A Modelling and Control Approach for a Type of Mixed Logical Dynamical System Using in Chilled Water System of Refrigeration System

Zhang, Yan; Chu, Xiaoli; Liu, Yang; Liu, Yongqiang

doi:https://doi.org/10.1155/2019/7131380

Mathematical Problems in Engineering

On this page

Abstract Introduction Conclusions Data Availability Conflicts of Interest Acknowledgments References Copyright Related Articles

Research Article | Open Access

Volume 2019 | Article ID 7131380 | https://doi.org/10.1155/2019/7131380

A Modelling and Control Approach for a Type of Mixed Logical Dynamical System Using in Chilled Water System of Refrigeration System

Yan Zhang,^1,2Xiaoli Chu,^3,4Yang Liu ,²and Yongqiang Liu²

Academic Editor: Akhil Garg

Received23 Sept 2018

Revised16 Jan 2019

Accepted25 Feb 2019

Published20 Mar 2019

Abstract

The chilled water system of central air conditioning is a typical hybrid system. The dynamic adjustment of cooling capacity based on hybrid system can achieve accurate temperature control and real-time energy saving. Mixed logical dynamical (MLD) systems have advantage for solving constrained optimization problems of this type of case by numerical methods. This paper proposes a novel modified type of MLD system, which enhances the model applicability and improves the switching flexibility and control effect for the framework. In order to meet the needs of cooling capacity and energy saving, the optimal control problem is transformed as MIQP problem by defined performance index. As a numeral example of application, the model and control method is used in pumps group control for variable water volume in chilled water system of central air conditioning. At last, the dynamic and energy-saving effects of the system are simulated, which shows the ideal control results.

1. Introduction

In recent years, hybrid systems have attracted many researchers from various fields, and more and more investigations focused on the theoretical exploration and engineering applications of hybrid systems in a variety of industries and fields, such as electric systems [1, 2], vehicle dynamic control [3, 4], aircraft design and control [5–7], traffic control [8, 9], robot systems [10, 11], chemical industries [12], and so on. Actually, modern complex industrial processes are typical hybrid systems, in which continuous dynamic processes interact with discrete events. According to the application background, the traditional single continuous or discrete systems hardly illustrate the process accurately, and the effect of corresponding control measure is not ideal. However, hybrid systems show advantages to reflect the issues of modelling and control for modern complex industrial applications. Because hybrid system includes both continuous and discrete components, and the control strategy is based on hybrid model. Hybrid systems adapt the characteristics of complexity in industrial control, therefore requiring further developments in the future.

For solving different fields of modelling and control problems, many researchers have built types of hybrid systems to fit different engineering cases, i.e., the mixed logical dynamical (MLD) systems [13]. MLD system is described by interdependent physical laws, logical rules, and operational constraints. In the process of modelling, qualitative knowledge and expert experience of the system are taken into account. The heuristic knowledge, logical judgment, and constraints that must be observed in operation of the object are transformed into propositional logic. The truth and falsity of propositions are represented by binary variables 1 and 0 introduced and expressed by logic. Connectors (and), (or), ~ (non), → (contain), ↔ (and only), (exclusive or) transform simple propositions into compound propositions and then express them in the form of integer linear inequalities containing both continuous variables and logical variables. Based on the MLD system, it is more convenient to establish and solve the problems concerned in the field of system, for example, reachability analysis [14], controller synthesis [15], state estimation [16], and system fault detection and so on [17].

Based on the above model rules and characteristics, MLD system can easily establish numerical methods for solving constrained optimization problems. The advantage of MLD model is that the systems can be represented by a hybrid logical dynamic model with the characteristics of linear dynamic systems in form after introducing appropriate auxiliary logical variables and auxiliary continuous variables. In order to achieve optimal control, the control measures [18–20] were transformed to dynamic programming problems, such as mixed-integer linear programming (MILP) form [21] and mixed-integer quadratic programming (MIQP) form [22, 23].

The study of MLD systems has gained growing interest in nearly two decades; some theoretical and applied results already exist in the literature. In [24], the model of a hydroelectric power plant in the framework of MLD systems was derived, and solved in a systematic way using a MILP method. In [25], a semi-active suspension model and constrained optimization problem was studied; the performance of different finite-horizon hybrid MPC controllers is tested in simulation using MIQP measure. In [26], an autonomous Lane-Change Controller via MLD systems was designed. An approach of complex mission optimization for multiple-UAVs using linear temporal logic was found in [27]. In [28], an algorithm named optimal pointer placement (OPP) scheduling algorithm was investigated and applied to the control and scheduling of a car suspension system. In [29], some recast types of biped motion were provided and controlled by on line optimal control approach. In [30], a unified modelling mode for control problems of multi-contact hand manipulation can be seen. In [31], a kind of MLD model to predict the back bead width of pulsed GTAW process with misalignment was addressed. In [32], a control strategy for a continuous stirred tank reactor (CSTR) system was introduced, and a controller of MPC was designed.

Many researchers focused on the existing MLD framework, and few people improve the model and control effect. Since the MLD system is an inequality form, the existing relationships between logical express and inequalities are based on zero, so there is only a couple of 0/1 switching signals. Based on that, there is only a pair of opposite switching directions of the system. The corresponding control strategy is also relatively simple, unable to deal with more complex control problems [33].

This paper proposes a novel type of MLD systems, which shows improvements for the inequalities. In framework of the inequalities, some new contain relations between logical expresses and inequalities are inferred by discussing non-zero switching point. A universal state-dependent switching law is obtained by dividing state region and setting three switching directions, which is not a 0/1 relationship confined to zero. For solving the optimal control problem of the improved MLD system, it is transformed as MIQP problem by defined performance index. The MLD system is firstly used in the refrigeration system for energy-saving control. Under the conditions of limited inputs and finite time, the inside temperature is reduced to setting value for minimizing power consumption. Via the state-dependent switching law and MIQP control strategy, pumps are used in variable water volume method for the chilled water system in central air conditioning, which transfer the cooling capacity to inside room by air system. Moreover, a simulation shows the control dynamical the energy-saving results.

The paper’s structure is as follows. Section 2 presents the framework of the type of MLD system with new switching points and law. The optimal control strategy is shown in Section 3. Section 4 uses the system in refrigeration system and gives a numeral inequality. Section 5 uses a simulation example to illustrate the dynamics of refrigeration system under the framework. Finally, some concluding remarks are made in Section 6.

2. Framework of the Improved MLD System

The improved MLD system contains n+1 subsystems, which is described as an inequality and composed of state variables, control variables, logical variables, auxiliary variables, novel defined switching points, and switching law. The constraints of the variables are gained via the inequality by (1), which is transformed from the constraints (2c) of the MLD standard form.

where state variable , input variable , logic variable , and auxiliary variable are defined as . is time scale, and with are state and input matrices of subsystems, respectively. is a small tolerance, and with are the upper and lower switching critical values, of course the switching points, respectively. and are state upper and lower bound values of vectors and , respectively. and are the switching laws/switching directions and the condition that it is triggered beyond the upper and lower boundaries, respectively. For convenience, equalities are set as , , , and .

The standard discrete-time MLD system is

where , , , is the continuous and discrete state of the system, respectively. , , , is the output vector. , , , is the continuous and discrete input variables, respectively. , , , , are matrices of suitable dimensions.

2.1. Mathematical Model

Consider the following dynamic system:

Each region is a continuous linear time-invariant subsystem, which indicates a mutually disjoint convex polyhedron, subjected to . , , and are the matrices of state, input and constraint, respectively. The system switches in such finite n+1 regions and satisfies the following conditions:

For each region , subjected to the exclusive-or condition:

Due to constraints of the subsystems and the transformational relation with the logic variables,

This infers that

Clearly, denotes the th row and th rank. According to the exclusive-or and equivalence relation [16], (7) can be transformed to equivalent mixed integer inequality constraints:

where , . Due to the logic variable , the standard form of the system can be rewritten as follows:

Equation (9) is nonlinear and contains a logic variable, state, and input for the sake of easy control. Thus, equation can be rewritten as mixed integer linear inequalities:

where and , which are finite-estimated or taken as the correct values by quadratic linear programming.

Assuming that

Finally, because of , the standard form can be updated as follows:

2.2. Novel Defined Switching Points

In [14], it just shows some simple inequalities (switched point is 0). For improving the freedom degree of switching point, and letting it not be confined to a specific point, a non-zero point should be presented. If the switched point is not 0 but S, can be set, is any value of the region, which infers some inequalities. The relationship between logical express and inequalities is shown in Table 1.

2.3. Switching Law

Based on the Novel defined switching points, switching law has new definition. Compared with one switching point (switched point is 0 mentioned in [16]), there are three intervals for switching decision, but not two. Thus, the switching law can be defined as follows: the state variable always changes in a region with time (whatever infinite or finite region). Here, two points can be set (the switching upper and lower bounds and ), which divide the region into 3 intervals, , , and . Once the state variable occurs or , the switching behavior will be triggered. If the state variable is , system still stays in current subsystem without triggering switching behavior.

If the system is in the th subsystem currently with , on account of the state value, switching points and intervals, the switching law can be set as follows [33]:

Remark(1)By the switching law/control strategy, is a mapping functions, the same as .(2) indicates that has different switching direction with .(3)If state variable , the system transfers from th subsystem to th subsystem; if state variable , the system transfers from th subsystem to th subsystem.(4) denotes the ith subsystem, the same as .(5) belongs to the integer set which contains n+1 numbers, just is one number of the set, the same as .(6)Generally, . Just under some special conditions, occurs. Otherwise, the law loses switching meaning.(7)Because the law is state-dependent, switching behavior is triggered by state variable, which is not decided by switching sequence and switching time.(8)Switching sequence and switching time are produced autonomously by state variable. Thus, both of them are random and synchronous.

The basic switching condition is known: only one subsystem can be selected each time, which abides by . or indicates the ith subsystem is selected or not, respectively.

Using the MLD logic expression, the law can be modified as follows:

with the constraint .

Because the system only selects one subsystem,, after switching behavior happens, the logical relations can be found:

From (14a), (14b), (14c), (15a), and (15b), we obtain the mixed integer inequality constraints:

This paper uses the continuous auxiliary variable and (12a), (12b), (12c), and (12d) to derive MLD inequality (1).

3. Optimal Control Strategy

Many literatures [13, 14, 18–23] show that mixed-integer quadratic programming (MIQP) method is a more effective way to deal with the problem of MLD system. This method can not only easily deal with the optimization control problem of quadratic performance objective, but, more importantly, it can conveniently handle the state of the system and the constraints of the control input and establish a unified method for the constraint optimization of the MLD system. Especially in continuous-time hybrid systems, it inevitably meets the problem of dimensionality in solving the Hamilton-Jacobi-Bellman and Euler-Lagrange differential equations by the principle of maximum value or the method of dynamic programming.

For a giving positive MLD system, the state transfers from initial state to final state in a finite time . We search (if it exists) the control sequence to ensure the state transferring and minimize the performance index:subject to and (2a), (2b), and (2c), where , , , are given weight matrices, and , , , , are given offset vectors satisfying (2b) and (2c).

From (2a), (2b), and (2c), the solution formula is with these following vectors:

Then, the object of constrained optimization control is transformed into a typical MIQP problem:where matrices , , are suitably defined.

4. Using in Refrigeration System

In the air conditioning system, the cooling capacity transfers from chiller unit to chilled water system by evaporator. The pumps drive water as cycle in the chilled water system, and send cooling capacity to air condition unit for exchanging from chilled water system to air system. After mixed with fresh air, the cold air is sent to air conditioning room to drop the temperature. In the process of temperature change in air conditioning room, the heat balance is influenced by several aspects: cold air, random heat of occupants and equipment, latent heat inside, and heat transfer from building structure. The processes of the cooling capacity transfer and temperature change are shown in Figure 1.

This paper uses constant air volume and variable water volume/constant temperature difference measures to transfer cooling capacity. In chilled water system, switching and variable frequency behaviors of pumps adjust the flow for saving energy, which denote the discrete and continuous time process, respectively. So we give an equation to show the heat balance as follows [34]:

On the left side of the equation, means the time differential of the heat capacity of a room. On the left side of the equation, reflects the cooling capacity for chilled water system; and denote the cooling capacity from return and fresh air systems, respectively; is random heat of occupants and equipment; shows latent heat inside; represents heat transfer from building structure. The description of the symbols and values can be seen in Table 2.

In the MLD system, the inequality is shown in (22), in which the values and calculations are from Table 2. is the inside temperature with variable boundaries, is the flow volume of the variable pump with limited range of 50-100%, denotes different amount of switchable pumps (fixed and variable frequency pumps are always work), and is auxiliary variable without physical meaning.

4.1. Model of Cooling Capacity Transferring

The principle of refrigeration is the progress of coldness transferring in essence; the cooling capacity is produced by chiller units, as a form of temperature difference between inlet and outlet water, which transfers from water system to air system. The cold air is sent into room to realize temperature shifting. Earlier central air conditioning usually uses a constant water volume and constant air volume to transfer coldness. Obviously, a constant water volume or air volume leads to a constant refrigerating or cooling capacity, even constant energy consumption. However, because of real-time variations in the cooling load, especially at low load, central air conditioning must reduce the refrigerating or cooling capacity to save energy.

In many modelling cases of central air conditioning, ether toward water system, or air system, few researchers combine both of them to model the system completely. Here, we describe the whole transfer process of coldness from water system to air system and model it overall. For air system, variable air volume technique usually uses several constant gears to adjust air speed; thus, it can be seen as constant air volume in some stage actually. For water system, variable water volume is an effective method too. Common techniques often use variable frequency pumps or switchable pumps (constant volume); few researchers combine variable frequency and switching measure together. Because central air conditioning with variable frequency pumps and switchable pumps has continuous and discrete features, a hybrid system can reflect the properties exactly. Here, we use this type of hybrid system and switching law to control it and save energy. We build the model between coldness transferring and inside temperature variation, by the mode of variable water volume and constant air volume. On account of same energy-saving principle, we just discuss chilled water system but cooling water system. And we focus on energy saving, for satisfying the requirement of temperature control, without regard to comfort, such as humidity.

The equation of heat balance in the test room is as follows:

In (23), it clearly shows that the cooling capacity comes from air being sent, and (24) indicates the transfer of the cooling capacity from the chilled water system to the air system. The model contains the chilled water system and the air system and uses a constant air volume and variable water volume as the transferring mode, to express the refrigeration progress as exactly as possible.

The returned air rate is set as ; hence, the new air rate is . The returned air volume can then be set as ; hence, the fresh air volume is . is the heat transfer efficiency from the water system to the air system (), is the sensible heat, and is the latent heat. Thus, the relationship for transferring coldness from the water system to the air system is as follows:

And because

we can substitute (24), (25), and (26) into (23) and obtain

The pumps are divided between fixed frequency pumps, variable frequency pump, and switchable pumps; thus, , where is the water volume of all fixed frequency pumps, is the water volume of one switchable pump, is the amount of working switchable pumps, and is the water volume of the variable frequency pump. Equation (27) can then be modified as (21).

4.2. Switching Law

In this example, there are three switchable pumps. Thus, the switching law can be defined asIn industrial system, state monitoring is periodical, assuming that the sampling period is . After each , the state value can be obtained. According to the switching law, system compares the value of with the switching critical value of each and decides which subsystem should be chosen next. Thus, switching behavior is decided autonomously.

If the system instantly switches to the th subsystem, the state variable can be divided into 3 intervals. After seconds, the system checks the value of . If , the system switches from the th subsystem to the i+1th subsystem; if , the system switches from the ith subsystem to the i-1th subsystem; if , the system remains in the ith subsystem. Note that if the system selects the subsystem and , the system still stay in the subsystem. Because the switchable pumps are out of work, the system is in the most energy-saving level. This is the same in the subsystem; i.e., if , the system will remain in subsystem, because the inside temperature must be dropped primarily. In a similar way, all the switchable pumps are full-load working for refrigerating. This phenomenon is the same for the next seconds.

The switching law can be expressed by the MLD logic expressions:

The MLD equation transforms the mixed integer inequality constraints:

Finally, the MLD inequalities and parameters are deduced in (22).

5. Numeral Example

A special items warehouse (length: 12 m, width: 6.5 m and height: 3 m) is the simulation object, which needs a constant temperature environment for storage. Simulink 2007 is used to simulate the modelling and controlling process, where the simulation time is 3000 seconds. The inside temperature starts off at 19.5, the setpoint, which has a small amplitude fluctuation near the set value in the simulation times due to the factors of outdoor temperature, indoor lighting, radiating equipment, and number of occupants. Assuming in a hot summer, the outdoor temperature rises from 35°C to 38.2°C in 3000 seconds (see Figure 2). Because of the two handling operations by the workers entering the warehouse, the indoor random heat capacity has two obvious growths (see Figure 3).

The switching law is state-dependent mode; the upper and low switching boundaries are defined as 0.2 and -0.2, respectively. Thus, the system checks the indoor temperature frequently and set the switch time as 60 seconds; once the boundaries are crossed, switching actions are triggered. The system begins at the 3th subsystem, in other works, rated power and flow volume. Under the optimal control strategy given in Section 3, we set Q₁=1, Q₂=0.01, Q₃=0.001, Q₄=10, Q₅=0 and use HYBRID TOOLBOX to simulate the dynamics. A common feedback control method is used to compare with the optimal control strategy, which shows the different effects in changes of flow volume, indoor temperature, and energy consumptions. The thermal parameters and control coefficients are given in Table 2.

Figure 4 shows the changes of indoor temperature near the set point under the two different control strategies above. It is obvious that the temperature has less amplitude of fluctuation by the optimal control than common feedback control method, which reflects more stable dynamical trajectory. Actually, the amplitude is approximate ±0.25°C during the whole process, which is unlike approximate ±0.35°C under common control method.

Figures 5 and 6 show the volume of chilled water system and the dynamic of switchable pumps under the two control strategies, respectively. These two figures look similar, because the total flow volume is composed of switchable, fixed, and variable frequency pumps; the change of whole volume is based on the dynamic of the switchable pumps. Under the optimal control strategy, it shows less switching behaviors and smaller flow volume than common control method. In engineering applications, the advantage presents less change of water pressure, which produces less impact and better hydraulic characteristics for pipe network.

Figure 7 shows the energy consumptions under the two control strategies in the 3000 seconds, which presents approximate 1.75×10⁵ and 1.84×10⁵ kWh, respectively. According to the cubic relationship between power and speed and the proportional relationship between speed and flow rate of pumps, the energy consumptions are calculated. Compared with the common control method, it provides approximate 5% energy saving. The power of various pumps is given in Table 2.

6. Conclusions

This paper reported a type of mixed logical dynamical system with novel structure, which shows an inequality contained state variables, control variables, logical variables, auxiliary variables, switching points, and switching law. The inequality is the constraints of system based on the state-dependent switching law, in where the switching points present more directions by the new relations between logical expresses and inequalities. A MIQP optimal control strategy is introduced for solving the problem. A case of refrigeration system is illustrated for the model and control dynamics, which use a thermal balance process to describe. From the simulation results, the changes of indoor temperature, flow volume, switching behaviors, and energy consumption show appeal effects for control and optimization of complex systems.

Data Availability

The data used to support the findings of this study are included within the article.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This work was financed by Foshan Social Science Fund (2018-ZDB05), Guangdong Youth Innovative Talent Project (2017WQNCX055), National Education Information Technology Research Fund (176140006), and National Nature Science Foundation of China (61104181).

References

K. Anoune, A. Laknizi, M. Bouya, A. Astito, and A. Ben Abdellah, “Sizing a PV-Wind based hybrid system using deterministic approach,” Energy Conversion and Management, vol. 169, pp. 137–148, 2018.
View at: Publisher Site | Google Scholar
H. Yang, L. Lu, and W. Zhou, “A novel optimization sizing model for hybrid solar-wind power generation system,” Solar Energy, vol. 81, no. 1, pp. 76–84, 2007.
View at: Publisher Site | Google Scholar
F. Borrelli, A. Bemporad, M. Fodor, and D. Hrovat, “An MPC/hybrid system approach to traction control,” IEEE Transactions on Control Systems Technology, vol. 14, no. 3, pp. 541–552, 2006.
View at: Publisher Site | Google Scholar
A. Khaligh and Z. Li, “Battery, ultracapacitor, fuel cell, and hybrid energy storage systems for electric, hybrid electric, fuel cell, and plug-in hybrid electric vehicles: state of the art,” IEEE Transactions on Vehicular Technology, vol. 59, no. 6, pp. 2806–2814, 2010.
View at: Publisher Site | Google Scholar
C. Friedrich and P. A. Robertson, “Hybrid-electric propulsion for aircraft,” Journal of Aircraft, vol. 52, no. 1, pp. 176–189, 2014.
View at: Publisher Site | Google Scholar
M. Soler, A. Olivares, and E. Staffetti, “Hybrid optimal control approach to commercial aircraft trajectory planning,” Journal of Guidance, Control, and Dynamics, vol. 33, no. 3, pp. 985–991, 2010.
View at: Publisher Site | Google Scholar
M. Heller, F. Schuck, L. Peter, and F. Holzapfel, “Hybrid control system for a future small aircraft,” in Proceedings of the AIAA Guidance, Navigation, and Control Conference, pp. 1–18, Portland, Oregon.
View at: Publisher Site | Google Scholar
L. Li, W. Dong, Y. Ji, Z. Zhang, and L. Tong, “Minimal-energy driving strategy for high-speed electric train with hybrid system model,” IEEE Transactions on Intelligent Transportation Systems, vol. 14, no. 4, pp. 1642–1653, 2013.
View at: Publisher Site | Google Scholar
L. M. Fernandez, P. Garcia, C. A. Garcia, and F. Jurado, “Hybrid electric system based on fuel cell and battery and integrating a single dc/dc converter for a tramway,” Energy Conversion and Management, vol. 52, no. 5, pp. 2183–2192, 2011.
View at: Publisher Site | Google Scholar
J. M. Toibero, F. Roberti, R. Carelli, and P. Fiorini, “Switching control approach for stable navigation of mobile robots in unknown environments,” Robotics and Computer-Integrated Manufacturing, vol. 27, no. 3, pp. 558–568, 2011.
View at: Publisher Site | Google Scholar
J. Tan, N. Xi, and Y. Wang, “A singularity-free motion control algorithm for robot manipulators—a hybrid system approach,” Automatica, vol. 40, no. 7, pp. 1239–1245, 2004.
View at: Publisher Site | Google Scholar | MathSciNet
J. Wang, J. Ye, E. Feng, H. Yin, and Z. Xiu, “Modeling and identification of a nonlinear hybrid dynamical system in batch fermentation of glycerol,” Mathematical and Computer Modelling, vol. 54, no. 1-2, pp. 618–624, 2011.
View at: Publisher Site | Google Scholar
A. Bemporad and M. Morari, “Control of systems integrating logic, dynamics, and constraints,” Automatica, vol. 35, no. 3, pp. 407–427, 1999.
View at: Publisher Site | Google Scholar | MathSciNet
A. Bemporad, G. Ferrari-Trecate, and M. Morari, “Observability and controllability of piecewise affine and hybrid systems,” IEEE Transactions on Automatic Control, vol. 45, no. 10, pp. 1864–1876, 2000.
View at: Publisher Site | Google Scholar | MathSciNet
A. Bemporad, A. Giua, and C. Seatzu, “Synthesis of state-feedback optimal controllers for continuous-time switched linear systems,” in Proceedings of the 41st IEEE Conference on Decision and Control, vol. 3, pp. 3182–3187, USA, December 2002.
View at: Google Scholar
A. Bemporad, F. D. Torrisi, and M. Morari, “Optimization-based verification and stability characterization of piecewise affine and hybrid systems,” Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics): Preface, vol. 1790, pp. 45–58, 2000.
View at: Google Scholar
A. Bemporad, D. Mignone, and M. Morari, “Moving horizon estimation for hybrid systems and fault detection,” in Proceedings of the American Control Conference (99 ACC), pp. 2471–2475, June 1999.
View at: Google Scholar
D. Frick, A. Domahidi, and M. Morari, “Embedded optimization for mixed logical dynamical systems,” Computers & Chemical Engineering, vol. 72, pp. 21–33, 2015.
View at: Publisher Site | Google Scholar
J. Lian, S. Liu, L. Li et al., “A mixed logical dynamical-model predictive control (MLD-MPC) energy management control strategy for plug-in hybrid electric vehicles (PHEVs),” Energies, vol. 10, no. 1, 2017.
View at: Google Scholar
T. Zou, X. Wang, and S. Y. Li, “A mixed logic enhanced multi-model switching predictive controller for nonlinear dynamic process,” Control & Intelligent Systems, vol. 35, no. 2, p. 154, 2007.
View at: Google Scholar
T. Van Den Boom and B. De Schutter, “Properties of MPC for max-plus-linear systems,” European Journal of Control, vol. 8, no. 5, pp. 453–462, 2002.
View at: Publisher Site | Google Scholar
S. Karaman, R. G. Sanfelice, and E. Frazzoli, “Optimal control of mixed logical dynamical systems with linear temporal logic specifications,” in Proceedings of the 47th IEEE Conference on Decision and Control, CDC '08, pp. 2117–2122, Mexico, North America, December 2008.
View at: Google Scholar
P. Tøndel, T. A. Johansen, and A. Bemporad, “An algorithm for multi-parametric quadratic programming and explicit MPC solutions,” Automatica, vol. 39, no. 3, pp. 489–497, 2003.
View at: Publisher Site | Google Scholar
G. Ferrari-Trecate, E. Gallestey, P. Letizia, M. Spedicato, M. Morari, and M. Antoine, “Modeling and control of co-generation power plants: a hybrid system approach,” Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics): Preface, vol. 2289, pp. 209–224, 2002.
View at: Google Scholar
N. Giorgetti, A. Bemporad, H. E. Tseng, and D. Hrovat, “Hybrid model predictive control application towards optimal semi-active suspension,” International Journal of Control, vol. 79, no. 5, pp. 521–533, 2006.
View at: Publisher Site | Google Scholar
Y. Du, Y. Wang, and C.-Y. Chan, “Autonomous lane-change controller via mixed logical dynamical,” in Proceedings of the 17th IEEE International Conference on Intelligent Transportation Systems, ITSC '14, pp. 1154–1159, October 2014.
View at: Publisher Site | Google Scholar
S. Karaman and E. Frazzoli, “Complex mission optimization for multiple-UAVs using linear temporal logic,” in Proceedings of the American Control Conference, ACC '08, pp. 2003–2009, USA, June 2008.
View at: Google Scholar
M. E. M. Ben Gaid, A. Çela, and Y. Hamam, “Optimal integrated control and scheduling of networked control systems with communication constraints: application to a car suspension system,” IEEE Transactions on Control Systems Technology, vol. 14, no. 4, pp. 776–787, 2006.
View at: Publisher Site | Google Scholar
G. Yin and S. Hosoe, “Mixed logic dynamical modeling and on line optimal control of biped robot,” in Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS '06, pp. 5895–5900, China, October 2006.
View at: Google Scholar
Y. Yin, S. Hosoe, and Z. Luo, “A mixed logic dynamical modeling formulation and optimal control of intelligent robots,” Optimization and Engineering. International Multidisciplinary Journal to Promote Optimization Theory & Applications in Engineering Sciences, vol. 8, no. 3, pp. 321–340, 2007.
View at: Publisher Site | Google Scholar | MathSciNet
H. Ma, S. Wei, T. Lin, and S. Chen, “Mixed logical dynamical model for back bead width prediction of pulsed GTAW process with misalignment,” Journal of Materials Processing Technology, vol. 210, no. 14, pp. 2036–2044, 2010.
View at: Publisher Site | Google Scholar
J. Du, C. Song, and P. Li, “Modeling and Control of a Continuous Stirred Tank Reactor Based on a Mixed Logical Dynamical Model,” Chinese Journal of Chemical Engineering, vol. 15, no. 4, pp. 533–538, 2007.
View at: Publisher Site | Google Scholar
Y. Zhang, X. Chu, and Y. Liu, “A class of mixed logical system and using in energy-saving control of central air conditioning,” in Proceedings of the IEEE 3rd Advanced Information Technology, Electronic and Automation Control Conference (IAEAC '18), pp. 1020–1026, Chongqing, China, October 2018.
View at: Publisher Site | Google Scholar
Y. Zhang, Y. Liu, and Y. Liu, “A hybrid dynamical modelling and control approach for energy-saving of central air conditioning,” Mathematical Problems in Engineering, Article ID 6389438, 12 pages, 2018.
View at: Google Scholar

Copyright

Copyright © 2019 Yan Zhang et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

PDF Download Citation

Download other formats

Order printed copies

Views

1361

Downloads

987

Citations