ELECTRIC SUPERCHARGER

Technical Field

The present invention relates to electric supercharg in particular electric superchargers including switched reluctance motors. Background of the Invention

Electric superchargers are becoming increasingly

attractive for use with internal combustion (IC) engines in the automotive industry. Firstly, they enable lower fuel consumption in the IC compared to conventional (direct engine- driven) superchargers and thus reduce carbon dioxide

emissions. They also tend to be able to be more responsive and may be able to attain higher speeds than direct engine- driven superchargers .

Using a switched-reluctance motor in an electric

supercharger (to drive the compressor element) has been found to be particularly beneficial. A previously-suggested control system for controlling the speed of a switched-reluctance motor in an electric supercharger is illustrated in Figure 1.

Referring to Figure 1, the control system 101 receives an input, setting the target speed 102 of the motor in the supercharger. This target speed 102 is determined by a control system for the engine (for example in response to the throttle position) . The control system for the engine is not shown in Figure 1.

The target speed 102 is compared to the actual speed 104 of the motor, to produce a speed error. The speed error 104 is received by a proportional integral (PI) controller 103. PI controller 103 then determines the appropriate torque for changing the speed of the motor from its actual speed 104 towards the target speed 102. The torque determined by the PI controller is referred to as the "torque demand" 106.

In some circumstances it is desirable to limit the maximum value of the torque applied by the motor. For example it might be desirable to reduce the resultant current

consumption of the motor when the state of charge of the battery is getting low. Accordingly, the control system 101 compares the magnitude of the torque demand 106 to a torque cap 108. If the magnitude of the torque demand 106 is greater than the torque cap 108, the torque demand 106 is reduced to the value of the cap. The torque, once checked against the maximum torque cap, is re-labelled and is referred to as the "torque set point" 110. If the magnitude of the torque demand 106 is less than the torque cap 108, the magnitude of the torque demand 106 remains unchanged (but it is nonetheless re ^¬ labelled as the torque set point 110) .

The behaviour of the torque cap 108 is often referred to as a "full-load curve". The full load curve 108 is determined by the torque above which the current starts to approach potentially damaging levels, or otherwise needs to be capped (for example to control electrical power consumption) . It is therefore the torque above which the supercharger is not permitted to operate. Therefore, if at any point, the torque demand 106 in the control system of Figure 1 is greater than the torque cap 108, it will automatically be reduced to that cap .

A full-load curve 108 corresponds to a specific current peak limitation. It may be desirable to provide a plurality of different peak current limitations (for example to offer more flexibility in the current consumption by the motor) . The system of Figure 1 therefore has several full-load curves 108, each corresponding to a different current limitation. The particular full load curve 108 to use, is selected in dependence of the peak current limitation.

Four different control variables of the input current (ON angle, OFF angle, Freewheel and Pulse Width Modulation (PWM) ) govern the torque generated by the switched-reluctance motor. Torque set point maps 105 are used to obtain the input values of each of these control variables required to achieve a particular torque set point 110. Once the control system 101 has obtained the value of the required variable (s), the current is supplied to a switched reluctance motor. This motor input gives rise to a physical response 111, creating a new actual speed 104. This actual speed 104 is fed back into the control system 101 and compared to the target speed 102, and the above-mentioned steps are repeated.

Electric superchargers tend to operate at high speeds (e.g. 50,000+ rpm) . When the motor needs to accelerate quickly, the torque set point is often limited by the

particular full load curve selected (i.e. capped at the maximum torque cap selected) . Figure 2 is a graph showing variation in target speed 102, actual motor speed 104, electrical current 112, torque setpoint 110 and the full load curve 108 as the supercharger motor controlled by the control system of Figure 1 is accelerated from 5000rpm to a target speed of 52000 rpm (for clarity, only y-axis values of speed are shown) . In this case, the full load curve 108 limits the torque set point during the first acceleration phase due to inertial forces, but then the torque set point 110 is below the max torque cap 108 (i.e. full load curve) during the second phase of the acceleration.

The above-described system of Figures 1 and 2 has some drawbacks: Firstly, each of the full-load curves 108 able to be used by the system are empirically determined. The process of empirically determining each curve can be time-consuming. Secondly, the system still only provides a finite number of peak current limits (each full-load curve corresponding to a particular limit) . This can restrict the system's flexibility which may be undesirable for some users (e.g. automotive manufacturers) who may want to tailor the supercharger to specific current usage at specific times. Finally, in the above-described system it is not possible to switch between full-load curves during non-idle operation of the supercharger because a step-change in torque demand could damage the mechanical assembly of the motor.

Summary of the Invention The present invention seeks to mitigate or overcome at least some of the above-mentioned disadvantages.

According to a first aspect of the invention, there is provided a method of controlling a motor in an electric supercharger when instructed to increase the speed of the motor to a target speed, the method comprising the steps of repeatedly :

(i) calculating an adjusted target speed;

(ii) determining the difference between the adjusted target speed and the actual speed of the motor;

(iii) determining a torque set-point for changing the speed of the motor from the actual speed towards the adjusted target speed, and

(iv) supplying electrical current to the motor dependent on said torque set point;

wherein the adjusted target speed is calculated as a function of the time elapsed since the instruction to increase the speed, the function being such that, over a response time, the adjusted target speed approaches the target speed. The present invention recognises that by providing an adjusted target speed (rather than aiming directly for the target speed) , the response of the motor can be controlled. More specifically, by calculating the adjusted target speed as a function of the time elapsed since the instruction to increase the speed, the adjusted target speed can be

controlled to gradually approach the target speed. This can be used to control the peak current supplied to the motor, thereby negating the need for multiple, bespoke, full-load curves.

The behaviour of the function is preferably selectable. The method may comprise the step of selecting the behaviour of the function. By having the behaviour of the function

selectable, the method enables the response time to be

selected (based on the selected behaviour) . The response time typically correlates with the peak current drawn by the motor (the faster the response, the higher the peak current and vice versa) , and thus the behaviour of the function may be selected such that the peak current supplied to the motor remains below the peak current threshold. The method may comprise the step of receiving a peak current threshold signal. The behaviour of the function may be selected such that the peak current supplied to the motor remains below the peak current

threshold. The behaviour of the function may be selected in dependence on the amount of power available in a power source. The power source is preferably the power source for powering the motor. The power source, may, for example be a battery or ultracapacitor.

The function may comprise a filter-function having a time constant. The behaviour of the function is preferably

determined by the magnitude of the time constant. Filter- functions having time-constants are well known per se. Such a filter function typically has a behaviour governed by the term (1 - e ^~t/T ) where t is the time elapsed (e.g. since instructed to increase speed) and τ is the time constant. The function may be such that the adjusted target speed is calculated as the target speed x (1 - e ^~t/T ) .

In preferred embodiments of the invention, the time constant is selectable to determine the behaviour of the function. Thus, the behaviour of the function can be selected based on a plurality, more preferably a multiplicity, or even an infinite number of possible time constants.

The invention may be particularly advantageous in methods in which a full-load curve is used when determining the torque set-point. The step of determining the torque set-point may comprise (a) determining the magnitude of a torque-demand required to change the speed of the motor from the actual speed towards the adjusted target speed, (b) imposing a torque cap on the torque-demand required, and (c) converting the torque-demand required, once capped, into a torque set-point. The torque cap may be based on a torque cap behaviour

previously selected from a plurality of different torque cap behaviours. Alternatively the system may comprise a single torque cap (the plurality of torque caps may be unnecessary by virtue of the advantages provided by the invention) .

The motor is an electric motor. The motor may be a permanent magnet motor. The motor may be a switched- reluctance motor (SRM) . The invention may be particularly advantageous in methods in which the motor is either a

switched reluctance motor or a PM motor because these motors can experience high peak currents (for example shortly after the current is switched ON in an SRM) .

According to a second aspect of the invention, there is provided a control system for controlling a motor in an electric supercharger when instructed to increase the speed of the motor to a target speed, the system comprising: a processor configured to calculate an adjusted target speed;

and wherein the system is arranged to determine a torque set-point for changing the speed of the motor from the actual speed towards the adjusted target speed, and is arranged to supply electrical current to the motor dependent on said torque set point,

wherein the processor is configured to calculate the adjusted target speed using a function of the time elapsed since the instruction to increase the speed, the function being such that, over a response time, the adjusted target speed

approaches the target speed.

By providing a processor configured to calculate an adjusted target speed using a function of the time elapsed since the instruction to increase the speed, the response of the motor can be controlled. More specifically, by

calculating the adjusted target speed as a function of the time elapsed since the instruction to increase the speed, the adjusted target speed can be controlled to gradually approach the target speed. This can be used to control the peak current supplied to the motor, thereby negating the need for multiple, bespoke, full-load curves.

The behaviour of the function is preferably selectable. The behaviour of the function may be selectable from a

multiplicity of behaviours. The multiplicity of behaviours may be such that for each behaviour the peak current supplied to the motor remains below a respective peak current

threshold .

The function may comprise a filter-function having a time constant. The behaviour of the function may be determined by the magnitude of the time constant. The system may comprise a memory module comprising a multiplicity of time constants. The time constant may be selectable from the memory module, each time constant being such that the peak current supplied to the motor remains below a respective peak current

threshold. Such an arrangement may enable the control system to select a particular time constant to ensure the particular peak current threshold is not exceeded. The memory module may comprise a lookup table comprising a plurality of time

constants and a plurality of peak current thresholds, each time constant peak corresponding to a peak current threshold.

The system is preferably arranged to determine the actual speed of the motor. The system preferably comprises a speed- measuring module configured to determine the difference between the adjusted target speed and the actual speed of the motor .

According to another aspect of the invention, there is provided an electric supercharger comprising a motor and a control system as described herein with reference to the other aspects of the invention. The motor in the electric

supercharger is preferably a switched reluctance motor.

The supercharger may be arranged to supply an internal combustion engine with a compressed charge. The motor is preferably arranged to drive a compressor element (such as a compressor wheel) .

The engine is preferably for use in an automobile. The engine is preferably a relatively small capacity engine. The engine is preferably 4 litres or less, more preferably 3 litres or less, and yet more preferably 2 litres or less) . The engine may be in an automobile. The automobile may be less than 3.5 tonnes, and more preferably less than 2 tonnes.

It will be appreciated that any features described with reference to one aspect of the invention are equally

applicable to any other aspect of the invention, and vice versa. For example any features described with reference to the method of controlling a motor according to the first aspect of the invention may be equally applicable to the control system of the second aspect and vice-versa.

Description of the Drawings

Embodiments of the invention will now be described, way of example only, with reference to the accompanying schematic drawings of which: Figure 1 is a schematic showing a previously-suggested control system;

Figure 2 shows the full load curve, torque set point, target speed, electrical current and motor speed during use of a supercharger controlled by the system of Figure 1 ;

Figure 3 is a schematic showing a control system

according to a first embodiment of the invention;

Figure 4 is a schematic showing the logic in a processor for providing the adjusted target speed in Figure 3;

Figure 5 is a schematic showing other logic in the processor for providing the adjusted target speed in Figure 3;

Figure 6 is a graph showing the speed response of the motor when using the system of Figure 3 with ten different time constants;

Figure 7 is a graph showing the current response in the motor corresponding when using each of the ten time constants;

Figure 8 is a graph showing the variation in peak current and response time for each of the ten time constants; and

Figure 9 is a schematic showing the logic in a processor for providing the adjusted target speed, in an embodiment according to a second embodiment of the invention. Detailed Description

As described in the introduction to this specification, in the previously-suggested control system of Figure 1, a control system 101 receives a target speed 102, and calculates a speed error by comparing the target speed with the actual speed. A PI controller 103 then determines the appropriate torque for changing the speed of the motor from its actual speed 104 towards the target speed 102 (i.e. the "torque demand" 106) . The system 101 applies a full load curve 108, which in this example is a constant torque cap. This torque cap is the torque at which a pre-determined maximum peak current is obtained when accelerating the motor. The torque cap 108 ensures this maximum peak current is not exceeded.

As shown in Figure 2, when the motor accelerates, the torque set point 110 matches the torque cap 108 during the whole time the current is supplied (0 to 1.6 seconds) . The current rises sharply to a peak (at around 0.3 seconds) and then settles once the compressor wheel has been accelerated. The magnitude of the peak current is determined by the full- load curve as this determines the torque set point.

The system of Figures 1 and has some disadvantages:

Firstly, each of the full-load curves 108 able to be used by the system are empirically determined. The process of empirically determining each curve 108 can be time-consuming. Secondly, the system still only provides a finite number of peak current limits (because each full-load curve corresponds to a particular limit) . This can restrict the system's flexibility which may be undesirable for some users (e.g.

automotive manufacturers) who may want to tailor the

supercharger to specific current usage at specific times.

Finally, in the above-described system it is not possible to switch between full-load curves during non-idle operation of - li ^¬ the supercharger because a step-change in torque demand could damage the mechanical assembly of the motor.

Figure 3 is a schematic of a control system 1 according to a first embodiment of the invention, which seeks to address the above-mentioned drawbacks. Features in the first

embodiment of the invention that correspond to similar

features in the system of Figure 1, are shown with the same reference numerals as in the first embodiment, but without the addition of the prefix ^λ 1' (or ^λ 10' where appropriate) .

The control system 1 is the same as that in Figure 1 except for the differences described below. Firstly, the control system further includes a processor 7 arranged to receive the target speed signal 2. The processor 7 is configured to adjust this target speed 2 to an adjusted target speed 13, prior to it being compared to the actual speed 2. The

invention recognises that by lowering the target speed to an adjusted target, the control system 1 can more gradually increase the motor speed towards the target speed 2. This, in turn, may reduce the peak current drawn by the motor.

The logic in the processor 7 for adjusting the target speed will now be explained in more detail with reference to Figure 4.

When the values of time constant 15 and time since the speed increase was instructed 19 are non-zero, the processor executes the control logic in Figure 4 to adjust the target speed. Referring first to Figure 4 the processor 7 receives the target speed (RPM_REQ) 9 and the speed at boost start (RPM_BOOST_START ) 17. The processor 7 also receives the time since the speed increase was instructed (TIME_SINCE_BOOST) 19 and a time-constant (TIME_CSTE) 15. As shown in Figure 4, the control logic performs the following control process to calculate an adjusted target speed (RPM_REQ_FLT ) 21: RPM_REQ_FLT = ( (RPM_REQ- RPM_BOOST_START ) (1- EXP (- Time Since Boost/ Time cste) ) + RPM BOOST START

It will be appreciated that the processor 7 thus calculates the adjusted target speed 21 using a function of the time elapsed since the instruction to increase the speed 19. More specifically, in the first embodiment, the function applied by the processor 7 comprises the filter function (l-e ^~t/T ) . This filter function has a time constant 15 such that over a response time, the adjusted target speed 21 gradually

approaches the target speed 2.

For different values of time constant 15, the response of the system will differ. Broadly speaking, the smaller the time constant 15, the less the target speed 2 will be lowered after each loop of the control system 1, so the faster the response time. Equally, the larger the time constant 15, the more the target speed 2 will be adjusted after each loop, so the slower the response time.

Figure 6 shows the speed response (i.e. actual speed 2 of the motor against time) for ten different values of time constant 15 (labelled PI - P10) . As the magnitude of the time constant 15 increases, the speed response becomes more and more gradual. The nature of the speed response directly impacts on the peak current drawn by the motor - a faster response will draw a higher peak current. This is

demonstrated in Figure 7, which shows the current response for each of the ten time constants 15 (P1-P10) in Figure 6 (time constant increasing left to right in Figure 6) . For the small time constants 15, there are notable peaks in current as the motor accelerates. As the time constants 15 increase, this peak current tends to decrease. This relationship is further demonstrated in Figure 7, which shows the relationship between time constant 15 (x-axis) and the peak current and the

response time.

Referring back to Figure 3, the control system 1 of the first embodiment comprises a memory module 23 comprising a lookup table containing the ten time constants 15 (P1-P10) and the peak currents corresponding to each of them (i.e. the data shown in Figure 8) . The data in the memory module 23 is accessible by the processor 7.

During use, the processor 7 receives a peak current threshold signal 25 from the engine control system (not shown) . This peak current threshold 25 is determined by the engine control system and is dependent on the power available to the engine (for example, it may depend on the magnitude of electrical power consumption from other systems in the

engine/automobile) . The processor 7 then selects a time constant 15, from the memory module 23, corresponding to the generation of that peak current threshold 25. Where the memory module does not hold data on the exact peak current threshold received, the processor interpolates between the nearest data above and below the requested peak current threshold, to select an interpolated time constant.

Having a time constant 15 that is selectable, enables the behaviour of the filter function (see Figure 4) to be selected such that the peak current supplied to the motor remains below the peak current threshold. Such an arrangement is especially beneficial because it negates the need for multiple, bespoke, full-load curves; instead the current usage can be varied simply by varying the time constant 15. The control system 1 of the first embodiment of the invention is therefore

especially flexible. Furthermore it does not require

empirically-determined multiple full-load curves.

Figure 5 is a second part of the logic in the processor 7. This is bypass-logic that is used in the event that either the time constant 15 or the time since the speed increase was instructed 13 are zero. If either of these are zero, the logic of Figure 4 is bypassed and the logic of Figure 5 is used such that the adjusted target speed 21 equals the target speed 9.

Figure 9 shows the logic in a processor 207 in a second embodiment of the invention. The other features in the second embodiment, which are not shown in Figure 9, are substantially identical to the first embodiment.

Unlike the first embodiment, the processor 7 does not receive a signal relating to the speed at boost start.

Instead, the control logic just performs the following control process to calculate an adjusted target speed (RPM_REQ_FLT ) 221 :

RPM_REQ_FLT = ( (RPM_REQ) (1- EXP ( -Time_Since_Boost/

Time_cste) )

It will be appreciated that the control system in the second embodiment thus also calculates an adjusted target speed as a function of the time elapsed since the instruction to increase the speed. The filter function is such that, over a response time, the adjusted target speed approaches the target speed.

Furthermore, the time constant 215 is selected from the lookup table 223 in dependence on a max current demand signal 225. The behaviour of the function can thus be selected to tailor the peak current in the motor (simply by using a different time constant) .

The second embodiment does not include the term

RPM_BOOST_START . However, it has additional logic (not shown) to ensure the adjusted target speed does not equal zero when Time_Since_Boost=0 ) ; this is important when the initial speed of the motor is non-zero (for example it has an idle speed of 5krpm) .

Whilst the present invention has been described and illustrated with reference to particular embodiments, it will be appreciated by those of ordinary skill in the art that the invention lends itself to many different variations not specifically illustrated herein. Where in the foregoing description, integers or elements are mentioned which have known, obvious or foreseeable equivalents, then such

equivalents are herein incorporated as if individually set forth. Reference should be made to the claims for determining the true scope of the present invention, which should be construed so as to encompass any such equivalents. It will also be appreciated by the reader that integers or features of the invention that are described as preferable, advantageous, convenient or the like are optional and do not limit the scope of the independent claims.