented controller (FOC) may be implemented for example as a software program contained in a memory of the control circuit 950, as a FPGA, or with other means.

[0195] In case of software implementation, a software program is executed by a processor as part of the control circuit 950 during the motor H55-18-PCT1controllers’ 93 operations and configured to control the electrical energy provided to the motor 94 in a control loop.

[0196] The FOC generates a three-phase voltage as a vector vS to control the three-phase stator current of the motor 94. The stator currents comprise two orthogonal components that can be represented with a vector. One component defines the magnetic flux of the motor 94, the other component of the vector the torque. By transforming the AC currents into rotational vectors using transforms, the flux and the torque components become time-invariant and thus allow the control with conventional techniques such as PI controllers, as with a standard DC motor. The term “controller”, such as flux controller, relates in the following to a software or hardware module, dependent on how the FOC is implemented.

[0197] In the following the term “value” represents electrical or physical quantities determined by measurement or set by the motor controller 93, whereas the term “variable” is a result of a calculation or transformation of a value represented in the processor.

[0198] All three-phase stator currents of the motor 94 may be determined by the current sensors 961-963 connected to the control circuit 950. These measurements provide values iU, iV and iW. In the prior art, it is often found that only two phase currents are measured and a third phase current is calculated based to the relationship iU+iV+iW = 0. This may only work in the absence of a failure in at least one of the field coils. If one of the phase lines is interrupted by one of the switches 961-963 it is only possible to measure the current in each of the remaining phase lines.

[0199] Alternatively, the phase currents of the motor may be determined using sensorless algorithms known in the prior art.

[0200] The three-phase stator currents of the motor 94 are converted to a two-axis coordinate system using the Clarke transformation 940. This H55-18-PCT1conversion provides the variables iα and iβ from the measured AC stator currents iU - iW. The variables iα and iβ are time-varying quadrature current values as viewed from the perspective of the stator. The rotor position Θ is directly measured by the encoder 960 or derived by integrating the speed determined by the encoder 960. Alternatively, the rotor position Θ is calculated and / or estimated by the Position and Speed estimator 943. In the latter case, an observer can be used by the Position and Speed estimator 943.

[0201] The two-axis coordinate system is rotated to align with the rotor flux using a transformation angle calculated at an initial or previous iteration of the control loop. The Park transformation 941, using the rotor position Θ, provides the id and iq variables derived from the variables iα and iβ. The id and iq variables are the quadrature currents transformed into the rotating coordinate system. For steady state conditions, id and iq are constant.

[0202] A speed setpoint ωRef corresponding to the rotor speed is set and an error signal is formed using the speed setpoint ωRef and the determined the rotor speed ωM*. The actual rotor speed ωM can alternatively be provided by a Position and Speed estimator 943, which uses a machine model for estimating the actual rotor speed. The velocity controller 948 is provided as a PI-controller and regulates its output, being the Δiq variable according to the error signal.

[0203] The speed of the rotor cannot be increased above the rated speed of the motor 94 without losing torque due to saturation of the ferromagnetic part of the motor with the magnetic flux generated by the rotation of the rotor. In case of coil failure, when a higher rotation speed is needed in order to compensate at least in part for the loss of torque due to the defective coil, the speed of the rotor can be further increased using a specific field-weakening control. The idea of a Field Weakening controller is to lower the resulting d-flux component (variable id, rotor magnetizing flux) by reducing the effect of the flux of the rotor. H55-18-PCT1

[0204] Therefore, a Field Weakening controller 949 controls the id variable. By setting the speed setpoint ωRef above the rated speed of the motor 94, the velocity controller 948 may decrease the q-component (variable iq, torque output) of the flux or may keep the q-component constant, whereas the Field Weakening controller reduces the d-flux component at the same time. Reducing means, that a negative d-flux component is applied to weaken the rotor magnetic field.

[0205] Fig.25 illustrates a graph showing quantitative values of the torque M available at the motor shaft, the rotor flux Φr, and the stator voltage VS, whereas the motor is controlled by using the FOC algorithm with the field-weakening control as disclosed before in the present example implementation.

[0206] When the FOC controls the rotor speed ω beyond the rated speed ωr of the motor, the torque M of the motor starts to decrease with the increasing rotor speed ω. The Field Weakening controller decreases the rotor flux Φr, by applying a negative d-flux component when entering the speed region above the rated speed ωr of the motor. The stator voltage VS is kept constant at its maximum in the overspeed region.

[0207] In the circuit of Fig.23, further error signals Δid, Δiq are formed using the variables id, iq and corresponding setpoints id*, iq*. The setpoint id*controls the rotor magnetizing flux and the setpoint iq*controls the torque output of the motor 94. The error signals Δid, Δiq are inputted to the Flux Controller 947 and the torque controller 946, wherein each is configured as PI controller. Other controller types, such as a bang-bang controller might be used instead.

[0208] The output of the Flux Controller 947 and the Torque Controller 946 provide the variables vd and vq, representing a voltage vector with two voltage vector components that will be set to the motor 94. The two voltage component vectors may be represented in the rotating d-q axis. H55-18-PCT1

[0209] A new transformation angle is calculated in a subsequent iteration of the control loop, where the variables vα, vβ, iα and iβ are considered as inputs. The new transformation angle guides the FOC as to where to place the next stator voltage vector vS.

[0210] The variables vd and vq provided by the Flux- and Torque Controller 947, 946 are rotated back to the stationary reference frame using the new transformation angle. The inverse Park transformation 942 provides the subsequent quadrature voltage values vα and vβ under consideration of the current rotor position Θ.

[0211] The subsequent quadrature voltage variables vα and vβ are transformed back to three-phase voltage values using an inverse Clarke transformation in the Pulse-Width Modulation (PWM) Modulator 944. New PWM duty cycle values vUC-vWC are calculated in the PWM modulator based on the transformed three-phase voltage values for signalling the inverter circuit 945.

[0212] The inverter circuit 945 sets its pulse pattern according to the provided duty cycle values vUC-vWC.

[0213] This process and the corresponding control loop are executed periodically, as long as the motor controller 93 provides the motor 94 with electrical energy.

[0214] In case of a failure in one of the field coils, the FOC control, as disclosed hereinbefore, is not suitable to control the motor operation, as the Clarke- and Park transformation requires the presence of a symmetrical three-phase system.

[0215] A failure in one of the field coils is detected by the control circuit 950, for example by determining the current in the phase lines during the operation of the motor 94. The control circuit 950 can detect an imbalance H55-18-PCT1in the three-phase system caused by the faulty field coil. Alternatively, or in addition, a fault in at least one of the field coils is detected with other means, such as a temperature or voltage sensor, as disclosed in one of the examples hereinbefore.

[0216] In this case scenario, the control circuit 950 controls the switch 931-933 in the corresponding phase line from a conductive state into a non- conductive state, isolating the phase line which supplies the defective field coil.

[0217] The control circuit 950 controls the motor 94 using the FOC algorithm as disclosed before with slight modifications. As the Clark- and Park transformations (including the inverse transformation) are not applicable in case of a loss of a phase line, the corresponding blocks 940, 941, 942, 944 in Fig.23 will be replaced or complemented by blocks, including new rotational transformations.

[0218] Examples of such rotational transformations are discussed and disclosed in the publication: “Performance Evaluation of the Field-Oriented Control of Star-Connected 3-Phase Induction Motor Drives under Stator Winding Open-Circuit Faults” Mohammad Jannati, et al., Journal of Power Electronics, Vol.16, No.3, pp.982-993, May 2016, http: / / dx.doi.org / 10.6113 / JPE.2016.16.3.982. Even though the publication discusses the implementation for an induction motor drive, the rotational transformations disclosed is suitable with the required adaptation for other 3-phase machines, such as the used permanent-magnet synchronous machine.

[0219] At the same time, when the control circuit 950 controls the motor 94 with the modified FOC algorithm, the speed setpoint ωRef, which corresponds in steady state operation to the rotor speed, is increased by the control circuit 950, such that the rotor speed exceeds the rated speed of the motor 94. Due to the higher speed that the rotor can reach, the loss of H55-18-PCT1torque caused by one failed coil is at least partially compensated, so that the aircraft will maintain its velocity, or reduces its loss of speed.

[0220] In conclusion, at least two FOC algorithms are implemented in the control circuit 950. A first algorithm is executed under normal operation, and a second algorithm is executed after detection of a failure of at least one field coil, where one phase line may be isolated.

[0221] Fig.24 illustrates a torque diagram of a motor controlled by the motor controller 93 using the first and second FOC algorithm comprised in the control circuit 950 as disclosed in the example before.

[0222] Fig.24 illustrates the torque M of the motor over the rotor speed ω. Electrical quantities, such as the absolute value of stator voltage vector vS, the power P and the absolute value of the stator current vector IS of the motor are also illustrated. The electrical quantities are controlled linearly (increased or decreased) over a speed range from zero speed to the rated speed ωr of the motor and having their maximum at the point of the rated speed ωr. The torque M of the motor is controlled to be constant in the speed range up to the rated speed ωr of the motor.

[0223] The grey highlighted area represents the speed range, where the motor is operated in field-weakening mode subsequently to a failure in one of the field coils. This speed range starts at the rated speed ωr of the motor and ends at the maximum allowable speed ωMax of the motor. The electrical quantities are controlled to be constant in this speed range, whereas the torque M of the motor decreases proportionally to the inverse value of the rotor speed ω. The absolute value of the stator voltage vector vS is proportional to the DC voltage vdc at the input side of the motor controller.

[0224] The power P of the motor can be controlled to be constant in the field-weakening mode while the rotor speed ω is increased. Therefore the aircraft keeps its velocity while the motor is operated in field-weakening H55-18-PCT1mode. In situations where the increase of the rotor speed ω is not sufficient to maintain the velocity of the aircraft, the blade pitch of the propeller 970 can be adjusted to move more air per revolution of the rotor to compensate for the loss of velocity. The adjustment of the blade pitch can be initiated by a higher level control means in response to the detection of a loss of velocity of the aircraft or in response to a failure in one of the field coils. However, it is apparent that the compensation with the adjustment of the blade pitch has its limit since the motor torque is reduced when operating in field-weakening (independently from the failure in one of the field coils).

[0225] The motor is controlled in the field-weakening mode using the FOC control algorithm as explained before, independently from a failure in one of the field coils of the motor.

[0226] Increasing the rotor speed ω beyond the maximum allowable speed ωMax of the motor would result in a drop of the power P and torqe M of the motor, which is not desirable as the speed of the aircraft would decrease. On the other hand, the motor might be damaged, as harmonics and alternating torques are produced.and the related method for improving the motor controlthe rotorand the rotor flux error

[0227] A step back needs to be taken to better illustrate the underlaying methodology of the vector control using the control diagram of Fig.23. The control diagram of Fig.23 illustrates the typical application of a closed- loop control, in which the setpoint ^^∗for the torque output of the motor is provided by the velocity controller 948, and the setpoint of field- weakening current^^∗is provided by the Field Weakening controller 949, in dependency of the speed setpoint ωRef. However, for the desired application of the control diagram of Fig.23 in an electric or hybrid aircraft a feed-forward control for the torque output and the rotor flux can be H55-18-PCT1more desirable. In particular the said setpoints ^^∗,^^∗might be directly set or commanded without a dedicated feedback loop comprised in the control circuit. For instance, the required torque for accelerating the aircraft might be directly set in dependency on a command torque request by the pilot. Therefore, there may not be a need for a closed-loop control. The setpoint ^^∗for the torque output can be referred in the following to as the q-axis current component setpoint ^^∗, and the setpoint for controlling the rotor magnetizing flux can be referred in the following to as the d-axis current component setpoint ^^∗.

[0228] Fig.26 illustrates a rotating dq-reference frame, atop a stationary αβ reference frame, which is usually aligned with the stator. The stator current space vector ^^is aligned with the stator magnetic field and rotates with the frequency of the said magnetic field. The d- and q-axis current component vectors ^^, ^^of the stator current space vector ^^are situated in the dq-reference frame and were obtained by measuring the stator phase currents and applying the Clark and Park transformation to the measurements. The dq-reference frame is rotating in the stationary αβ reference frame system with a speed that corresponds to the rotor speed ωM, under steady-state conditions. Thus, the magnetic stator field and the rotor rotate at the same speed in steady state. In addition, the rotating dq- reference frame is chosen such that the d-axis is aligned with the rotor flux vector Φ^.

[0229] In the control circuit, as of Fig.23, the d- and q-axis current component vectors ^^, ^^of the stator current space vector IS are controlled by their amplitudes (absolute values) and thereby affecting the torque output or rotor flux of the motor. However, for the control of the d- and q- axis current component vectors ^^, ^^the knowledge of the effective rotor position θ^^^is a prerequisite, since any deviation therefrom results in a deterioration of the motor control quality.

[0230] The rotor position θ^^^conventionally is determined with a rotary encoder mechanically connected to the rotor or is estimated using a H55-18-PCT1machine model and stator voltage and current measurements. Please note, that the d-axis current component vector ^^of the stator current space vector ^^has been illustrated exemplarily. For controlling the motor with maximum output torque, the d-axis current component vector ^^is maintained at zero. Especially in the case as illustrated, the d-axis current component vector ^^points in the opposite direction to the rotor flux vector Φ^, which can result in a weakening of the rotors' magnetic field.

[0231] Fig.27 illustrates a variant of the control diagram of Fig.23, in particular a simplified control diagram for the motor controller 93 connected to the motor 94. The motor controller 93 and the motor 94 can be configurated as disclosed in one of the previous examples, and the control circuit 950 is configured as explained in the example for Fig.23, comprising the modifications as explained in the following.

[0232] In the present example, each phase line also includes a current sensor 961-963 arranged between the inverter circuit 945 and the motor 94, for measuring an electrical current flowing in each of the phase lines during operation of the motor 94. The currents can refer to the stator currents.

[0233] Each current sensor 961-963 is connected to the control circuit 950. The current sensors 961-963 may be Hall effect current transducers. Other types of current sensors might be used instead, such as current transformers, Rogowski coils or shunt resistors. The voltage in each phase line is optionally measured using voltage sensors 964-966. The voltage sensors 964-966 can be configured as insulated voltage sensors to avoid a potential carryover or for protection purposes and they may be used for the sensor less control only. When a sensor-based vector control is used, the voltage sensor can be omitted.

[0234] The motor 94 includes a rotary encoder 960 connected to the shaft of the motor for determining the rotor speed ωM and the rotor position θ^^^. The rotary encoder 960 is connected to the control circuit 950. H55-18-PCT1The rotor speed ωM and the rotor position θ^^^is determined and outputted by the Position and Speed estimator 943 based on rotary encoder signals. Alternatively, the rotor speed ωM and the rotor position θ^^^can be calculated and / or estimated by the Position and Speed estimator 943 based on measured stator currents and voltages and a machine model, a rotor position estimator, or a rotor position observer known from the prior art.

[0235] In the present example, the control diagram 950 illustrates key elements only for a field-oriented control of the motor 94. The field- oriented controller may be implemented for example as a software program contained in a memory of the control circuit 950, such as a microcontroller, FPGA, or other means.

[0236] In the case of software implementation, a software program is executed by a processor as part of the control circuit 950 during the motor controllers’ 93 operations and configured to control the electrical energy provided to the motor 94 in a control loop. The term “controller”, such as Flux Controller 947, also relates in the following to a software or hardware module, dependent on how the vector control is implemented.

[0237] Compared to the control diagram illustrated in Fig.23, the present control diagram lacks the velocity controller. A by the pilot demanded torque is received by the motor controller 93, e.g. from a higher level control means, and translated into the q-axis current component setpoint ^^∗. The Field Weakening controller 949 also lacks the input for the speed setpoint ωRef. Therefore the torque and the flux are controlled in an open-loop manner, assuming a certain torque constant. The rotor speed ωM is calculated in the position and speed estimator 943 and considered by the Field Weakening controller 949. In the following steady state of the motor 94 is assumed, and the rotor speed ωM is synchronous to the electrical frequency of the rotating stator field, which is typical for synchronous motors in a steady state. The absence of field-weakening is furthermore assumed, which means that the rotor speed ωM assumed is situated well below the critical speed for the field weakening. H55-18-PCT1

[0238] The control circuit 950 has been complemented with a feed- forward decoupling structure configured to decouple the Flux Controller 947 and the Torque Controller 946 from each other, in particular their crossed-coupled voltage terms. This structure is used, amongst other things, to compensate for unmodeled dynamics in the motor 94. The feed-forward decoupling structure, in principle, models the differential equations defining the synchronous motor operation (i.e. governing the current change in the dq-reference frame). The said equations can be presented as:where: ^^denotes the stator resistance, ^^, ^^denote the d- or q-axis component of the stator inductance in the rotating dq reference frame, where ^^is the inductance of a winding in alignment with the rotor flux and ^^corresponds to the self-inductance of the winding in-quadrature with the rotor flux; ^^, ^^denote the d- or q-axis component of the stator voltage, ^^, ^^denote the d- or q-axis component of the stator current, ω^denotes the rotor speed or electrical frequency of the rotating stator field, Φ^denotes the rotor flux linkage.

[0239] One can notice that d-axis and q-axis current components are not independent of each other. From equation (1) one can notice that the d- axis component ^^of the stator voltages is not the only voltage term competing for control of the d-axis current component. There is also a speed-dependent term that contains the q-axis component ^^of the stator currents in it.

[0240] From equation (2), the q-axis component ^^of the stator voltage is also competing with a voltage term containing the d-axis component ^^H55-18-PCT1of the stator currents. For both the Flux Controller 947 and the Torque Controller 946, this cross-coupling effect manifests itself as an unwanted disturbance most prominent during transient conditions at high rotor speeds.

[0241] To correct this situation, the feed-forward decoupling structure, as illustrated in Fig.27, is applied which exactly cancels these cross-coupled voltage terms. The feed-forward decoupling structure in principle implements the equations (1) and (2). The product of the rotor speed ω^and the parametrized rotor flux linkage Φ^,^added at the output of the Torque Controller 946 is strictly speaking not part of the feed-forward decoupling structure since it does not contribute to the cross-coupling effect. In the course of the present disclosure, however, the product may be considered to be part of the feed-forward decoupling structure. Since the outputs of the Flux and the Torque Controller 947, 946 should drive the currents to their steady-state values, and by assuming that the terms for the stator resistor can be neglected, as the voltage drop is typically very small, equations (1) and (2) can be further simplified to:where: ^ ^^ ^ , ^ ^^ ^ denote the d- or q-axis component of the feed-forward stator voltage reference, apparent at the summing point at the outputs of the Flux Controller 947 and the Torque Controller 946, respectively; ^∗ ^, ^^∗denote the d- or q-axis component of the stator voltage reference, apparent after the summing points; Δ^^, Δ^^denote d- or q-axis component of the stator voltage reference at the outputs of the Flux and the Torque Controller 947, 946 respectively.

[0242] In the control circuit 950, the rotor flux linkage Φ^refers to the parametrized rotor flux linkage Φ^,^. The parametrized rotor flux linkage H55-18-PCT1Φ^,^can also be expressed as the back-EMF constant over the number of pole pairs of the rotor. The motor parameters, in particular, the d- or q-axis component of stator inductance ^^, ^^, and the parametrized rotor flux linkage Φ^,^are available in the control circuit 950 in the form of variables, whereby the underlying physical motor parameters were obtained by measurement of the motor 94, such as in an end-of-line test after final assembly.

[0243] The equations (1) - (4) are applicable when the rotor position θ^^^outputted by the Position and Speed estimator 943 optimally aligns with the effective rotor position of the motor. However, in case there is a deviation between the outputted (given) rotor position θ^^^and the effective rotor position θ^(as illustrated in Fig.28), the equations as presented might not be valid anymore, and the outputs of the Flux and the Torque Controller 947, 946 might not be able to compensate for this deviation. As a result, the motor cannot be controlled for its maximum torque output since the d-axis of the dq-reference frame can’t be precisely aligned with the rotor flux anymore. The deviation between the outputted rotor position θ^^^and the effective rotor position might be caused, for instance, by a permanent misalignment of the rotary encoder and the shaft of the motor. In this case, the deviation might be static and persists while the motor is operated.

[0244] Fig.28 illustrates this circumstance in more detail, in particular by two rotating dq-reference frames rotating in the stationary αβ reference frame, as also explained for Fig.26. The dq-reference frame illustrated in solid lines represents the reference frame effectively aligned with the rotor, and the corresponding rotor flux vector Φ^of the rotor. The equations (1) and (2) apply correspondingly for the dq-reference frame aligned with the effective rotor position. The dq-reference frame illustrated in dashed lines represents the reference frame as used for controlling the d- and q-axis components of the stator current, controlled by the control circuit of Fig. 27. Obviously, the effective rotor position θ^deviates from the rotor position θ^^^outputted by the Position and Speed estimator 943 by some H55-18-PCT1degree, in particular by the rotor position deviation ∆θ^^^. For this case, the equations (1) and (2) can be re-written and applied to the control circuit as: ωM^^d− ^q^ ωM^^d− ^q^ (5) ^d= ^dsin 2 2(Δθdet)+ ^qcos 2(2Δθdet)^M^^d− ^q^ ^M^^d− ^q^ (6) ^q= ^qsin 2Δθ + ^ cos 2Δθ 2(det)d2(det)^^d+ ^ +q2 + +where: ∆θ^^^= θ^^^− θ^.

[0245] It can be noticed that the differential equations (1) and (2) defining the AC motor operation include various parasitic terms that arise from the deviation between the effective rotor position θ^and the rotor position θ^^^outputted by the Position and Speed estimator. As long as the deviation is static or has slow-changing dynamics, the Flux and the Torque Controller might be able to ensure that the setpoint for the d- and q-axis current components are tracked. Assuming that the rotor position θ^^^outputted by the Position and Speed estimator typically is almost accurate, yet there might exist a static deviation of a few degrees due to different effects. If the deviation ∆θ^^^is considerably small, e.g. not higher than ±0.06 rad (±3.4°), the cosine of double the angle can be approximated as one (cos(2Δθdet)≈ 1), while the sine can be approximated by the angle itself H55-18-PCT1(sin(2Δθdet)≈ 2Δθdet). This assumption can be used to better approximate the d- and q-axis components of the stator voltages by:

[0246] Even though an assumption about the small angular error is made, the expressions for the voltages in the dq-reference frame are still relatively complex. Some useful conclusions can be made by observing how equations (5) and (6) behave in the case when the d- and q-axis components of the stator current are set to zero and the steady-state condition is reached. In this case equation (5) simplifies to: ^d= ωMΦM^^^(Δθdet)(9)

[0247] The rotor speed ωMand the parametrized rotor flux linkage Φ^,^are conventionally available in the control circuit with high accuracy, wherein the tolerance of the parametrized rotor flux linkage Φ^,^can be assumed to be less than ±10%. Reflecting on the control circuit of Fig.27, the d-axis component of the stator voltage ^dis available at the output of the Flux Controller, presuming the setpoints of the d- and q-axis component of the stator currents were set to zero, and the outputs of the Flux and the Torque controller reached a steady state. In this case, and under consideration of the control circuit, equation (9) corresponds to:H55-18-PCT1The last expression shows that the d-axis voltage under the given circumstances contains information regarding the rotor flux ΦMand the sine of the angular error Δθdet.

[0248] The disclosure makes use of a further possibility to take benefit of the feed-forward decoupling structure and the controllers’ outputs, since with the use of the said structure it is possible to extract an essential motor parameter under operation of the motor 94, while controlling the motor 94 with the use of the control circuit 950 as illustrated in Fig.27. While setting the d- and q-axis component of the stator currents to zero, equation (6) is also simplified, and can be written as:The last expression shows that the q-axis voltage reference under the given circumstances contains the same information as the previous one, with the only difference that it is proportional to the cosine of the angular error.

[0249] Consequently, the deviation of the rotor position can be determined by:

[0250] Equation (12) indicates that the deviation of the rotor position can also be determined for errors greater than 0.3rad. In conclusion, the error between the effective rotor position and the rotor position outputted by the Position and Speed estimator can be determined by the control circuit of Fig.27, especially by setting the setpoint of the d- and q-axis component of the stator current to zero and waiting until the outputs of the Flux and Torque controller are settled. The settlement might H55-18-PCT1also be detected in the control circuit internally. The deviation can then be determined (for instance) at the outputs of the Flux and Torque controllers and can further be used to compensate for the static deviation, e.g. as correcting term in the Position and Speed estimator. The control structure enables all this.

[0251] In addition, combining the equations (10) and (11), one can determine the actual rotor flux linkage ΦMin the following manner:Determination of both the rotor flux linkage ΦMand the position error Δθdetcan be completed in a single measurement time window, before the flight, or during the flight.

[0252] In cases when the angular position error is low, the expression (13) can be simplified by neglecting the influence of the d-axis current controller contribution Δ^d. This can typically be done when^Δ^d / Δ^q^≤ 0.1. In such a case, the rotor flux linkage ΦMis determined with an accuracy greater than ±0.5%, using the following expression:

[0253] While the detection method of the rotor position error requires that both the d and q-axis current references are set to zero for a short amount of time (roughly 10-100 current control time constants), calculation of the rotor flux linkage ΦMand comparing it with the parametrized rotor flux linkage Φ^,^apparent in the control circuit, the parametrized rotor flux linkage Φ^,^can be corrected to better model the effective corresponding electrical parameter of the motor. H55-18-PCT1

[0254] In summary, the outputs of the Flux and the Torque Controller (which may or may not include the feed-forward decoupling structure) can be used to correct a deviation between the effective rotor position and the representation of the rotor position used in the control circuit of Fig.27. In addition; it can be used to correct a deviation between the effective rotor flux linkage and the representation of rotor flux linkage used in the said control circuit. By correcting the rotor position deviation and the parametrized rotor flux, the torque output of the motor can be optimized and leading to an improved accuracy of the motor torque (by having a better knowledge of the rotor flux linkage and position information). In addition, the optimal transition into the field-weakening mode of operation can be achieved, and the higher safety can be achieved (by avoiding motor saturation or magnets demagnetization for example).

[0255] Fig.29 illustrates a flowchart of a process for determining the static rotor position deviation ∆θ^^^and an effective rotor flux linkage, which can be used to correct or replace the parameterized rotor flux linkage ΦM,Papparent in the control circuit of Fig.27. The process starts while the aircraft is grounded and the rotor (including the propeller mechanically attached thereto) is at standstill.

[0256] In step i) the rotor is turned to a first rotational speed, which is smaller than the maximal speed of the motor but much greater than zero speed. The said speed also needs to be lower than the speed, in which field- weakening is activated. Consecutively, the setpoints of the d- and q-axis components of the stator current are set to zero, and it is waited for the settlement of the Flux and Torque controller outputs. Alternatively, the steady state values of the controller outputs might be detected in the control circuit internally by a corresponding means. Subsequently, the static rotor position deviation ∆θ^^^is determined or calculated using the value outputted at Flux Controller and Torque Controller outputs, under consideration of equation (12). The rotor position deviation ∆θ^^^consecutively is taken to correct the rotor position θ^^^outputted by the Position and Speed estimator of Fig.27. This can be achieved by adding a correction term to the Position and Speed estimator to add the determined H55-18-PCT1static rotor position deviation ∆θ^^^. The deviation might be stored and permanently applied for further controlling the motor using the control circuit of Fig.27 and under consideration of the setpoints of the d- and q- axis components of the stator currents.

[0257] Step i) can be repeated every time the aircraft is prepared for flight, while taxing or while flying as indicated by the dashed line, provided that the motor is not operating in field weakening. Consecutively the regular motor operation is started under the control of the control circuit using the corrected rotor position θ^^^, such that the aircraft can take off.

[0258] Upon execution of step i), step ii) is initiated, all while maintaining the same conditions as in step i). The effective rotor flux linkage ΦMis calculated using the value outputted at Torque Controller output under consideration of equation (14) or the values outputted by the Torque and Flux Controllers using equation (13). Consecutively, the parametrized rotor flux linkage ΦM,Pof the control circuit is corrected by the determined effective rotor flux linkage ΦM. The control of the motor is continued by conventionally controlling the motor with the control circuit of Fig.27, once the deviation has been corrected, under consideration of the setpoints of the d- and q-axis component of the stator current, in particular, the setpoint for the torque.

[0259] Step ii) can be repeated during flight, e.g. every 10 minutes or even more frequently, until the aircraft is grounded. The airstream passing by the propeller supports in maintaining the rotor speed. The process of step i) and ii) can be processed by the control circuit in less than 10ms, thereby not influencing the flight quality of the aircraft due to the inertia of the mechanical system. The process as explained, might be initiated by the control circuit itself, or by a higher-level control instance.

[0260] Even though the steps i) and ii) are illustrated as being performed in separate steps, they can be processed one after the other in one single process step. They can also be repeated every time the aircraft is H55-18-PCT1prepared for flight, while taxing or flying in the absence of field- weakening operation. In addition, or alternatively, the order of steps i) and ii) might be changed or selected individually when technically required. Example implementation of a propulsion system and a related method for improving the motor control by controlling the motor in the field- weakening region under consideration of the actual DC link voltage and the feed-forwardstructure

[0261] The field-weakening operation of the motor has been discussed previously in light of a failure in one (or more) field coil(s) of the motor. However, the feed-forward decoupling structure can also contribute to an improved performance of the motor in field-weakening operation in the absence of a failure in one (or more) of field coil(s) of the motor. The feed- forward decoupling structure, however, is not a mandatory part of the control structure as such and may be omitted.

[0262] The control circuit of Fig.27 shall be considered as the basis for the following explanation, in particular with respect to the related equations (3) and (4). One should recognize that the field-weakening operation of the motor follows the dynamics of the mechanical system, i.e. the speed of the motor shaft. Therefore, the time constants of the mechanical system are 2-3 orders of magnitude lower than the time constants of the current control loop. Consequently, all the transients in the electrical system can be neglected, and only the steady-state values may be taken into consideration.

[0263] Due to the motor parameters' or position deviation that is not compensated, the stator voltage equations in the dq-reference frame take different form than suggested in equations (1)-(2). Accounting for a position information mismatch of less than 3 electrical degrees, and expressing the stator voltages as in (3) and (4), the following expressions are yielded. This may cause a deviation, even when the static deviation has been corrected using the process as discussed in Fig.29. H55-18-PCT1where: ^^^^^denotes the actual stator resistance, ^^^^ ^ , ^^^^^denote the actual d- or q-axis component of the stator inductance, where ^^^^ ^ is the inductance of a winding actually in alignment with the rotor flux and ^^^^^corresponds to the self-inductance of the winding where the rotor flux is effectively aligned with the in-quadrature winding, Δ^^, Δ^^denote the error of the d- or q-axis component of stator inductance, Δ^^, Δ^^denote the actual deviation of the d- or q-axis component of the stator voltage, Δ^^, Δ^^denote error terms of the d- or q-axis component of the stator voltage, in particular an induced back-EMF due to the position deviation, Φ^^^ M denotes the actual rotor flux linkage, ΔΦMdenotes the rotor flux linkage error.

[0264] In equations (15) and (16), parameters with the superscript act are the actual values of the parameters, which are not necessarily equal to the values used for the feed-forward compensation as of the related equations (3) and (4). Variables Δ^^, Δ^^are the errors in the back-EMF introduced by the deviation in the rotor position.

[0265] The equation and the underlaying physics are better visualized in Fig.30. Again, two rotating dq-reference frames are illustrated that rotate in the stationary αβ reference frame. The dq-reference frame shown in solid lines represents the reference frame effectively aligned with the rotor. The dq-reference frame illustrated in dashed lines represents the reference frame as used for controlling the d- and q-axis components of the stator current, controlled by the control circuit of Fig.27. The two reference frames are displaced by the rotor position deviation ∆θ^^^. It is visible that an induced back EMF component Δ^^only exists in the actual q-axis (^^^^), H55-18-PCT1advancing the rotor position for 90 degrees. However, due to the deviation in the rotor position, the control circuit will be assuming the same amount of the induced back EMF, yet in a displaced axis ^^^^. In fact, this creates a small error in the q-axis back EMF, labeled as Δ^^. Even greater deviation exists in the back EMF induced in the d-axis ^^^^, proportional to the sine of the angular rotor position deviation. These influences are compensated by the Flux and Torque Controllers, in particular as defined in equations (15) and (16).

[0266] Finally, as only the steady-state values of the Flux and Torque controller are of interest, one can obtain the expressions for the said controller outputs in d- and q-axis under steady-state conditions. Assuming that the said controllers ensure ideal reference tracking (^^= ^^∗; ^^= ^^∗) the controller outputs are equal to:where: ^∗ ^ , ^^∗denote the setpoint of the d- or q-axis component of the stator current.

[0267] Equations (17) and (18) illustrate that the Flux- and Torque Controller in the d- and q-axis will ensure that all the deviations in the parameters will be compensated in terms of their influence on the current control. This fact can be used to enhance the state-of-the-art open-loop field-weakening control strategy, such that it becomes insensitive to the parameters deviations. Therefore, instead of setting the setpoints for the d- axis and q-axis current component, using ideal parameters assumption, e.g. as in the Field Weakening block of Fig.27, one can benefit from the output values of the Flux and Torque controller to mitigate the dependence of the cross-coupled voltage terms on various non-idealities. H55-18-PCT1

[0268] Fig.31 represents a diagram of a field-weakening block for adapting the control parameters (variables) when a field-weakening operation of the motor is required, e.g. when the motor is controlled in the constant power region, using the control circuit as illustrated in Fig.27. Also in this example, the motor is a synchronous machine, preferably as a permanent magnet synchronous machine.

[0269] Preferably, it is assumed that the rotor position θ^^^outputted by the Position and Speed estimator and the parametrized rotor flux linkage Φ^,^are optimally set, e.g. by the process as outlined in Fig.29. However, this is not a prerequisite, as the enhanced open-loop algorithm, as being presented in the following paragraphs, is insensitive to the deviations, especially in the rotor flux linkage.

[0270] The process 951 as illustrated in Fig.31 is preferably executed on the same control device as the control circuit of Figs.23 or 27. For executing the process 951, multiple input parameters, setpoints, and measurement signals are required. The block (S) inputs the q-axis current component setpoint ^^∗in dependency of a torque requested by the pilot. The block (P) inputs multiple parameters, such as the parametrized rotor flux linkage Φ^,^, the absolute value of the stator current limit ^^,^^^, further parametrized motor parameters, state-variable limits, etc.. The block (M) inputs measurement signals, such as the actual DC link voltage vDC, or the actual speed of the rotor ω^outputted by the Position and Speed estimator, and further values determinable from the control circuit of Fig.27. The output block (O) outputs the optimal (or maximal) setpoint for the d-axis component ^^∗of the stator current and the limit for the q-axis component ^^,^^^of the stator current in field-weakening control of the motor. The outputs of the output block (O) can be used in the control circuit of Fig.27, in particular for the Field Weakening block, as it sets the setpoint for the d-axis component ^^∗of the stator currents, as well as a limit to the q-axis current component setpoint ^^∗. H55-18-PCT1

[0271] In the initialization step (I), a first value is calculated, which is further used in the course of the process 951. The value den accounts for:where: ^^denotes the parametrized q-axis value of the stator inductance, ^^^^, ^^denote machine constants, in particular maximal expected torque of the machine, and the machine torque constant, respectively, Φ^,^denotes the parametrized rotor flux linkage. The ultimate q-axis current limit ^^,^^^also is calculated in the initialization step (I), by:

[0272] In the first step i) of the process 951, the maximum allowable phase voltage ^^^,^^^is calculated in dependency on the actual DC link voltage ^^^as measured by:where: ^^^denotes the actual DC link voltage, Δ^^^^^denotes a predetermined voltage control margin, e.g.40V.

[0273] In the second step ii) of the process 951, the critical field- weakening speed ^^,^^is calculated by: H55-18-PCT1ii) Wherein the constant 0.85 represents the safety margin and can be varied or set differently depending on the configuration of the motor and / or motor controller. The critical field-weakening speed ω^,^^is the minimal motor speed at which the motor controller and motor are likely to enter the field-weakening operation (when exceeding the critical field- weakening speed).

[0274] In the third step iii) the actual speed ω^of the rotor is compared with the critical field-weakening speed ω^,^^, where field-weakening control might become required. If the actual rotor speed ^^is smaller than the critical field-weakening speed ω^,^^, then no field-weakening operation is required, and the setpoint for the d-axis component ^^∗of the stator current is set to zero, and the limit for the q-axis component ^^,^^^of the stator current is set to the predetermined maximum allowable stator current is,max(which depends on the motor characteristics at the highest torque and may depend on the maximum current suppliable by the motor controller to the motor). This is done in step iv).

[0275] The q-axis current component setpoint ^^∗in dependency of the torque requested by the pilot, might be overwritten in the control circuit of Fig.27 by the limit for the q-axis component ^^,^^^of the stator current, in case its absolute value is higher than the calculated limit. The process 951 might continue with step i) or iii). However, if the actual rotor speed ω^is greater than the critical FW speed ω^,^^, the process 951 continues with step v), in which the setpoint for the d-axis component ^^∗of the stator current is calculated by:H55-18-PCT1where: ^^denotes the parametrized d-axis inductance, Δ^^, Δ^^denote filtered d- and q-axis components of the stator voltage outputted by the Torque and Flux Controllers of Fig.27, ^^,^^denotes a filtered d-axis feed-forward component of stator voltage as outputted at the summing point of the Torque Controller of Fig.27. Both voltage components are filtered using low pass filters in order to account only for their steady-state values.

[0276] In step vi) the d-axis component ^^∗of the stator current is further determined. If the d-axis current component ^^∗is greater than zero, no field-weakening operation is required and the process 951 continues with step iv). However, if the d-axis current component ^^∗, as calculated in step v) is smaller than zero, the field-weakening operation is initiated.

[0277] In a first step of the field-weakening operation, step vii), it is determined if the calculated d-axis current component ^^∗is smaller than a predetermined limit ^^,^^^. If the calculated d-axis current component ^^∗is equal or greater than the said predetermined limit ^^,^^^, the process 951 continues with step viii) in which the q-axis component limit ^^,^^^of the stator current in field-weakening operation is calculated by:

[0278] The control of the motor is continued in field-weakening operation, with the d-axis component ^^∗as determined in step v) and with the limit of the q-axis component ^^,^^^as determined in step viii). The process 951 can consecutively continue in step i) or iii). Preferably the process 951 always restarts from i), in order to update the maximal phase voltage value. If the calculated d-axis current component ^^∗is equal or greater than the predetermined limit ^^,^^^, the process 951 continues with step ix), in which the q-axis current component ^^,^^^of the stator current in field-weakening control of the motor is determined by: H55-18-PCT1where: ^^denotes the parametrized q-axis inductance, Δ^^, Δ^^denote d- or q-axis component of the stator voltage terms reference at the outputs of the Flux and the Torque Controller of Fig.27, ^^,^^denotes the filtered d-axis feed-forward component of stator voltage as outputted at the summing point of the Torque Controller of Fig.27, (whereby both voltage components are filtered using low pass filters in order to account only for their steady-state values) and the d-axis component ^^∗as determined in step v) is limited to a predetermined d-axis component current limit ^^,^^^.

[0279] In step x) the actual stator current is determined and compared to the predetermined maximum allowable stator current ^^,^^^by:

[0280] If the actual stator current is smaller than the maximum allowable stator current ^^,^^^, step xi) is followed, where the existing references and determined limits of step ix) are used. However, if the actual stator current ^^is greater than the maximum allowable stator current ^^,^^^, xii) is followed, in which the q-axis component ^^∗of the stator current is further limited by: ^^,^^^xii)

[0281] The term limiting can refer to overwriting or manipulating a setpoint or value, such as overwriting the q-axis current component setpoint ^^∗as set in dependency of the commanded torque request, based on the q-axis current component limit ^^,^^^. H55-18-PCT1

[0282] In summary, the following steps are performed by the process 951 as outlined in Fig.29 under consideration of the outputs by the Flux- and Torque Controller and the feed-forward decoupling structure: i) Calculate the maximally available stator voltage in dependency of the actual DC link voltage with a voltage reserve, e.g.40V. ii) Calculate the critical speed where the field-weakening operation could start. iii) Check for the actual rotor speed- if not attained, field-weakening operation is not initiated; continue with step iv); but if the actual rotor speed exceeds the critical FW speed, calculate d-axis currentin step v). iv) set the d-axis current setpoint to zero, and the q-axis current limit the the maximal stator current. vi) If the calculated d-axis current component ^∗ ^ exceeds zero, no field- weakening operation is required; continue with step iv), but if the calculated d-axis current componentis lower than zero, transition into field-weakening operation, by evaluating the calculated d-axis current component ^∗ ^ in step vii). vii) If the calculated d-axis current component ^∗ ^ is not lower than its lowest limit, use the calculated d-axis current component ^∗ ^ to calculate the q-axis current component limit ^^,^^^in step viii), but if the calculated d-axis current componentis lower than its lowest limit, apply the limit to the d- axis current componentand calculate the q-axis current component limit ^^,^^^in step ix). x) Calculate the absolute value of the actual stator current and determine if the absolute value exceeds the maximum value. If the absolute value of the actual stator current does not exceed the maximum value, use the calculated q-axis component current limit ^^,^^^and the corresponding limited d-axis component ^∗ ^ in step xi), but if it does exceed the maximum value, recalculate the q-axis current component limit ^^,^^^in step xii).

[0283] Fig.32 illustrates a comparison between a conventional feed- forward-based method for controlling a motor in field-weakening operation and the method of Fig.31 for controlling a motor in field- weakening operation in presence of motor parameters deviation. The conventional feed-forward based method does not take into account any H55-18-PCT1variabilities of the motor or the control circuit. The setpoints for the d-axis component ^^∗of the stator current in the conventional control are directly set by a Field Weakening block, whereas the q-axis component ^^∗of the stator current might be directly commanded, e.g. independency of a torque request set by a pilot. For the conventional feed-forward based methods, however, no parameter deviations, such as the deviation in the rotor flux linkage or stator inductances, are not considered for the control.

[0284] On the axis of the abscissas of all graphs (a) - (c), the rotor speed of the motor is presented in a range from standstill to 7000 rpm. The range between 6000 rpm and 7000 rpm may denote the maximum rotor speed achievable with the specific motor, whereas the field-weakening operation of the motor starts at approximately 4500 rpm.

[0285] Graph (a) illustrates in dotted lines the maximal attainable stator phase voltage with a stable given DC link voltage. The axis of ordinates denotes the voltage. In addition, the steady-state stator phase voltage is illustrated in dashed lines, controlled by a conventional feed-forward control circuit under consideration of the parameter deviation. Finally, the steady-state stator phase voltage (solid line) is illustrated, controlled by the control circuit as of Fig.27, under consideration of the process as illustrated in Fig.31 and related to the same parameter deviation. One can notice that in the speed region below 4500 rpm the steady-state stator phase voltage using both control approaches is perfectly aligned. However, in the speed region between 4500 rpm and approximately 6800 rpm there is a deviation between the steady-state stator phase voltage. In particular, the steady- state stator phase voltage of the motor controlled by the control circuit as of Fig.27 aligns with the maximal attainable stator phase voltage. The other steady-state stator phase voltage is situated below. In summary, the capabilities of the motor are better utilized by using the process as illustrated in Fig.31, especially in case of parameter deviation of the motor.

[0286] The axis of ordinates in graph (b) on the left denotes the d-axis current component and the axis of ordinates on the right denotes the q- H55-18-PCT1axis current component, in the unit ampere. The dotted line at the bottom of the graph denotes the limit for the d-axis current component. References are being made in the following to the speed region above 4500 rpm.

[0287] The dotted line on the left represents the d-axis current component setpoint under consideration of the parameter deviation using the conventional feed-forward control. The nearby dashed line represents the d-axis current component setpoint using the process as illustrated in Fig.31, also under consideration of the parameter deviation. Furthermore, the solid line on the right relates to the q-axis current component setpoint under consideration of the parameter deviation using the conventional feed-forward control, whereas the nearby long dashed line illustrates the q-axis current component setpoint using the process as illustrated in Fig.31. It can be noticed, that the setpoints of the d-axis and q-axis components of the stator currents using the process as illustrated in Fig.31 start to degrade at a higher motor speed and thereby better utilize the motor capabilities.

[0288] The axis of ordinates in graph (c) on the left denotes the maximum of the stator current in ampere, whereas the axis of ordinates on the right denotes the torque in Newtonmeter. The dotted line at 280 amperes denotes the stator current limit. References are being made in the following to the speed region above 4500 rpm.

[0289] The dotted line on the left represents the stator current reference under consideration of the parameter deviation using the conventional feed-forward control. The nearby dashed line represents the stator current reference using the process as illustrated in Fig.31, also under consideration of the parameter deviation. Furthermore, the solid line on the right relates to the maximal attainable torque under consideration of the parameter deviation using the conventional feed-forward control, whereas the nearby long dashed line illustrates the maximal attainable torque using the proposed as illustrated in Fig.31. It can be noticed, that H55-18-PCT1the maximal achievable torque, relating to the torque output of the motor, can be longer maintained constant using the process as illustrated in Fig.31 before the maximal attainable torque starts to degrade at a higher motor speed and thereby better utilizing the motor capabilities.

[0290] In summary, it can be observed that the process, as illustrated in Fig.31, using the control circuit as illustrated in Fig.27 embodied with the feed-forward compensation structure, improves the utilization of the available voltage and current by using the values from the Flux- and Torque Controller. Namely, it fully uses the available DC link voltage, resulting in an extended region of operation without the field weakening, i.e. lower losses and higher torque capabilities. The benefits of using this process can be even more pronounced for larger parameter deviations.

[0291] Finally, the presented graphs were created using simulation models. The results are illustrated in Fig.32 over the whole speed range of interest. Initially, the commanded torque was set to the nominal torque of 143 Nm, which resulted in the setpoint for the q-axis current component of 258 A (graph (c)). Until the speed of around 4500 rpm, the setpoint for the d-axis current component was set to zero, and the rotor was accelerating with the maximal torque (graph (b)). At that speed, the stator voltage reached the defined limit, and the negative d-axis current component was injected (graph (b)). Due to the further increase of the motor speed, the d- axis current component setpoint further increased, resulting in the increased stator current. At the speed of roughly 5600 rpm, the stator current limit was reached, resulting in the reduction of the q-axis current limit. With the further increase of speed, the d-axis current was further increased in absolute terms, reaching its maximal value around the speed of 6300 rpm, where it was limited by the process of Fig.31. H55-18-PCT1and related method for

[0292] Fig.33 illustrates a propulsion system of an aircraft, comprising two battery packs 91, 92 connected in series and functioning as an energy source for the motor controller 93 connected to the battery packs 91, 92 at an input end. The motor 94 is connected to an output end of the motor controller 93 and features a propeller 970 mechanically connected to the motor shaft of the motor 94. The motor 94 is configured as a three-phase permanent magnet synchronous motor. The motor controller 93 features an inverter circuit with a plurality of power semiconductors and a DC link, which includes the DC link capacitor CDC for buffering electrical energy. The motor controller 93 also comprises a braking chopper circuit, which includes a power semiconductor Sbr in series with the braking resistor Rbr, connected in parallel to the DC link capacitor CDC. When the power semiconductor Sbr is activated (state conducting) a current can flow through the braking resistor Rbr and electrical energy of the resistor is converted into heat. The motor controller 93 also comprises one diode D1, which is situated between the battery packs 91, 92 and the braking chopper circuit. The diode D1 prevents a flow of electrical energy in the reverse direction, i.e. from the DC link into the batteries 91, 92. The motor controller 93 comprises a non- illustrated control device, which implements the control and power circuitry of Fig.23 or 27, respectively.

[0293] On an aircraft, the direction of rotation of the propeller 970 is indicated from the pilot's point of view. The most commonly used direction of rotation is clockwise, i.e. as seen by the pilot from the cockpit. It might happen that the propeller 970 rotates in a counter-clockwise direction, e.g. when driving the aircraft backward to leave the parking position, or the propeller is accelerated by other air movements, such as backwind on the runway. However, before the propeller is forced to rotate in a clockwise direction, e.g. for generating thrust for propelling the aircraft, the propeller needs to be decelerated (braking). H55-18-PCT1

[0294] For braking the propeller 970, it is necessary that the motor 94 applies a certain torque to counteract the movement of the propeller 970 and the rotating rotor, respectively. Conventionally the motor 94 is controlled by the motor controller 93 in generator mode, to generate a toque for decelerating the propeller 970. While the motor 94 is operated in generator mode, the motor 94 supplies electrical energy into the direct current link, such that the DC link voltage vDC increases. To prevent that the DC link voltage exceeds its predetermined limit, the braking chopper circuit is activated, such that the excess energy is wasted in form of heat in the braking resistor Rbr, and thereby lowering the DC link voltage vDC. One design criterion for the choice of braking resistor is to select the maximum power that can be generated by the motor 94 under consideration while the motor is turning counter-clockwise at a certain speed.

[0295] For example, if the mechanical power provided by the rotation of the propeller accounts for Pmech=2 kW and the motor losses (in generator mode) account for Pmotor=500 W, the braking resistor Rbr is selected to dissipate Pdisi=1500W. However, selecting a braking resistor Rbr that is capable of dissipating that amount of electrical energy might be bulky and adds additional weight to the aircraft, which is not desirable at all. Furthermore, the motor controller 93 must be designed to evacuate the heat generated by the braking resistor Rbr. Forced cooling might be required, which increases the complexity of the construction of the motor controller 93 and / or the aircraft.

[0296] Alternatively, the propeller can be decelerated to standstill, by short circuiting the terminals of the motor 94. The terminals are short- circuited by activating all semiconductors of the inverter circuit connected to either positive or negative dc pole at the same time. In this operation, a high amount of torque is generated, without that the DC link voltage vDC is increased. However, this might not be desirable as it can result in undesirable pendulum torques in the propulsion system. More preferred is to control the motor 94, as illustrated in Fig.34. H55-18-PCT1

[0297] Fig.34 illustrates a process for decelerating the propeller with the use of the propulsion system in the configuration shown in Fig.33 and the control circuit in Fig.23 or 27, respectively. When the propeller needs to be decelerated (suppose it is turning in the counter-clockwise direction), the motor is controlled in a first step i) by the motor controller in generator mode and thereby allowing a reverse flow of energy, such that energy generated by the motor is supplied into the direct current link. Step i) in principle, is the step where the propeller braking is initiated, and the positive torque command is given while the speed is negative (=counter- clockwise). At the same time, or slightly delayed step ii) is initiated. However, step i) and ii) might be combined in one single step.

[0298] In the second step ii), the d-axis current component setpoint (please refer to control circuit of Fig.23 or 27) is set to a negative value and the q-axis current component setpoint is set to a value sufficient to generate an adequate torque for braking the propeller by applying a torque for counteracting the movement of the propeller and the rotating rotor. The required q-axis component of the stator current can be calculated by

[0299] where: ^^^^denotes the required q-axis component of the stator current; ^^^denotes the required braking power; ^^denotes the motor torque constant (ratio between the torque and the q- axis current), ω^denotes the actual rotor speed, and ^^^^^denotes absolute value of the maximal stator currents.

[0300] While the propeller rotates counter-clockwise, the setpoint for the q-axis current component ^^∗needs to be limited, as indicated in equation (19), so as not to produce power higher than the required braking power or generate the stator current in excess of the maximal stator H55-18-PCT1current. The setpoint for the d-axis current component ^∗ ^ of the stator current, on the other hand, can be calculated by:where: 1.1 denotes a margin of 10% in order to compensate for tolerances in the machine; ^^^^^denotes the actual q-axis component of the stator current; ^^^^ ^^denotes the actual braking power; and ^^denotes the stator resistance.

[0301] By setting the d-axis component of the stator current to a negative value, the rotor magnetic field influence on the stator is weakened, and the motor is operated in an operation mode, where additional losses are generated in the stator windings. As a result, the energy supplied by the propeller is no longer available as an excess voltage in the DC link but is dissipated in the motor itself. The setpoint for the d- axis component of the stator current is maintained to be negative as long as the propeller is turning in the counter-clockwise direction. As long as the rotor and propeller are turning in the counter-clockwise direction, the motor is controlled as a generator, and the rotor magnetic field is weakened.

[0302] In step i) and ii) the braking chopper circuit remains deactivated as long as the DC link voltage persists below a predetermined threshold, but can be activated if the said threshold is exceeded. In an optimal control point, the braking chopper circuit remains deactivated because the excess energy is entirely dissipated through motor losses. The process is terminated when the propeller is at standstill, or when the propeller turns in the clockwise direction. The process can also be applied when the propeller needs to be decelerated from turning in the clockwise direction to standstill The only difference would be that the applied torque by the H55-18-PCT1motor would be negative, hence, the q-axis current component setpoint ^^∗ is calculated the same way as in equation (19), yet multiplied by -1 at the end. On the other hand, the d-axis current component setpoint ^∗ ^ is calculated in the same manner as in equation (20), and it has demagnetization properties.

[0303] In the present example, when the process illustrated in Fig.34 is used, the braking resistor is selected for a lower power, e.g. for dissipating Pdisi=300W of excess power; without that the DC link voltage vDC exceeds its limits. In summary, the braking chopper circuit is configured to dissipate less than ¼ of the power supplied by the motor when the propeller is decelerated. The final objective, however, might be to omit the braking chopper circuit entirely and only dissipate the excess power in the motor in the form of intentionally generated losses. The circumstance that the motor controller is designed to allow a unidirectional flow of energy may be a hypothetical case scenario. The process as illustrated can also be useful to prevent an overcharging of the battery (when used as a DC energy source), as a flow of energy into the DC link and finally into the battery can be prevented.

[0304] Fig.35 illustrates the influence on a DC link voltage while braking the rotor of the motor with the use of the presented braking method.

[0305] The figure shows the results of a simulation, whereby the simulation was performed assuming a propeller rotating in counter- clockwise direction (=negative direction), resulting in a motor speed of -750 rpm. The q-axis current limit ^^^^^is calculated to produce a torque that would result in the braking power not being higher than the propeller power at the maximal negative speed (for example -1000 rpm).

[0306] At the time instant t=0.01s, the pilot applies a step torque reference of 30 Nm, which is passed with a certain slew rate to the control circuit of Fig.23 or 27, respectively. The torque reference is translated into H55-18-PCT1the setpoint for the q-axis current component ^^∗, which is followed precisely owing to the Torque Controller. On the other hand, the d-axis current referenceis calculated using the process as indicated in Fig.34, and soon reaches its maximal value, where it is maintained. The motor decelerates.

[0307] At the time instant t=0.1s, with the motor speed of around - 600 rpm, the pilot sets the torque reference at 80 Nm, which is translated into the corresponding setpoint for the q-axis current component ^^∗. However, the said setpoint is limited by the maximal braking power limit (corresponding to ^^^^^).

[0308] The d-axis current necessary for the braking is still maintained at its maximum. At the time instant t=0.2s, the q-axis current component limit ^^^^^is high above the commanded setpoint for the q-axis current component ^^∗. This current reference results in a torque equal to 80 Nm, while the power taken from the shaft decreases with the rotor speed. This results in a decrease of the setpoint for the d-axis current component ^∗ ^ to zero. The motor further continues to accelerate to positive speeds in clockwise direction. As it can be noticed at the bottom of the figure, the DC link voltage vDC remains nearly unchanged, and the main objective is fulfilled, namely to avoid an increase in the DC voltage.and a the motor control in case of a loss of the speed orerroneoussensor signal error

[0309] Figs.20 and 21 illustrate a motor 94 connected to the motor controller 93 in different arrangements.

[0310] Fig.20 illustrates a first battery pack 91 and a second battery pack 92 connected in series for providing a DC voltage Vdc at the input side of the motor controller 93. The motor controller 93 supplies the motor 94 on the output side with electrical energy. H55-18-PCT1

[0311] The motor 94 in the following examples is configured as a three- phase permanent-magnet synchronous motor. Other motor types, such as induction machines, might be used instead in all examples. The motor 94 can propel the aircraft with the use of the propeller.

[0312] The motor controller 93 provides the motor 94 with three-phase alternating currents and voltages with varying amplitudes and varying frequencies. The motor controller 93 is also configured to vary the fundamental frequency of the alternating currents and voltages provided to the motor 94. A propeller is attached to the shaft of the motor 94 and an air stream surrounds the propeller. The situation appears during the flight of the airplane.

[0313] Fig.21 illustrates a motor controller 93 with a DC voltage Vdc at the input side and a motor 94 connected at the output side. The motor controller 93 comprises multiple switches 931-933 configured to interrupt or arrange the connection between electrical circuits (not shown) comprised in the motor controller 93. The electrical circuits can comprise power semiconductors for converting the DC voltage VDC at the input side into alternating quantities, such as alternating voltages and / or currents.

[0314] The switches 931-933 may be high power contactors. Other forms of switches, such as semiconductor switches, might be used instead.

[0315] A control circuit (not shown) is connected to the switches 931-933 and configured to control the said switches from a non-conductive state into a conductive state and vice versa. The control circuit may include digital components, including for example a processor, a FPGA circuit, and / or any combination of digital and / or analogue components for controlling the switches. Also, in this example, passes the airstream the propeller.

[0316] A speed sensor in the form of a rotary encoder is mechanically connected to the motor shaft, as in the examples illustrated in Figs.20 and H55-18-PCT121. The angular position of the rotor of the motor 94 can be determined with the use of the said rotary encoder. In addition, can the speed of the rotor (angular frequency) be determined. The rotary encoder is electrically connected to the control circuit of Fig.21.

[0317] Fig.36 illustrates a simplified control diagram for a motor controller 93 connected to the motor 94 of Figs.20 and 21.

[0318] The motor controller includes an inverter circuit 945 configured to convert the DC voltage Vdc at the input side into three-phase alternating currents and voltages at the output side. Three phase lines interconnect the inverter circuit 945 of the motor controller 93 and the motor 94. Each of the three phase lines includes a switch 931-933 connected in series with each phase line. Each switch 931-933 is connected to the control circuit 950 and can be controlled by the control circuit 950 between a conductive and non-conductive state and vice versa.

[0319] In this example, each phase line further includes a plurality of current sensors 961-963 between the inverter circuit 945 and the switches 931-933, for measuring an electrical current flowing in each of the phase lines during the operation of motor 94. Each current sensor 961-963 is connected to the control circuit 950. The current sensors 961-963 may be hall effect current transducers. Other types of current sensors, such as current transformers, Rogowski coils, or even shunt resistors, might be used instead. The voltage of each phase line is also measured with the use of a plurality of voltage sensors 964-966, such as voltage transducers.

[0320] The motor 94 includes the rotary encoder 960 mechanically connected to the shaft of the motor for determining the rotor speed ωM and the rotor position Θ. The rotary encoder 960 is electrically connected to the control circuit 950. The rotor speed ωM and the rotor position Θ can be alternatively calculated and / or estimated by the control circuit 950 using a position and speed estimator 943. H55-18-PCT1

[0321] The position and speed estimator 943 calculates and / or estimates the rotor speed ωM*and the rotor position Θ*based on measured phase currents or phase voltages and a machine model known from the prior art. The position and speed estimator 943 is essential for the operation of the motor controller 93 and motor 94 in case of a failure of the rotary encoder 960 and will be explained in greater detail in the course of this example.

[0322] Three control modes are considered. In the sensor-based control mode the motor controller 93 controls the motor 94 with the use of a sensor-based vector control method utilizing the sensor signals provided by the rotary encoder 960. In case of a loss of the sensor signal or a defect of the rotary encoder 960 a sensorless control mode needs to be considered, wherein the position of the rotor needs to be estimated. The same control mode might be used during the start-up of the motor controller 93, in particular when the aircraft is grounded, and the exact rotor position is not known. The initialization control mode is considered when the sensor signal of the rotary encoder 960 is permanently lost or disregarded for other purposes, such as a defect of the rotary encoder 960. In this mode, the motor controller 93 controls the motor 94 permanently with the sensorless vector control method.

[0323] The control diagram illustrates key elements for a field-oriented control of the motor 94 only. A field-oriented controller (FOC) may be implemented, for example, as a software program contained in a memory of the control circuit 950, as a FPGA, or with other means. As said, the FOC can be used for all control modes in different variations.

[0324] In case of software implementation, a software program is executed by a processor as part of the control circuit 950 during the motor controllers’ 93 operations and configured to control the electrical energy provided to the motor 94 in a control loop.

[0325] The FOC generates a three-phase voltage as a vector vS to control the three-phase stator current of the motor 94. The stator currents H55-18-PCT1comprise two orthogonal components that can be represented with a vector. One component defines the magnetic flux of the motor 94, the other component of the vector the torque. By transforming the AC currents into rotational vectors using transforms, the flux and the torque components become time-invariant and thus allow the control with conventional techniques such as PI controllers, as with a standard DC motor. The term “controller”, such as flux controller, relates in the following to a software or hardware module, dependent on how the FOC is implemented.

[0326] In the following the term “value” represents electrical or physical quantities determined by measurement or set by the motor controller 93, whereas the term “variable” is a result of a calculation or transformation of a value represented in the processor.

[0327] During the sensor-based control mode, all three-phase stator currents of the motor 94 are determined by the current sensors 961-963 connected to the control circuit 950. These measurements provide values iU, iV and iW. In the prior art, it is often found that only two phase currents are measured and a third phase current is calculated based to the relationship iU+iV+iW = 0.

[0328] The three-phase stator currents of the motor 94 are converted to a two-axis coordinate system using the Clarke transformation 940. This conversion provides the variables iα and iβ from the measured AC stator currents iU - iW. The variables iα and iβ are time-varying quadrature current values as viewed from the perspective of the stator. The rotor position Θ is directly measured by the encoder 960 or derived by integrating the speed determined by the encoder 960.

[0329] The two-axis coordinate system is rotated to align with the rotor flux using a transformation angle calculated at an initial or previous iteration of the control loop. The Park transformation 941, using the rotor position Θ, provides the id and iq variables derived from the variables iα and H55-18-PCT1iβ. The id and iq variables are the quadrature currents transformed to the rotating coordinate system. For steady-state conditions, id and iq are constant.

[0330] A speed set point ωRef corresponding to the rotor speed is set and an error signal is formed using the speed set point ωRef and the determined the rotor speed ωM*. The velocity controller 948 is provided as a PI- controller and regulates its output, being the iq variable according to the error signal.

[0331] The speed of the rotor cannot be increased above the rated speed of the motor 94, due to saturation of the ferromagnetic part of the motor with the magnetic flux generated by the rotation of the rotor. However, with the use of the field weakening controller 949 the torque of the motor 94 can be exhaustively utilized in all operational ranges. The idea of a field weakening controller 949 is to lower the resulting d-flux component (variable id, rotor magnetizing flux) by reducing the effect of the flux of the rotor.

[0332] Therefore, the field weakening controller 949 controls the id variable. By setting the speed set point ωRef above the rated speed of the motor 94, the velocity controller 948 increases the q-component (variable iq, torque output) of the flux, whereas the field weakening controller reduces the d-flux component at the same time.

[0333] Further error signals are formed using the variables id, iq and corresponding set points id*, iq*. The set point id*controls the rotor magnetizing flux and the set point iq*controls the torque output of the motor 94. The error signals are inputted to flux controller 947 and into the torque controller 946, wherein each is configured as PI controller. Other controller types, such as a bang-bang controller might be used instead.

[0334] The output of the flux controller 947 and the torque controller 946 provide the variables vd and vq, representing a voltage vector with two H55-18-PCT1voltage vector components that will be set to the motor 94. The two voltage component vectors may be represented in the rotating d-q axis.

[0335] A new transformation angle is calculated in a subsequent iteration of the control loop, where the variables vα, vβ, iα and iβ are considered as inputs. The new transformation angle guides the FOC as to where to place the next voltage vector vS.

[0336] The variables vd and vq provided by the flux- and torque controller 947, 946 are rotated back to the stationary reference frame using the new transformation angle. The inverse Park transformation 942 provides the subsequent quadrature voltage values vα and vβ under consideration of the current rotor position Θ.

[0337] The subsequent quadrature voltage variables vα and vβ are transformed back to three-phase voltage values using an inverse Clarke transformation 942 in the Pulse-Width Modulation (PWM) Modulator 944. New PWM duty cycle values vUC-vWC are calculated in the PWM modulator based on the transformed three-phase voltage values for signalling the inverter circuit 945.

[0338] The inverter circuit 945 sets its pulse pattern according to the provided duty cycle values vUC-vWC.

[0339] This process and the corresponding control loop are executed periodically, as long as the motor controller 93 provides the motor 94 with electrical energy.

[0340] In the sensorless control mode, the speed signal provided by the rotary encoder 960 is not available and / or is disregarded for other purposes. The loss of the sensor signal can occur during the sensor-based control mode, in particular when the motor controller 93 controls the motor 94 with the use of the sensor-based vector control method. H55-18-PCT1

[0341] In this case, the control circuit 950 needs to transition into the sensorless control mode. In this mode the motor controller 93 can control the motor 94 using a V / f or I / f open loop control. This can be achieved by using the control diagram as illustrated but disabling the feedback loop (the lower part in the diagram, including the Clarke transformation 940, the Park transformation 941, and the position and speed estimator 943) and thereby directly influencing the variables vd and vq (V / f control) or id and iq (I / f control) and finally the voltage vector vS.

[0342] Controlling the motor with the use of the method is necessary to turn the rotor, for instance, during start (from a standstill to a first rotational speed, sufficient to measure the back-EMF generated by the motor). However, and in case the airplane is cruising, the application of this control method is not required, as the rotor remains turning, due to the airstream passing the propeller. If the V / f or I / f open loop control is used to control the motor 94, then the control is stopped. The switch (as illustrated in Fig 21) is opened and the motor 94 generates a voltage, that can be determined with the voltage sensors 964-966, as the rotor of the motor 94 keeps turning.

[0343] This also applies to the situation in which the propeller turns the rotor during flight. The step of interrupting the phase lines can be necessary to ensure that the motor 94 is still turning, and the faulting signal is not caused by the defect of the motor 94. Furthermore, the voltages measured vu-vw can be used to estimate the rotor position Θ* and the rotor speed ωM*by the position and speed estimator 943 with the use of the measured voltages.

[0344] The position and speed estimator 943 is functioning in this case as a simple back-emf observer, only relying on the voltages that are induced by the rotating rotor into the stator coil of the motor 94. For this purpose the measured voltages vu-vw are transformed with the use of the Clarke transformation 940 into variables vα*, vβ*in the αβ coordinate system and finally into variables vd*, vq*which represents quadrature voltages H55-18-PCT1transformed to the rotating coordinate system (d-q reference frame). The position and speed estimator 943 can then estimate using the variables vd*, vq*the rotor position Θ* and the rotor speed ωM*with methods known from the prior art.

[0345] However, interrupting the phase lines is not the only possibility to determine the rotor position in situations of the loss of the signal of the rotary encoder 960. There are further solutions known from the prior art for which it is not necessary to interrupt the phase lines.

[0346] For instance, the motor controller 93 can comprise a resistor that can be switchable activated (also called braking resistor). When the motor 94 supplies electrical energy, the said energy can be wasted in the said resistor, for the sake of causing determinable currents in the phase lines.

[0347] By determining the currents and voltages of the phase lines, the rotor position Θ* and the rotor speed ωM*can be estimated with the use of the method, such as disclosed in the publication "Zhiqian Chen, M. Tomita, S. Doki, et al. “An extended electromotive force model for sensorless control of interior permanent-magnet synchronous motors”. In: IEEE Transactions on Industrial Electronics 50.2 (Apr.2003), pp.288–295. https: / / doi.org / 10.1109 / TIE.2003.809391 /

[0348] Alternatively, the motor controller 93 can feed the energy that is supplied by the motor 94 into the battery and the related currents and voltages measured in the phase lines can also be used to estimate the rotor position Θ* and the rotor speed ωM*, also with the use of the extended back-EMF observer as disclosed in the said document.

[0349] Once the rotor position Θ* and the rotor speed ωM*is estimated using one of the before outlined possibilities, the initialization control mode is entered and / or used. The motor controller 93 controls the motor 94 with the use of the sensorless bases vector control method, taking benefit H55-18-PCT1of the estimated rotor position Θ* and the rotor speed ωM*as starting point for the conventional sensorless based motor control.

[0350] The example outlined herein before takes benefit of the behavior of the permanent-magnet synchronous motor. However, the control structure is also suitable for other types of electrical machines, such as induction machines or DC machines, but the control structure would require a slight adaptation of the control in the sensorless control mode. Additional Features and Terminology

[0351] Although examples provided herein may be described in the context of an aircraft, such as an electric or hybrid aircraft, one or more features may further apply to other types of vehicles usable to transport passengers or goods. For example, the one or more futures can be used to enhance construction or operation of automobiles, trucks, boats, submarines, spacecraft, hovercrafts, or the like.

[0352] Many other variations than those described herein will be apparent from this disclosure. For example, depending on the embodiment, certain acts, events, or functions of any of the algorithms described herein can be performed in a different sequence, can be added, merged, or left out altogether (for example, not all described acts or events are necessary for the practice of the algorithms).

[0353] The various illustrative logical blocks, modules, and algorithm steps described herein can be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. The described functionality can be implemented in varying ways for each H55-18-PCT1particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the disclosure.

[0354] Conditional language used herein, such as, among others, "can," "might," "may," "e.g.," and the like, unless specifically stated otherwise, or otherwise understood within the context as used, is generally intended to convey that certain embodiments include, while other embodiments do not include, certain features, elements or states. Thus, such conditional language is not generally intended to imply that features, elements or states are in any way required for one or more embodiments or that one or more embodiments necessarily include logic for deciding, with or without author input or prompting, whether these features, elements or states are included or are to be performed in any particular embodiment.

[0355] The terms "comprising," "including," "having," and the like are synonymous and are used inclusively, in an open-ended fashion, and do not exclude additional elements, features, acts, operations, and so forth. Also, the term "or" is used in its inclusive sense (and not in its exclusive sense) so that when used, for example, to connect a list of elements, the term "or" means one, some, or all of the elements in the list. Further, the term "each," as used herein, in addition to having its ordinary meaning, can mean any subset of a set of elements to which the term "each" is applied. H55-18-PCT1Reference symbols in the figures 91 First battery pack 92 Second battery pack 93 Motor controller 94 Motor 931, 932, 933 Switch 940 Clarke transformation 941 Park transformation 942 Inverse Park transformation 943 Rotor Position and Speed estimator 944 PWM modulator 945 Inverter circuit 946 q-axis current controller, Torque Controller 947 d-axis current controller, Flux Controller 948 Velocity controller 949 Field Weakening controller 950 Control circuit, control structure 960 Rotor position sensor, resolver 970 Propeller id d-axis component of the stator current (dq-reference frame, variable or value) iq q-axis component of the stator current (dq-reference frame, variable or value) iq*, id*Setpoint for the d-axis or q-axis component of the stator current (dq-reference frame, variable) iU, iV, iW Current phase U, V, W iα, iβ Time-varying quadrature current (variable or value) in the αβ-coordinate system iS Stator current vector (absolute value) M Torque, Motor torque, Rotor torque P, p Power vd, vq Components of voltage vector vS in dq-reference frame (variable or value) vα, vβ Components of the voltage vector (variable or value) in H55-18-PCT1the stationary αβ-coordinate system vdc, Vdc DC voltage, DC link voltage vS Stator voltage, or stator voltage vector (absolute value) Θ, θ, θ^^^Rotor position ∆^^^^Deviation between rotor position outputted by the Position and Speed estimator and the effective rotor position Φ^rotor flux vector, rotor flux linkage, effective rotor flux linkage Φ^,^parametrized rotor flux linkage ΔΦMrotor flux linkage error, rotor flux linkage deviation ω, ωM, ωM*Rotor speed, actual rotor speed, stator field electrical frequency ωMax Maximum rotor speed ωRef, ωr Speed setpoint, rated speed Further references are specifically defined in the corresponding equations. H55-18-PCT1

Claims

Claims 1. A propulsion system for an electric or hybrid aircraft (100), comprising: - a motor (94) comprising a rotor and a stator; - a motor controller (93) connected to the motor (94), wherein the motor controller (93) is arranged for supplying driving signals (is, vs) to the motor (94), wherein the motor controller (93) comprises a control structure (950) configured to implement a field-oriented control for controlling the driving signals (is, vs) in dependency of a given rotor position (Θdet), wherein the control structure (950) is adapted to determine a rotor position static deviation (ΔΘdet) between the given rotor position (Θdet) and an effective rotor position (Θact) of the rotor, wherein the control structure (950) is further arranged to correct the given rotor position (Θdet) based on the determined rotor position static deviation (ΔΘdet).

2. The propulsion system of claim 1, comprising a rotor position sensor (960) configured to output rotor position sensor signals for indicating a rotor position, wherein the motor controller (93) is arranged to receive the rotor position sensor signals and the control structure (950) is adapted to determine the given rotor position (Θdet) and / or the rotor angular speed, based on the rotor position sensor signals.

3. The propulsion system of claim 1, the motor controller (93) comprising a measurement circuit for measuring the driving signals (is, vs), wherein the control structure (950) comprises a rotor position estimator (943) configured to determine the given rotor position (Θdet) based on measured driving signals.

4. The propulsion system of any one of the claims 1 to 3, wherein the driving signals (is, vs) are supplied by the motor controller (93) to the motor (94) in form of stator voltages (vs) and stator currents (is), wherein the control structure (950) comprises a d-axis current controller (947) configured to control a d-axis voltage component (vd) of the stator voltages H55-18-PCT1(vs) in dependency of a d-axis current component setpoint (id*) of the stator currents (is) and a q-axis current controller (946) configured to control a q- axis voltage component (vq) of the stator voltages (vs) in dependency of a q- axis current component setpoint (iq*) of the stator currents (is), the control structure (950) being optionally further arranged with a feed-forward decoupling structure configured to decouple the d-axis current controller (947) and the q-axis current controller (946) from each other, in particular cross-coupled voltage terms thereof.

5. The propulsion system of claim 4, wherein the control structure (950) is further adapted to set said current component setpoints (id*, iq*) to zero and being configured to detect a steady state at outputs of the d-axis current controller (947) and / or of the q-axis current controller (946), wherein the control structure (950) being provided to determine the rotor position static deviation (ΔΘdet), when the steady state of the outputs of the d-axis current controller (947) and q-axis current controller (946) is detected.

6. The propulsion system of claim 5, wherein the control structure (950) is adapted to determine the rotor position static deviation (ΔΘdet) based on a ratio of the output value of the d-axis current controller (947) and the output value of the q-axis current controller (946).

7. The propulsion system of claim 5 or 6, wherein the control structure (950) is adapted to determine an effective rotor flux linkage (Φ^), the control structure (950) being further adapted to correct an parametrized rotor flux linkage (Φ^,^) with the use of the determined effective rotor flux linkage (Φ^) for controlling the driving signals (is, vs) based on the corrected given rotor position (Θdet) and the corrected parametrized rotor flux linkage8. The propulsion system of claim 7, wherein the control structure (950) is provided to determine the effective rotor flux linkage (Φ^) based on the H55-18-PCT1output value provided by the d and / or q-axis current controller (946) and an actual rotor speed (ωM), when the steady state is detected.

9. The propulsion system of any one of the claims 1 to 8, wherein the motor (94) is a synchronous motor, preferably a three-phase permanent magnet synchronous motor.

10. The propulsion system of claim 1 comprising a variable-pitch propeller (970) and control means, wherein the variable-pitch propeller (970) is mechanically connected to an output shaft of the motor (94), wherein the control means is configurated to adjust a blade pitch of the variable-pitch propeller (970).

11. A method for controlling a propulsion system according to any one of the claims 1 to 10 in an electric or hybrid aircraft (100), , the method comprising the steps of: - supplying the motor (94) with driving signals (is, vs); - rotating the rotor of the motor (94), preferably at a constant speed; - setting the current component setpoints (id*, iq*) of the d-axis current controller and the q-axis current controller (946) to zero; - detecting a steady state at the outputs of the d-axis current controller (947) and the q-axis current controller (946); - determining the rotor position static deviation (ΔΘdet) between the given rotor position (Θdet) and the effective rotor position (Θact) of the rotor when the steady state is detected; - correcting the given rotor position (Θdet) using the determined rotor position static deviation (ΔΘdet); - controlling the driving signals (is, vs) using the corrected given rotor position (Θdet).

12. The method of claim 10, further comprising the step of: - calculating a ratio of the output value of the d-axis current controller (947) and of the output value of the q-axis current controller (946), the H55-18-PCT1rotor position static deviation (ΔΘdet) being determined using said calculated ratio.

13. The method of claim 11 or 12, further comprising the steps of: - determining an effective rotor flux linkage (Φ^); - correcting a parametrized rotor flux linkage (Φ^,^) using the determined effective rotor flux linkage; - controlling the driving signals (is, vs) using the corrected given rotor position (Θdet) and the corrected parametrized rotor flux linkage (Φ^,^).

14. The method of claim 13, wherein the effective rotor flux linkage (Φ^) is determined using the output value provided by the d and / or q-axis current controller (946) and the actual rotor speed (ωM).

15. The method of any one of the claims 11 to 14, wherein the steps of determining the rotor position static deviation (ΔΘdet) between the given rotor position (Θdet) and the effective rotor position (Θact) are performed while the electric or hybrid aircraft (100) is grounded or flying, and / or the step of determining the effective rotor flux linkage (Φ^) is performed while the electric or hybrid aircraft (100) is grounded or flying at a constant speed.

16. The method of claim 15, wherein the effective rotor flux linkage (Φ^) is determined during flight of the electric or hybrid aircraft (100), when the rotor is rotated at constant speed by a propeller exerting a couple on the rotor by an airflow passing the propeller during flight. H55-18-PCT1

Images