Graduate School of Information Technology

The starting point for the analysis and experiments presented in this paper is a simplified elevator control problem, calledS-ring'. As in many other real-world optimization problems, the exact fitness function evaluation is disturbed by noise. Evolution strategies (ES) can generally cope with noisy fitness function values. It has been proposed that theplus'-strategy can find better solutions by keeping over-valued function values, thus preventing inferior offspring with fitness inflated by noise from being accepted. Theplus'-strategy ...

Threshold selection-a selection mechanism for noisy evolutionary algorithms-is put into the broader context of hypothesis testing. Optimal selection thresholds were derived theoretically. These theoretical results were used to find threshold values for a simple model of stochastic search and for a simplified elevator simulator. Design of experiments methods are used to validate the significance of the results

Efficient elevator group control is important for the operation of large buildings. Recent developments in this field include the use of fuzzy logic and neural networks. This paper summarizes the development of an evolution strategy (ES) that is capable of optimizing the neuro-controller of an elevator group controller. It extends the results that were based on a simplified elevator group controller simulator. A threshold selection technique is presented as a method to cope with noisy fitness function values during the optimization run. ...

Today's urban life cannot be imagined without elevators. The central part of an elevator system, the elevator group controller, assigns elevator cars to service calls in real-time while optimizing the overall service quality, the traffic throughput, and/or the energy consumption. The elevator supervisory group control (ESGC) problem can be classified as a combinatorial optimization problem [Bar86, SC99, MN02]. It reveals the same complex behavior as many other stochastic traffic control problems, ie materials handling systems (MHS) with ...

Efficient elevator group control is a complex combinatorial optimization problem. Recent developments in this field include the use of reinforcement learning, fuzzy logic, neural networks and evolutionary algorithms [Mar95, CB98]. This paper summarizes the development of a parallel approach based on evolution strategies (ES) that is capable of optimizing the neurocontroller of an elevator group controller [SWW02]. It extends the architecture that was used for a simplified elevator group controller simulator [MAB+01, ...

Genetic network programming (GNP) has been proposed as a new method of evolutionary computation. Until now, GNP has been applied to various problems and its effectiveness was clarified. However, these problems were virtual models, so the applicability and availability of GNP to the real-world applications have not been studied. In this paper, as a first step of applying GNP to the real-world applications, elevator group supervisory control systems (EGSCSs) are considered. Generally, EGSCSs are complex ...

In order to increase the transportation capability of elevator group systems in high-rise buildings without adding elevator installation space, double-deck elevator system (DDES) is developed as one of the next generation elevator group systems. Artificial intelligence (AI) technologies have been employed to find some efficient solutions in the elevator group control systems during the late 20th century. Genetic Network Programming (GNP), a new evolutionary computation method, is reported to be employed as the ...

Genetic Network Programming (GNP), one of the extended evolutionary algorithms was proposed, whose gene is constructed by the directed graph. GNP is distinguished from other evolutionary techniques in terms of its compact structure and implicit memory function. GNP can perform a global searching, but it lacks of the exploitation ability. Since the behavior of GNP is characterized by the balance between exploitation and exploration in the search space, we proposed a hybrid algorithm in this paper that combines GNP with Ant ...

Abstract: The latest elevator systems have structural differences from traditional systems, such as the use of destination calls, or multiple cars in the same hoistway. This requires the development of new elevator group control systems, which is best done by adopting modern soft-computing methods. We review some of these systems and the results of research so far, and suggest further directions of research.

Abstract: Multi-Car Elevator (MCE) is the novel system and attracts attention for improvement of transportation in highrise buildings, but the design of controller for MCE is very difficult engineering problem. In this paper, the works of the authors are summarized: The control of MCE is divided to four phases and the assignment of a car to a hall-call is focused. The exemplar-based policy is used for the assignment, the policy is evaluated through discrete event simulation, and the parameter is optimized by genetic algorithm. To design traffic- ...

Multi-car elevator is an emerging technology consisting of two or more elevator cars moving independently in an elevator hoistway, which has become more appealing as building heights increase. In this paper, the design and drive methodologies for a linear motor driven multi-car elevator system with independently moving cars is introduced together with experimental results. Additionally, a safety method developed for the linear motor elevator and the conditions necessary for its proper operation are discussed. The new results ...

This paper summarizes the development of aparallel approach based on evolution strategies (ES) that is capable of optimizing the neurocontrollerof an elevator group controller [SWW02]. It extends the architecture that was usedfor a simplified elevator group controller simulator [MAB01, MN02, BEM03]

It has been clarified by researchers that insulators having contact angles, e.g. ± 45 ° with regard to the plane electrodes, is hard to acquire charge when they are exposed to high voltages in vacuum. For example, an insulator in the shape of a truncated frustum is free from charging. However, since such insulator has inevitably large volume, it obstructs the compactness. On the other hand, roughening insulator surface is also effective to prevent the surface from charging. In this case, the cost for roughening the entire surface may be high when the insulator is large. In this paper we review our previous investigation results and describe new experimental results aimed at developing feasible and compact insulators used in vacuum. Surface charging characteristics of partially and mechanically processed insulators have been studied for this purpose. That is, the charging phenomena when such insulators are exposed to a ramped dc or ac voltage have been observed by using a capacitive probe embedded in the cathode. Firstly, we describe of a partially roughened cylindrical insulator, where the height of the roughened surface measured from the cathode is varied keeping the total length constant. Secondly, we describe a cylindrical insulator having a truncated conical frustum or a chamfer at its cathode-side end. Based on these experimental and simulation results we prove that the partial mechanical processing is effective to control and suppress the insulator charging in vacuum.

The multi-car elevator system is a revolutionary new technology for high-rise buildings, promising outstanding economic benefits, but also requiring new technology for propulsion, safety and control. In this paper we report on experimental results with new components for linear motor driven multicar elevators. We show that linear synchronous motors with optimized design and with our new safety and control system can be considered as core components of a new generation of elevator systems. The main new results concern the development of a safety system integrated into the propulsion system, the design methodology of a linear motor optimized for the multicar elevator task, and the motion control system that is expected to be usable for extra high-rise buildings.

The double-deck elevator system (DDES) has been invented firstly as a solution to improve the transportation capacity of elevator group systems in the up-peak traffic pattern. The transportation capacity could be even doubled when DDES runs in a pure up-peak traffic pattern where two connected cages stop at every two floors in an elevator round trip. However, the specific features of DDES make the elevator system intractable when it runs in some other traffic patterns. Moreover, since almost all the traffic flows vary continuously during a day, an optimized controller of DDES is required to adapt to the varying traffic flow. In this paper, we have proposed a controller adaptive to traffic flows for DDES using Genetic Network Programming (GNP) based on our past studies in this field, where the effectiveness of DDES controller using GNP has been verified in three typical traffic patterns. A new traffic flow judgment part was introduced into the GNP framework of DDES controller in this paper, and the different parts of GNP were expected to be functionally localized by the evolutionary process to make the appropriate cage assignment in different traffic flow patterns. Simulation results show that the proposed method outperforms a conventional approach and two heuristic approaches in a varying traffic flow during the working time of a typical office building.  This paper is organized as follows: The overview of DDES is introduced in Section 2, followed by descriptions of the proposed method in the next section. Section 4 shows the simulation results and some discussions. Finally, conclusions and future work are mentioned in Section 5.

The multi-car elevator system is a revolutionary new technology for high-rise buildings, promising improved space utilization but requiring new control methods. We have developed a control algorithm for operating a multi-car system efficiently, by solving jointly the problems of allocating calls to cars, and planning elevator routes, using a continuously running real-time GA method. We report on simulation experiment showing the superiority of the proposed method, as compared with the best existing heuristic method.

In order to increase the transportation capability of elevator group systems in high-rise buildings without adding elevator installation space, double-deck elevator system (DDES) is developed as one of the next generation elevator group systems. Artificial intelligence (AI) technologies have been employed to find some efficient solutions in the elevator group control systems during the late 20th century. Genetic Network Programming (GNP), a new evolutionary computation method, is reported to be employed as the elevator group system controller in some studies of recent years. Moreover, reinforcement learning (RL) is also verified to be useful for more improvements of elevator group performances when it is combined with GNP. In this paper, we proposed a new approach of DDES using GNP with RL, and did some experiments on a simulated elevator group system of a typical office building to check its efficiency. Simulation results show that the DDES using GNP with RL performs better than the one without RL in regular and down-peak time, while both of them outperforms a conventional approach and a heuristic approach in all three traffic patterns.