This action might not be possible to undo. Are you sure you want to continue?
Abstract In this paper, we address the design and development of robust multivariable control design to highlight some of the practical issues with an emphasis on a non-trivial application of Hybrid Electric Vehicle (HEV) Energy Management. Most of the existing energy management approaches are not model-based dynamic systems and/or do not address the concepts of robust stability and performance. By considering a multi-input multi-output (MIMO) dynamics of the HEV model, subject to uncertain parameters and dynamics, it is demonstrated that the two subsystems of the HEV dynamics with strong interactions (ie internal combustion engine and integrated DC motor/generator) are eﬀectively controlled over a deﬁned drive cycle. The employed MIMO mixed-µ control design provides guaranteed stability- and performance-robustness for the plant under investigation and is shown to yield satisfactory performance subject to parametric uncertainties (such as combustion lag, rotational moment of inertia and damping), unmodeled complexvalued uncertainty (due to fueling delay) and unmeasurable exogenous plant disturbance (vehicle load). Simulation results verify that the multivariable control synthesis as a promising approach could yield signiﬁcant improvements on the emission, fuel consumption and calibration time to production of hybrid electric vehicles. Index Terms robust feedback control; multivariable control systems; uncertain systems; hybrid electric vehicle; energy management; fuel eﬃciency; diesel engine modeling.
I. Introduction Hybrid Electric Vehicles (HEVs) have been received with great enthusiasm and attention in recent years . On the other hand, complexity of hybrid powertrain systems have been increased to meet end-user demands and to provide enhancements to fuel eﬃciency as well as meeting new emission standards  which accordingly require more complex (advanced) control system designs to achieve satisfactory performance for the hybrid powertrain. While the main objective of hybrid electric vehicles is to reduce Carbon Dioxide (CO2 ) emissions with an optimised fuel consumption , a signiﬁcant amount of research has been devoted to the ﬁeld of energy management for full Hybrid Electric Vehicles (HEVs) and Electric Vehicles(EV) . There are a number of energy management methods in the literature of hybrid electric vehicles . Among these energy management strategies, a number of heuristics techniques, say e.g. using Rule-based or Fuzzy logic, have claimed to oﬀer some improvements in the HEV energy eﬃciency ,  where the optimisation objective is, in a heuristic manner, a function of weighted fuel economy and driveability variables integrated with a performance index, to obtain a desired closed-loop system response. However, such heuristics based energy management approaches suﬀer from the fact that they guarantee neither an optimal result in real vehicle operational conditions nor a robust performance if system parameters deviate from their nominal operating points. Consequently, other strategies have emerged that are based on optimisation techniques to search for sub-optimal solutions using programming concepts such as linear programming, quadratic programming and dynamic programming , , . Nonetheless, they may not lead to the feasible casual solutions, since they assume that the future driving cycle is entirely known. Moreover, the required burdensome calculations integrated with these approaches have imposed a high demand on practical computational resources which prevent from being implemented on-line in a straightforward manner. Two new HEV energy management approaches have been recently introduced in the literature. In the ﬁrst approach, instead of one speciﬁc driving cycle a set of driving cycles are considered for calculating an optimal control law resulting in the stochastic optimisation approach . A solution to this approach is calculated oﬀ-line and stored in a statedependent lookup table. Similar approach in this course employs Explicit Model Predictive Control , . In this approach, an entire control law is computed oﬄine, whilst the controller will be implemented online based on the lookup table, similar to the stochastic optimisation approach. The lookup table provides a quasi-static control law which is directly applicable to the on-line vehicle implementation. Although this method has potential to perform well for systems with fewer states, inputs, constraints, and ”suﬃciently short” time-horizons , it cannot be utilised in a wide variety of applications whose dynamics, cost function and/or constraints are time-varying due to e.g. parametric uncertainties and/or unmeasurable exogenous disturbances. In other words, any lookup-table-based optimisation approach may end up with severe diﬃculties in covering a real-world driving situation with a set of individual driving cycles.
By utilising this proposed design. An alternative approach is to extend the objective function with a fuel equivalent term. optimisation of the operating points of the each component can still be beneﬁcial. while the second player aims to minimize this performance index. to replace the existing ”non-dynamic” and/or ”non-robust” energy management design approaches. Among these modeling tools developed to represent the hybrid systems. there is a vital need of model-based dynamic controllers with an increased processing speed. Similar to aforementioned methods. to hedge for the uncertain real parameters and the inevitable unmodeled dynamics.g. Hybrid modeling tools have been recently developed to analyse and optimise a number of classes of hybrid systems. have not addressed the MIMO ”directional properties” quantiﬁed by the singular value decomposition (SVD). to name but a few. simulation. have made on-line implementations of predictive control approaches almost impossible. and also due to there being a quasi-static method. A weighting factor is typically included to prevent a drift in the battery from its nominal energy level and also to guarantee a charge sustaining solution. a change in the dynamics of the plant. “directionality” issues predominate and we must use singular value plots to understand the dynamic properties in the frequency domain. Towards this end. Currently applicable techniques. it cannot address the vehicle deriveability requirements. the main drawbacks of the game-theoretic approach are the lack of robustness. Hybrid Toolbox for MATLAB  is developed for modeling. for the real-time application of HEV energy management. one is capable to pose a quantiﬁed robustness analysis in the presence of uncertainties due to. In addition. Any solution to this approach is calculated oﬀ-line and stored in a state-dependent lookup table. The proposed control design. utilise quasi-static models which are not suﬃcient to address driveability requirements . We shall highlight some of the aspects of the multivariable robust control design by representing typical simulation results. the robust controller will be integrated into a HEV system as the ”torque manager”. generation and utilisation of maps defeat the original purpose of designing a dynamic compensator which maintains driveability. and verifying hybrid dynamical models and also for designing hybrid model predictive controllers. The ﬁrst player (drive cycle) wishes to maximize the performance index which reﬂects the optimisation objectives. In a number of applications. is derived based on the dynamic models of the plant and hence driveability requirements are taken into consideration as part of the controller design. HYbrid Systems Description Language (HYSDEL) . such as processing speed and memory. such as game-theoretic based optimisation. The mixed-µ compensator design process in fact ”detunes” an optimal H∞ compensator . However. transfer function matrices describe all relevant dynamic systems.g. For this reason the standard MPC algorithms have been retained away from modern production vehicles. After the design of the robust multivariable controller. MPC is currently applied oﬀ-line to generate the required maps and then these maps are used on-line . Almost all of these hybrid tools. We strongly believe that the results of this work could .A recent approach has endeavored to decouple the optimal solution from a driving cycle in a game-theoretic (GT) framework . In this approach. say e. to minimise emission constraints and fuel consumption. there are a number of model-based energy management methods such as Model Predictive Controls (MPC). eﬀect of time-varying parameters (due to drive cycle) is represented by the actions of the ﬁrst player while the eﬀect of operating strategy (energy management) is modeled by the actions of the second player. The main shortcoming of model predictive control methods is that they can only be used in applications with ”suﬃciently slow” dynamics . designed for the nominal generalised plant. Therefore. for the use of practical hybrid electric vehicle energy management application. have dealt with multivariable hybrid systems as decoupled single-input single-output (SISO) loops and. This approach has been considered in the past. e. we shall overview some of the theoretical and practical aspects of the robust multivariable control using mixed-µ synthesis  for a practical problem of hybrid electric vehicle energy management. we shall refer to Mixed Logical Dynamical (MLD) . and hence are not suitable for many practical applications including HEV energy management problem. Thus.. Towards a practical model-based dynamic controller approach. A recently developed package for the hybrid MPC design is referred to as Hybrid and Multi-Parametric Toolboxes  which is based on the traditional model predictive control optimisation alternatives using generic optimisers. and robust to uncertainties. This lookup tables provides a quasi-static control law which is directly employed for on-line implementation in vehicle. in any MIMO case. . thus. and Piecewise Aﬃne (PWA) models . some methods deﬁne an optimisation criterion to minimise the vehicle fuel consumption and reduce exhaust emissions . however. e. or lack of proper estimation of the vehicle load torque (plant disturbance). In this paper.. This term includes the corresponding fuel use for the energy exchange with the battery in the optimisation criterion . In fact. but it is still remained an arduous task to select a weighting factor that is mathematically sound . a number of inherent hardware constraints and limitations integrated with the vehicle electronic control unit (ECU).g. which makes up the main part of the energy management solution. It should be self-evident that. Two fundamental drawbacks of aforementioned strategies are ﬁrstly their consideration of driveability being an afterthought and secondly the driveability issue is deemed in an ad-hoc fashion as these approaches are not modelbased dynamic. The novelty of this work is indeed in the design and development of the robust multivariable control system with practical signiﬁcance of addressing dynamic vehicle driveability. If only the present state of the vehicle is considered.
and technical reports which have not been taken into account in this study.org for more information. Also. from only 2 papers in 1985 to 552 papers in 2010. when required. 1. hybrid propulsion systems are potentially capable of improving fuel eﬃciency for a number of reasons: they are able to recover some portion of vehicle kinetic energy during braking and use this energy for charging the battery and hence. Section VI will present some of our conclusions. From Figure 1. if the torque request (demanded by driver) is below a threshold torque. Hybrid Electric Vehicles Modern day automotive engineers must. The vehicle energy management development is a challenging practical control problem for which a signiﬁcant amount of research has been devoted in regard to the full HEVs and Electric Vehicles (EVs) in the last decade . It turns out that the internal combustion engine of the hybrid electric vehicle can be potentially designed with a smaller size and weight which results in a superior fuel eﬃciency and reduced emissions . utilise the electric motor at a later point in time. Hybrid vehicle research trend based on the number of publications of the IEEE over the period 1985 to 2010. Compared to the vehicles with conventional ICE. II. we shall extend and analysis the study carried out in  on the number of IEEE publications published between 1985 and 2010. ﬁled patents. This paper is organized as follows.make automotive engineers motivated to continue the design and development of robust multivariable control design and. Hybrid Electric Vehicles (HEVs) have been introduced which typically combine two diﬀerent sources of power. In Section II. These are some representative advantages of the hybrid vehicles compared to those of conventional vehicles. maximize fuel economy and sustain a reasonably responsive car (i. engine downsizing and utilising the electric motor/motors to make up for the lost torque. theses. the traditional internal combustion engine (ICE) with one (or more) electric motors. articles. The concept of sharing the requested power between the internal combustion engine and electric motor for traction during vehicle operation is referred to as ”vehicle supervisory control” or ”vehicle energy management” . we discuss brieﬂy the characteristics of the robust multivariable control system design. of Publications 400 300 200 100 0 1985 1990 1995 Year 2000 2005 2010 Actual Data Linear Fitting Fig. Figure 1 depicts the number of publications recorded at the IEEE database1 whose abstract contains at least one of the strings ”hybrid vehicle” or ”hybrid vehicles”. Recall that these are only publications of the IEEE database . There are other beneﬁts oﬀered by hybrid electric vehicles in general. In Section III. Simulation results will be demonstrated in Section V. we provide in detail the mathematical description of the simpliﬁed dynamical model of the hybrid diesel electric vehicle (parallel type) along with analysing the open-loop dynamics. e. a linear regression analysis of the IEEE publications shown in Figure 1 indicates that research in the ﬁeld of 1 See http://ieeexplore. there are currently extensive academic and industrial research interests ongoing in the area of hybrid electric vehicles as these vehicles are expected to make considerable contributions to the environmentally conscious requirements in the production vehicle sector in the future – see  and other references therein. . is particularly referred to as a control allocation for delivering the required wheel torque to maximize the average fuel economy and sustain the battery state of charge (SoC) within a desired charging range . To tackle such a challenging problem. we will overview the literature of the hybrid electric vehicles.e. using its potential capability. conference papers. In Section IV.g.there are many other publications than those of the IEEE including books. mainly for optimising fuel eﬃciency and reducing CO2 and greenhouse gases (GHG) . to tackle a wide range of applications in the automotive control system designs.ieee. The latter term. 600 500 No. Towards this end. To clarify the above claim. employed throughout this paper. it is obvious that the number of publications in the area of hybrid electric vehicles (HEVs) has been drastically increased during this period. Besides. among other objectives. the ICE can be switched oﬀ as well as during vehicle stop for avoiding engine idling. maintain driveability) while still meeting increasingly stringent emission constraints mandated by the government.
The design of a multi-input multi-output (MIMO) dynamic compensator. utilizing the so-called D.g. . for ﬁxed performance weights. G.and performance. • It is essential to meet end-user demands for increasingly complex new vehicles towards improving vehicle performance and driveability . could potentially fail in providing stability. The greater the parametric uncertainty. these can be analyzed and evaluated by the SVD methodology. Of course. The theoretical foundation for the mixed-µ analysis and synthesis is too copious for a comprehensive review in the framework of this paper.G-K iteration).hybrid vehicles has been accelerated remarkably since 2003.this is a clear evidence to acknowledge that HEVs research and development is expected to make considerable contributions to both academia and industry of production automotive sector in the future. The robust complex µ -synthesis. designed for the nominal generalised plant. In regard to hybrid electric vehicles. . One may also predict that the number of publications in this area could be increased up to about 1000 articles in 2015. when real parameter uncertainty is present in the system. the following key engineering design issues of multivariable feedback control system for the HEV energy management must be taken into consideration: • Controllers that are not robust. are very conservative and yield inferior setpoint tracking performance compared with the compensators designed by the mixed-µ software that implements the full D. . are designed without taking into account the uncertainties. Hence.e. only a summary is provided here. changes in the level of uncertainty associated with the real parameters can have a signiﬁcant impact upon the ‘directional properties’ of the closed-loop system. there are tremendous mandates on Original Equipment Manufacturers (OEMs) to deliver fuel-eﬃcient and less-polluting vehicles at lower costs. the impact of advanced control design for the application of the hybrid vehicle powertrain systems has become signiﬁcantly important .g. Refs. as well as the characteristics of unmeasurable exogenous plant disturbances and sensor noises. 2 has substantial disadvantages. In summary. • There is a continuous increase in the gap between the theoretical control advancements and the practical control strategies being applied to the existing production vehicles. Hence. have been developed that can indeed be used to design robust MIMO feedback control systems with the requisite stability. It is signiﬁcantly important to quantify the performance tradeoﬀ of the reference tracking problem. . that is. in the past decade or so. must possess guarantees of both stabilityand performance-robustness with respect to the explicit performance speciﬁcations posed by the control system designer SSIP05.and performance-robustness. in its present form and as presently is implemented in current numerical Matlab software package . our robust mixedµ design has been carried out using an advanced D. Balas  of the University of Minnesota. the purpose of this paper is to highlight importance for analysis and concrete results related to the performance tradeoﬀs that are always present in MIMO feedback designs with the guaranteed stability. fuel economy. • In robust MIMO feedback designs. i. . performance or both of them. Fortunately. using the so-called D − K iteration. which illustrate the strong interactions of the subsystems. This gap is due to a number of missed opportunities in addressing fundamental functionalities. Multivariable feedback control design All linear time-invariant (LTI) models of real dynamic systems are subject to uncertainty.robustness guarantees. emissions. ‘directional properties’ quantiﬁed by the singular value decomposition (SVD) are important. For a more in-depth discussion the reader is referred to e. that is nearly twice as many as in 2010 .G-K iteration software provided to us by Prof. Moreover. uniﬁcation of control architecture and integration of the Automotive Mechatronics units on board vehicle. e. since we deal with a MIMO design. In particular. the smaller the guaranteed performance. III. the mixed-µ design methodology .  and MATLAB software . tested and implemented in the production hybrid vehicles. to hedge for the worst uncertain real parameters and the inevitable unmodeled dynamics. 2 Although a recent version of Matlab Robust Control Toolbox  includes the mixed-µ design (D. G − K iteration. . G − K iteration . For each LTI model. and of the resulting robust feedback control system. • Controllers which are designed based on the unstructured H∞ or even robust D-K iteration (complex µ ) . the resulting compensator can be arbitrarily conservative.J. • The robust MIMO control system design seem promising advanced tools to bridge this gap should they could be developed. we must take into account both unmodeled dynamics and uncertain real parameters. The robust mixed-µ compensator design process detunes an optimal H∞ compensator. driveability. here are the facts and regulations which must be taken into consideration by automotive engineers: • Due to the ever increasing stringent regulations on fuel consumption and emissions. while still continuing to reduce costs and meeting new emission standards.
G−K iteration leading to the best possible robust compensator. As our intention in this study is to investigate the ”full-hybrid” mode. In this section. we shall investigate how to model a simpliﬁed hybrid electric vehicle by replacing the sophisticated nonlinear dynamic of the diesel internal combustion engine. ). For this reason. the gearbox is shown in Figure 2 but no gear setting was considered in our simpliﬁed HEV demonstration. These low-level subsystems include . These can lead to some unexpected behavior . Diesel ICE Clutch MIMO Robust Controller CIMG Gear Setpoint Torque Commands (high-level static optimisation) Fig. designer-imposed preferences for relative performance associated with diﬀerent signals will also have a signiﬁcant impact on the ‘directional properties’ of the closed-loop system. This is due to the fact that our empirical diesel engine model is derived with engine speed range of ω = [1200. Likewise. • All unmodeled dynamics must be bounded by their maximum singular values and properly introduced in the mixed-µ synthesis. by exploiting some new analysis tools recently developed for the mixed-µ upper bound . it could be widely adopted in new problems of the automotive industry to be addressed by robust multivariable control design. We have employed the mixed-µ software  which fully implements the D. integrated within the typical HEV dynamics. for the ﬁrst time. has been successfully applied to a broad variety of problems (see. Before describing the structure of the simpliﬁed diesel hybrid electric vehicle. most of the common low-level subsystems. clarify the importance of our work carried out in the ﬁeld of advanced energy management for the HEV systems. these can be analyzed and evaluated by the SVD concepts. we used a variety of initial conditions in order to avoid ﬁnding a local minimum. we design a two-input two-output (TITO) control system for the application of energy management of an uncertain hybrid electric vehicle and demonstrate the very important tradeoﬀs between performance and uncertainty. G − K optimization problem is not convex. In this paper. we overviewed the literature of the hybrid electric vehicles (HEVs) which could. Low-level control components such as high voltage electric battery. in some extent. Schematic structure of a parallel 4 × 4 Hybrid Electric Vehicle (HEV). The complex-µ synthesis procedure as an eﬃcient robust design tool. electric rear axle drive etc are excluded in this high-level energy management conﬁguration. and also for generating electric energy via charging a high-voltage battery (not shown in the ﬁgure). The hybrid electric vehicle structure shown in Figure 2 is equipped with a turbocharged diesel engine and a crankshaft integrated motor/generator (CIMG) which is directly mounted on the engine crankshaft. Diesel HEV model In Section II. Again. since the formulated D.In robust MIMO feedback designs with ﬁxed uncertainty for the real parameters. we shall assume that the integrated ICE-CIMG clutch is fully engaged and hence our descriptive HEV dynamical model (see §IV-C) excludes a clutch dynamics as it is shown in Figure 2. 2. which is (optimally) robust to structured mixed uncertainty is very diﬃcult. let us ﬁrst overview a generic HEV structure. however. The mixed-µ synthesis problem extends the above procedure to the mixed real/complex uncertainties case. • Examine and determine the actual closed-loop stability regions for diﬀerent types of ‘legal’ unmodeled dynamics. The problem of synthesizing a controller with existing control theory and computational tools. Because of this non-convexity. It is also worthwhile to emphasise that our design methodology on the development of the HEV energy management is a high-level design strategy. A representative conﬁguration of an advanced 4 × 4 parallel hybrid electric vehicle conﬁguration is shown in Figure 2. in our studies. are not considered in the HEV conﬁguration as shown in Figure 2. • IV. We shall integrate this tractable uncertain HEV model (subject to real and complex uncertainties). The CIMG is used for starting and assisting the engine in motoring-mode. for example. 2000]rpm running at the ﬁrst gear. with recent advances on robust multivariable control design using mixed-µ synthesis.
a diesel internal combustion engine (ICE) and an armature-controller DC electric motor. A. ς [msecs]. to name just a few. this model shall be used for the simulation of the overall vehicle motion (at longitudinal direction). Such a total torque loss is typically a nonlinear function of the engine speed and could be represented by table mapping. the severe cross coupling. control engineers should possess a good comprehension of the physics of the plant under investigation. arising e. but rather a high-level torque manager model that will generate control eﬀorts based on a given set-point torque commands. air ﬂow). The speed-dependant fueling delay (τ ) and combustion lag (ς ) are given in Table I. For instance. which are the quantities of interest. low level clutch control. The mathematical modeling of these two subsystems will be discussed in the remainder of this section. For designing a well balanced feedback control law. it is not usually possible to obtain a detailed diesel engine data (or model) from the production vehicle manufacturer . could be applied. However. it is essential to obtain a straightforward and realistic model of the propulsion system to which an eﬃcient control strategy. τ is the speed-dependant time-delay due to fueling dem course and TB is the mapped fueling input representing the required ICE crankshaft (brake) torque. The ICE indicated torque Tind is assumed to be mapped from the delayed fueling input proportionally. τ [msecs]. obtaining a precise mathematical model of a HEV powertrain is a very challenging task particularly due to multienergetic nature and switching dynamics of a hybrid powertrain. CIMG low-level motor control. combustion process. As stated earlier. Our simpliﬁed diesel engine model is empirically derived using a turbo-charged diesel engine at speed range of ω = [1200. we will have Tind (t) = 1 T dem (t − τ (ω )) ς (ω )s + 1 B (1) where ς is the speed-dependant time constant due to combustion lag. the frequency range of the engine system. In a mathematical representation. 100]NM. Simpliﬁed diesel engine model In this section. Our diesel model also contains a speed-dependant torque loss TLoss arising due to friction torque. for the purpose of the studied operating range of engine speed and brake torque. There are two main reasons to highlight the importance of simpliﬁed HEV dynamical models: First. This HEV dynamical model is modeled using two subsystems. due to combustion and turbo lag. and the limited output sensor dynamic capabilities all contribute to make this modelling step a most arduous task . we shall present a simpliﬁed dynamical model of a turbo-charged diesel engine. Generally speaking. For the above reasons. such as our proposed robust mixed-µ design methodology.ω 1200 1200 1600 1600 2000 2000 TB 50 100 50 100 50 100 τ 100 140 84 96 80 72 ς 144 142 140 137 140 134 TABLE I Experimental results of fueling delay.g.e. ancillary torque and pumping loss. Recall that this torque management structure could be easily adopted to other engine conﬁgurations in a straightforward manner. high voltage battery management. 2000]rpm with operational brake torque acting at TB = [50. and electrical distribution including DC-DC converter. Therefore. the inherent sampling nature of the four cycle internal combustion engine. low level transmission control. heat distribution. One of the challenging aspect of any model based engine control development is the derivation of a simpliﬁed yet insightful model. and for the ease of development of an advanced HEV energy management system. Secondly. Furthermore. This simpliﬁed model is based upon the nonlinear diesel engine dynamics and the fact that it must capture both the transient and steady-state dominant modes of the diesel engine during operational conditions. our developments towards a simpliﬁed hybrid model are based on a high ﬁdelity simulation model of the overall diesel hybrid electric vehicle. we do not intend to utilise any detailed model of the internal engine processes. These results are captured by measuring the step response of the engine to a step change in the engine brake torque. and combustion lag. and has limited bandwidth due to internal combustion dynamic eﬀects. the nonlinearities associated with the internal engine processes (i. namely . the dynamics of the engine model includes the average torque responses of both diesel engine and CIMG over all four cylinders. as functions of diesel engine speed [rpm] and brake torque [NM].
as discussed earlier. without loss dem dem of generality. Also. and TL is the vehicle load torque representing the plant disturbance. ﬂywheel. due to there being in hybrid mode. The diesel dynamic shown in Equation (2) will be used in the overall conﬁguration of the HEV dynamics. respectively. not discussed in this study (see also Figure 2). C. Regarding the fact that the engine speed is synchronised with that of the CIMG in the ”full-hybrid” mode. va is control eﬀort as of armature voltage. Simpliﬁed CIMG Model Assuming that the hybrid electric drivetrain includes an armature-controlled CIMG (DC motor). the details of estimation approach are not discussed here. the total engine torque loss is well approximated by a linear function of ω modeled as TLoss = mω where m = 0. A schematic representation of the simpliﬁed parallel diesel hybrid electric vehicle model is shown in Figure 3. Ia is armature current. J is the eﬀective combined moment of rotational inertia (including CIMG rotor. For the sake of simplicity. one could assume that both ICE and CIMG output torque signals are measured with inherent sensor noises to represent the estimation errors. the rotational dynamics of the hybrid drivetrain (including inertial crankshaft and motor) is given as follows: Jω ˙ + bω = TB + TM − TL (4) where ω is the drivetrain speed. The simpliﬁed diesel engine model can be described as the following state-space equations: 1 dem x ˙1 = −1 (t − τ ) ς x1 + ς T B Tind = x1 TB = Tind − TLoss (2) where x1 is the state associated with the combustion lag dynamics. respectively. in Figure 3 the engine brake torque TB and the CIMG torque TM are estimated torque feedback signals. However.ω = [1200. b is the eﬀective joint damping coeﬃcient. the gear setting is disregarded at this simpliﬁed model. 100]NM. B. see  for more details. driveshaft and wheels). respectively. From Equations (2) and (5). and La and Ra are inductance and resistance of the armature. Furthermore. it is assumed that the ICE/CIMG clutch is fully engaged and hence the clutch model is excluded from the main HEV dynamics . . The armature-controlled CIMG model as of Equation (3) along with the rotational dynamics of Equations (4) could be integrated into the following state-space equations: kb 1 dem x ˙2 = L − Ra x2 − L v x3 a a a km b+m 1 1 TL x ˙ 3 = J x1 + J x2 − J x3 − J TM = km x2 (5) where x2 and x3 are the states associated with the armature current (Ia ). and drivetrain speed (ω ). driveline. The mathematical dynamics of the CIMG could be represented as Ia vemf TM 1 (v dem − vemf ) La s + Ra a = kb ω = km Ia = (3) dem where km and kb are torque and back emf constants. we can now build the proposed simpliﬁed 3-state HEV model. a simpliﬁed but realistic simulation HEV model with a detailed component representation of diesel ICE and DC electric motor (CIMG) will be used as a basis for deriving the HEV model in the subsequent section. the applied voltage va controls the motor torque (TM ) as well as the angular velocity ω of the shaft. req req Recall that the setpoint torque commands (TB and TM ) are provided to the controller by a high-level static optimisation algorithm. In addition. crankshaft. 2000]rpm and TB = [50. Also.it was previously shown in HEV schematic diagram of Figure 2. Simpliﬁed diesel hybrid electric vehicle model Based on the state-space representation of both the diesel ICE and electric CIMG. given in Equation (2) and Equation (5).12 (ω in [rad/sec]). the look-up mapping table of CIMG torque request (TM ) vs armature voltage request (va ) is simply assumed to be an unit gain in this model. vemf is the back emf voltage.
It is worthwhile to emphasise that combustion lag ς is not a constant uncertain parameter . in practice. the plant includes one actuator unmodeled dynamics described by the uncertain value in the fueling control time delay. hence. from a theoretical perspective. respectively. suﬃcient slowly-varying parameters could be dealt as constant uncertain parameters . the real uncertainties ς . J b are the uncertain ranges and (δς . driveline. and the eﬀective damping coeﬃcient b. By omitting the dependance on time-delay for the sake of simplicity. However. For the mixed-µ synthesis modeling. crankshaft. while the J uncertainty occurs in the quotient form 1/J .δb ) ∈ R determine the structured uncertainty values and satisfy |δς | ≤ 1. Simpliﬁed model of the parallel Diesel Hybrid Electric Vehicle. We shall assume the similar case for combustion lag ς and fueling delay τ to be modeled as constant uncertain parameters in the robust mixedµ design.TLoss TBdem e -τ( ω )s ςω ( 1 )s Tind +1 Engine torque loss vs driveline speed mapping Fueling delay Combustion/turbo lag TB TL 1 Js + b ω dem TM dem va Motor torque vs armature voltagemapping kb La s + Ra TM Load (including motor rotor. u = [TB ( t − τ ) TM ] and T y = [TB TM ] are control signals and HEV torque outputs. using either structured or unstructured uncertainty models. By considering the linear fractional transformation (LFT) for the above three real parametric uncertainties.it is rather a time-varying speed-dependant parameter. 3. described by the lower-loop LFT transfer functions  as . the hybrid inertial load J . |δJ | ≤ 1 and |δb | ≤ 1.δJ . it turns out that the parameters ς and b appear in multiplicative form. three uncertain real parameters associated with the diesel engine combustion lag ς . The robust LTI MIMO control synthesis cannot deal with slowly-varying parameters. As shown in Equation 6. as also evidenced through our simulations presented in Section V. given by the delay time τ . 1) Real parameter uncertainty: The real uncertain parameter ς . J b are the nominal values. J and b are described by ς J b = ς ¯ + δς ς ˜ ¯ + δJ J ˜ = J ¯ ˜ = b + δb b (7) ¯ and ¯ ˜ and ˜ where ς ¯. ς ˜. flywheel. the overall state-space equations of the simpliﬁed HEV model is represented by 1 1 0 0 −ς 0 0 ς kb 1 x + 0 TL u+ 0 x ˙ = 0 −Ra − L La a 1 km 1 m+b −J (6) 0 0 − J [ J ] J 1 0 −m y= x 0 km 0 dem dem T where x ∈ R3 is the state of the HEV system obtained from Equations (2) and (5). These are modeled according to the mixed-µ synthesis framework. driveshaft and wheels) v emf Armature kb Fig. 1/J and b are regarded as scalar gain blocks.
The maximum value of the fueling time-delay uncertainty is τ ≤ τmax = 140 msecs. The transfer function Wun (s) is multiplied by a delta block ∆un (s) ∈ C that satisﬁes ||∆un (s)||∞ ≤ 1. 0. which introduces a phase uncertainty in the range of ±180.5N m/(Rad.s).5]N m/(Rad. J ˜ = 0. The frequency-domain upper bound for the unmodeled time-delay is required for mixed-µ synthesis design and is the magnitude of the transfer function 2. 10 0 −10 Magnitude (dB) −20 −30 −40 eM(jω) wτ(jω) −50 −1 10 10 0 10 Frequency (rad/sec) 1 10 2 10 3 Fig. so the pole of the Wun (s) transfer function is set near the maximum time-delay frequency of 15rad/secs – likewise.5.] ) ς ¯ ς ˜ ς = FL .25]N m/(Rad. The multiplicative error magnitude. (10) Fig.1 ¯ = 1N m/(Rad.s2 ) J ∈ [0.s) → ¯ b = 0.15]N/m → ς ¯ = 0. eM (s).14N/m. there is a time-delay τ in the ICE fueling control channel which represents an unmodeled complex-valued dynamics.15s Wun (s) = (11) s + 15 which is shown in Figure 5 along with the multiplicative error magnitude due to unmodeled dynamics arising from ICE fueling control channel delay. the gain is adjusted until Wun forms a tight upper bound for the magnitude of the eM (s) transfer function.15N m/(Rad.05.s2 ) → J b ∈ [0.s2 ). 0. δJ 1 ] −J ([ ) ¯ b ˜ b b = FL . Block diagram of the fueling time-delay error model. 4. by a complex-valued uncertainty. ς ˜ = 0. δb 1 0 The real uncertain parameters values for our case study are ς ∈ [0. . δς ( 1 ) [ 0 ] ˜ 1 −J 1 1 ¯ J = FL J ˜ . Frequency response of unmodeled fueling time-delay dynamics for τmax = 140 msecs and its frequency-domain upper bound |Wun (jω )| of Equation (11). 1.1N m/(Rad.13. which is an indeﬁnite dimension block. we shall consider a frequency upper-bound to be represented as a multiplicative error eM (s) = e−τ s − 1 as shown in Figure 4. In order to model such a ”pure” ICE fueling time-delay. can then be approximated by a high-pass transfer function Wun (s) with a real pole. 5. ˜ b = 0.s) ([ (8) (9) 2) Complex-valued unmodeled dynamics: In addition to the three real uncertainties.
2(s + 100) Wact (s) = diag . 0. The greater the performance parameter. they have ±400NM and ±200NM limits on their deﬂection which are considered in designing the frequency-domain control design weights. with continuous-time white noise ξ (t). we have chosen constant measurement noise of Wn =diag(0. • The CIMG handling quality response from the CIMG torque request input to the CIMG torque output should 144 match the damped second-order response HQCIM G = s2 +16 . Also. denoted by Wact (s). The desired setpoint tracking problem requires that the eﬀects of plant disturbance TL (t) (primarily) and also sensor noise (estimation errors) be minimized. the (unmeasured) plant disturbance (vehicle load TL (t) in our case study). Nonetheless.05) indicating of %5 error due to the estimation (or measurement noise) on the torque output signals.1(s + 60) 0. as follows: TL (s) = Wd (s) ξ (s) (15) α (16) s+α where the frequency range ω ≤ α is where the vehicle load has most of its power. We select a control frequency weight. In order to design the robust feedback controller the following type of performance weight upon the output vector is used: ] [ 1 1 Ap 0 (12) Wp (s) = 0 A2 s+α p which reﬂects our performance speciﬁcation (setpoint tracking) for the frequency range ω ≤ α rad/sec where the vehicle load disturbance TL (t) has most of its power. There are also frequency-domain weights on the measurement noise Wn . We have assumed that performance variables (ICE torque and CIMG torque outputs) are measurable. In either case. thus. D. The ICE torque control and CIMG torque control actuators have ±100NM/sec and ±300NM/sec limits on their deﬂection rates. an estimation algorithm is required to estimate the HEV output torques. we must also select frequency-dependent weight matrices upon the disturbances. 0. the selected measurement noise can be translated into modeling of the torque estimation errors.0033 (14) (s + 600) (s + 1000) so that we allow larger controls in lower frequencies and we penalize for large controls at much higher frequencies. • The ICE desired torque (handling quality (HQ)) response from the ICE setpoint torque to the ICE brake torque 100 TB should match the damped second-order response HQICE = s2 +16 s+100 . with zero mean and unit intensity. this might be infeasible and. Wd (s) = with . Aj p are the performance parameters – see Table II. To carry out the mixed-µ design. Frequency-dependant weights Our design goal is to have the ”true” MIMO HEV respond eﬀectively to the ICE and CIMG setpoint torque requests. The form of Wact (s) used is ( ) 0. is modeled by a stationary stochastic process generated by driving a low-pass ﬁlter. In reality. Also. in order to penalize the control actions diﬀerently in diﬀerent frequency regions. We do not intend to discuss details of the designed estimators in this paper for the sake of readability. This is used to limit the bandwidth of the closed-loop system. 0. To capture the limits on the actuator deﬂection rates. Wd (s). which determines the pole location of the ﬁlter in the derivative action.01.An actual time-delay of τ (ω ) mapped from engine speed and brake torque (see Table I) will be used in our simulation. . without loss of generality. the following control rate weights are used in the robust mixed-µ design: [ ] 1 AE = s s [ 1+ N ] (13) 1 AM = s s 1+ N where N speciﬁes the derivative coeﬃcient.05. These performance speciﬁcations include: • Decoupled responses from ICE setpoint torque to the ICE brake torque TB and from the CIMG setpoint torque to the CIMG output torque TM . sensor noises. outputs and control signals as follows subsequently. the superior the torque reference tracking performance.8s+144 .
Note that the worst case uncertainties have most eﬀect over the frequency range ω ≤ 0. E. 6. Moreover.The frequency-dependant design weights of Equations (11)-(16) together with the state-space equations of Equation (6) are shown in Figure 6 which will be used in designing the robust multivariable feedback control using mixedµ synthesis described in Section III.1 rad/sec. Frequency-domain analysis dem dem T The open-loop singular value decomposition of the uncertain MIMO HEV model from the input u(t) = [TL . even for the simpliﬁed LTI HEV system. δb . 20] rad/sec. . there are large variations in the resonant peaks over the frequency range [2. there is a huge gap between the maximum and the minimum singular values of the worst case uncertain parameters (around 3 dB) which will create signiﬁcant challenges for control system design. and complexvalued unmodeled dynamics δς and ∆un . Singular values decomposition of the open-loop uncertain hybrid electric vehicle dynamic. TM ] are shown in Figure 7 for diﬀerent values of real uncertainties δJ . Some representative simulation results of HEV energy management case study will be shown in Section V to highlight some advances of our proposed MIMO control system design. TM ] T to the output y (t) = [TB . 30 20 10 Singular Values (dB) 0 −10 −20 −30 −40 −50 −60 10 −2 10 Frequency (rad/sec) 0 10 2 10 4 Fig. Also. Wp Robust Control Design for HEV energy management problem. 7. TB . δ I 0 J 2× 2 δb δς ∆ 0 un w I CE Torque Setpoint ICE Torque Dem and z TL CIMG Torque Setpoint K AE CI MG Torque Demand AM TBdem dem TM HEV eact TB TM W n Sensor noise Wact HQICE HQCIMG ep Fig.
The plant initial condition x0 ∈ R3 is assumed zero in the simulations.01 20 190 50 Unit msecs msecs msecs Ohms Henrys NM. If these torque output signals are not measurable.Amp−1 Volts.25 0. Parameter Sampling time (ts ) Maximum ICE fueling delay (τ ) ICE combustion lag (ς ) Motor armature resistance (Ra ) Motor armature inductance (La ) Motor torque constant (km ) Motor back emf constant (kb ) Eﬀective drivetrain inertia (J ) Eﬀective drivetrain damping (b) Bandwidth design frequency (α) ICE torque performance parameter (A1 p) CIMG torque performance parameter (A2 p) Derivative ﬁlter coeﬃcient (N ) Value 8 140 [130-150] 1 0.0. We shall analyse and evaluate the closed-loop behavior of the HEV energy management system using the robust feedback control design applied to the high-ﬁdelity simpliﬁed model of the HEV described in Section IV. V.g. nB and nM can also be translated as appropriate torque estimation errors. δς . nB and nM are torque sensor noises when the torque output signals are available to measure (e. Our simulations have been carried out in Simulink and implemented in discrete-time using a zero-order hold with a sampling time of ts = 8 msecs – see Table II. 8.secs. ∆un ) and the robust TITO controller K (s) to be designed using mixed-µ synthesis.A mixed-µ synthesis controller was designed for this system using the frequency-dependant weighting functions integrated into the generalised plant P (s).3 0. TL TBReq T Req M ep nM nB P( s ) eact T dem eM K(s) Fig. we shall present the results of robust multivariable control design described in Section III for the application of the simpliﬁed HEV energy management system discussed in Section IV.5] [0.sec−1 – – – TABLE II Physical constants of the HEV model and control system design parameters. Simulation results In this section. as shown in Figure 8.05.25 [0. the uncertainty block ∆ = diag (δJ I2×2 . For the uncertain HEV energy management application an equated LTI discrete-time system of the continuous-time state-space dynamics described in Equation (6) is obtained using a sampling interval ts (see Table II).25] 0. Robust control design structure using mixed-µ synthesis for the HEV energy management problem.m2 /s2 Nms rad. The parameters used in the control system design along with other physical constants of the HEV model dynamics are provided in Table II. δb . The robust control diagram shown in Figure 8 includes the generalised HEV model P (s). along with the HEV model uncertainty and performance objectives. when using dyno). .5-1.rad−1 kg.
as illustrated in Figure 9(a).We shall emphasis that the developed robust mixed-µ synthesis. by requesting large torque commands. Figure 9 shows typical simulation results for the period of 20 secs in tracking the HEV requested setpoint torques. which illustrate the strong interactions of the two subsystems. we have in fact violated our empirical HEV modeling assumption in that drivetrain speed is limited to ω = [1200. at 2. both the diesel ICE and the DC electric motor could satisfactorily track the requested torque setpoints. in designing the robust MIMO control system we require to utilise an LTI model of the HEV energy management plant hedging on worst case of the ICE speed-dependant parameters. directional properties quantiﬁed by the singular value decomposition (SVD) are signiﬁcantly important. As shown in Figure 9. Figure 10(c) shows that the robust MIMO mixed-µ fully satisﬁes the required optimisation constraints due to actuator limitations. Simulation results of the HEV Torque setpoints and outputs using our proposed robust control design. It is worthwhile to point out that as illustrated in Figure 11(a). Fig. in the simulation.4GHz under Windows 7 Pro platform. Figure 10 shows the vehicle load torque used for the purpose of our simulations. The vehicle speed shown in Figure 11(b) has been calculated using a dynamic model of the vehicle as a function of the drivetrain speed which is not discussed here. does not require knowledge about future driving conditions. However. It is also important to mention that fueling delay and combustion lag are functions of engine speed and brake torque – see Table I. Conclusions The aim of this paper was to present a robust MIMO feedback control design with an application for the energy management of hybrid electric vehicles (HEVs). 60 45 40 40 35 TB [NM] 20 TM [NM] 30 25 20 20 15 0 5 10 Time (secs) 15 20 Treq M TM 0 Treq B TB −20 0 5 10 Time (Secs) 15 (a) Engine Brake Torque. The designed robust controller is then enquired to optimise control eﬀorts so as to track the requested torque references at the HEV model outputs. We shall assume that the plant disturbance (vehicle load) is unknown while robust controller attempts to minimise its eﬀects on the performance outputs. This fact will make implementation of all sort of H∞ optimisation based algorithms more tractable to be applied in real dynamical applications. During this simulation period. assumes that setpoint torque commands are provided by some sort of static optimisation algorithms. due to there being a dynamic control system. The main goal of energy management in hybrid electric vehicles is to . the system is in hybrid mode as both ICE torque and CIMG torque are requested. respectively. However. Such future driving conditions in our case study include setpoint torque commands (requested by driver) and vehicle load torque. at times t = 5 secs and t = 15 secs . Towards this end. Figure 11 shows simulation results in regard to drivetrain speed and vehicle speed. VI. we assume that ς and τ are suﬃciently slow. Recall that. the TITO controller is requested to follow the commands for an increased and decreased ICE torques. and the controller has successfully delivered this torque request. However. which are accordingly considered as constant uncertain parameters in our design. we have utilised the actual time-varying (speed-dependant) parameters. as depicted in Figure 9(b). Regarding the real-time simulations in Simulink (ﬁxed-step) using a sampling time of ts . since we deal with a MIMO design. as discussed earlier. the simulation time required for a single run of 20 secs was approximately 20 times faster than real-time running a Toshiba Portege laptop with an Intel(R) Core(TM) i5 processor. Similarly. Recall that our torque manager structure. there was an increased request for the CIMG torque (from 20 NM to 40 NM) at time t = 10 secs. 2000]rpm. it can be seen that the robust controller could satisfactorily control the HEV energy management dynamics in real-time. 9. (b) CIMG Torque. Despite the fact that the HEV energy management is a highly-coupled MIMO dynamical system. ICE torque loss and control eﬀorts generated by the robust mixed-µ controller.
10. The proposed robust design architecture could resolve shortcomings of the existing non-robust control design methodologies. Torque loss. 11. Simulation results illustrated that the robust multivariable control design could be a very promising control system design methodology while it could play a key role in a wide variety of . reduce the CO2 emissions with an enhanced fuel consumption for a hybrid powertrain control system. Simulation results of parallel diesel HEV drivetrain speed and vehicle speed. The applicability of existing controllers in the energy management setting. 300 u(t) dem T B dem TM 200 100 0 0 5 10 Time (Secs) 15 20 (c) Control signals. Simulation results of vehicle load. 3000 2500 w (rpm) v (mph) 0 5 10 Time (Secs) 15 20 2000 1500 1000 500 25 20 15 10 5 0 0 5 10 Time (Secs) 15 20 (a) Drivetrain Speed.50 40 TLoad [NM] 30 20 10 0 TLoss [NM] 40 Ancilary Torque Friction Pumping loss Total losses 30 20 10 0 5 10 Time (Secs) 15 20 0 0 5 10 Time (Secs) 15 20 (a) Vehicle (load) Torque. The singular value decomposition analysis were shown to address directionality so as to understand the dynamic properties in the frequency domain. the dynamical robust multivariable control design does not explicitly utilise any knowledge in regard to the future driving cycle. has shown a main drawback of these algorithms as most of them do not address the problems of stability. Fig. however. Moreover. and Control eﬀorts.and performance-robustness tradeoﬀ and are not dynamic modelbased approaches – this turns out that the driveability aspect in the existing control designs is an afterthought. (b) Vehicle Speed. Fig. 400 (b) Torque Loss.
Khayyam et al. CRC Press. Prentice-Hall. Warwick. Int. Young. McFarland & Co Inc. On Control Systems Technology. Apr. 1989. Sciarretta and L. Feb. Prentice-Hall. Of the American Control Conf. Korea. 1991.M. In Proc. USA. Morary. Structured singular value approach for systems with parametric uncertainty.J. 2002. Jan.  V. Hybrid Vehicles. Bailey.  Bemporad et al. June 1994.  Iqbal Husain. De la Pena et al. 36(1):25–38. AK.G. and Predictive Control: An Application to Hybrid Electric Vehicle Energy Management. 2004. HEV control strategy for real-time optimization of fuel economy and emissions. Guzzella. Optimization of powertrain operating policy for feasibility assessment and calibration: Stochastic dynamic programming approach. June 2004. In Proc. In In the 2nd International IEE Conference on Automotive Electronics.  M.C. Robust and Optimal Control. Of the IEEE Int. The Netherlands. Cundev.C. IEEE Control Systems Magazine. IEEE Transactions on Automatic Control.unisi. 2006. IEEE Trans. 1981. 1996. 11:653–680. Regenerative braking system for a hybrid electric vehicle. Bemporad and M. Kessels.dii. Sontag. 2006.  C. Control of systems integrating logic. Ripaccioli et al..  S. . 26(2):346–358. http://www. InTech Publisher. Doyle. pages 1425–1430. Florida. In In Proc. Fan et al. June 2011. Optimal energy management for mild hybrid operation of vehicles with an integrated starter generator. Kolmanovsky et al. Dextreit et al. pages 2333–2337. Controller design with mixed uncertainties.J. UK. Oct.  K. State-space solutions to standard H2 and H∞ control problems.  D. Electric and Hybrid Vehicles: Design Fundamentals. Use of approximate dynamic programming for the control of a mild hybrid. of the IEEE Conf. Hysdel — a tool for generating computational hybrid models. In In Proc. June 2008.  G. Bemporad. Hybrid Toolbox . In Proc. In In Proc. on Automatic Control. chapter Fast Model Predictive Control and its Application to Energy Management of Hybrid Electric Vehicles.  A. µ analysis with real parametric uncertainty. Paris. Van Druten.J. Design optimisation and optimal control for hybrid vehicles. Assadian. In In Proc. M. pages 1251–1256.  F. Evaluating the impact of driveability requirements on the performance of an energy management control architecture for a hybrid electric vehicle. InTech Publisher. of Robust and Nonlinear Control. Dec. USA. µ -Analysis and Synthesis Toolbox of Matlab. pages 1230–1235. 2004. Fekri. Robustness in the presence of mixed parametric uncertainty and unmodeled dynamics. of the International Conference on Engineering Optimization. 2009. May 2002. IEE Savoy Place. Automatica. The MathWorks Inc. Hofman and R. Rousseau et al. Control of hybrid electric vehicles. 38:3–20.  S. Steinmaurer and L. Vasconcelos and S. Electric and Hybrid Cars: A History. on Automatic Control. of the IEEE Conf. April 2005. Cacciatori et al. Energy Management for Automotive Power Nets.  G. Aug. 34(8):831–847.D.and performance-robustness tradeoﬀs: Mimo mixed-µ vs complex-µ design.  A. chapter Hybrid Modeling.  I. In Proc. Young and J. on Decision and Control. pages 58–72. Fuzzy logic control for parallel hybrid vehicles.  J. MD.  P. April 2000. 51:1399–1403. 1st edition.E. 2004. on Decision and Control.it/hybrid/toolbox. On Vehicular Power and Propulsion. USA. Private communication re mixed-µ software.  E. Int. Baltimore. Brazil. Symp. June 2007. 1999. In Proceedings of the 17th IFAC World Congress. Beyond singular values and loopshapes. PhD thesis.  M. Anchorage. J. Englewood Cliﬀs. AIAA Journal of Guidance and Control. Jan. Seoul. 1991.  G.C. of WMG Hybrid Conference. SpringerVerlag.M. Model Predictive Control.  B. Faculteit Electrical Engineering. 1998. chapter Intelligent Energy Management in Hybrid Electric Vehicles. 35:407–427. Fast model predictive control using online optimization. Mechatronic design and control of hybrid electric vehicles. UK. dynamics.  P. 2010. Hybrid electric vehicle energy management using game theory.  J. J.  N. May 2002.M. Dec. Proc. of the Future Car Congress. Young. Fekri and F. Schouten et al.  A. pages 3129–3134.  H.  M. Aug. Cikanek and K.H. LAP LAMBERT Academic Publishing. Johnson et al. NJ. In In Proc. Michigan. France. Nonlinear regulation: The piecewise linear approach. References  Curtis D. Energy Management. In In Proc. Orlando. 2007.  E. 1990. June 2008. Explicit Hybrid Model Predictive Control of the dc-dc Boost Converter. Technische Universiteit Eindhoven. 2003. On Mechatronics. 2003. In IEEE PESC.challenging complex automotive applications. 1991.H. Narciso et al. of the 12th International Conference on Hybrid Systems Computation and Control (HSCC 2009).  Allen Fuhs.M. of the SAE World Congress. March 2010. Athans J. Detroit.K.  Wang and Boyd. 2008. 2001. Young et al. NJ. Torrisi and A.. Hybrid-electric Cars and Control Systems: Control strategy of car hybrid system and its experimental conﬁrmation. Identiﬁcation. May 2002. 2008. AK. Baumann et al. 12(2):235–249. Ramsbottom and F. Zhou et al. University of Warwick.  P. Stein and J. Anderson and Judy Anderson.C. of the American Control Conf.  G. Doyle. pages 6974–6979. Rio de Janeiro. on Control Systems Technology. IEEE Trans.A.R. Doyle et al. Computation of µ with real and complex uncertainties. Energy analysis of hybrid vehicle powertrains.M. volume 5.  G. July 2008.User’s Guide. Washington DC.  D. March 2000. In IEEE/ASME Trans.C. of IEEE International Engineering Management Conference.  K. April 2009.  P. Assadian. April 2007. Essentials of Robust Control. EngOpt 2008. IEEE Trans.  A.  G. In In Proc... User’s Guide. Automatica.  D. of the American Control Conf. IEEE Trans. Zhou and J. Balas et al. and constraints. USA. pages 1 – 5. Anchorage. Jan. In SAE World Congress & Exhibition. Stability. Balas. 2nd edition. 10(3):460–468.  T. Del Re. pages 60–70. of Robust and Nonlinear Control. Robust explicit MPC based on approximate multiparametric convex programming.D. 2008. IEEE Trans. A framework for multi-parametric programming and control — an overview. IEMC Europe 2008. Beccuti et al. Bemporad. 2006. 1st edition. The explicit linear quadratic regulator for constrained systems. Doyle. CRC Press. Englewood Cliﬀs. on Automatic Control. 14(1):5–16.
of IEEE International Conference on Control Applications (CCA 2008). 1996.  J. of Electrical Engineering and Computer Science. Kaszynski and O.  M. G. Massachusetts Institute of Technology. Sawodny. USA. The Control Handbook. MA. pages 7–12. 1980. 2008. Lewis. Packard. 2005. Portugal. pages 671–687. The Structured Singular Value (µ ) Framework. Balas and A. Instituto Superior T´ ecnico (IST). Fekri. Levine. PhD thesis. Modeling and identiﬁcation of a built-in turbocharged diesel engine using standardized on-board measurement signals. Dept. Automotive engine control: A linear quadratic approach. Sept.  S. CRC Press Inc. (W.).S. ed. In Proc. Lisbon.J. Robust adaptive MIMO control using multiple-model hypothesis testing and mixed-µ synthesis. . B. PhD thesis.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.