HK1126745B - Elevator system - Google Patents

Description

Elevator system

Technical Field

The invention relates to the control of an elevator group. In particular, the subject of the invention is a method and an apparatus for controlling an elevator group by allocating landing calls (landing calls) specifically taking into account the potential energy and the kinetic energy of the elevators. The term "elevator" here refers to the sum of moving masses (moving masses) moving in an elevator shaft (elevator hoistway), irrespective of the arrangement of one, two or more elevator cars (elevator cars) in the elevator hoistway. In double-deck cars or multi-deck cars, the cars are fixed to each other one above the other so that they serve multiple floors in synchronism.

Technical Field

The allocation of landing calls is one of the most basic of the many different tasks of elevator group control. The targets of the allocation are: the elevator car serves the call in such a way that the performance indicator (performance indicator) describing the elevator system is as good as possible. Traditionally, the most common performance indicators relate to call times (times) and passenger waiting times. Typically, the averages of these times are calculated, and their distributions (distributions) are established.

There are many types of methods for allocation of landing calls, and each elevator manufacturer has its own method of accomplishing this task. However, a common feature of all these different methods is that they include a set of parameters specific to each method that can affect the operation of the method used.

The targets of monitoring are typically landing calls, car calls (car calls), elevator load and elevator motion status. During peak hours, the target may give priority to minimizing the travel time (travel time) of the elevator users. Another commonly used optimization objective, especially in the interests of the building owner, is the energy consumption of the elevator system.

Many optimization objectives can be found, such as call time, expected waiting time of passengers, running time (runtime), travel time, number of stops, car load, number of car calls and landing calls synchronized, etc. What has to be determined is which of these targets should be given priority and under which traffic conditions (traffic) how much priority should be given.

Since non-renewable energy resources on earth are limited and increased energy consumption leads to many indirect effects, such as in the form of the greenhouse effect, energy consumption is an important minimization target. The operating costs and maintenance costs of a building can be influenced in respect of their operating costs and maintenance costs by using an elevator system that is economical in respect of its energy consumption. According to a study carried out in hong Kong (Yim, Leung' Building for the 2001 summons in London 21^stPaper at century' conference: "Building services summary in public Housing"), the elevator system consumes about 18% of the energy consumed in the public space of a typical 40-story residential Building (i.e., excluding the consumption of personal electrical energy used by the residents themselves). Another study estimates that: the elevator system uses 5-15% of the total energy consumption of the building. For the energy consumption of an elevator, the transport capacity (transport capacity) of the elevator, the motor, the power input ratio and, in general, the design of the mechanical parts of the system must be taken into account.

International patent application WO 02/066356 presents a control method of an elevator system in which the energy consumed by the elevator system is minimized so as to meet the desired elevator passenger service time requirement on average. In the method, given a target value for a certain service time of the elevator group, landing calls are allocated to different elevators so that the condition of the service time to be detected is fulfilled over a longer time interval, while, however, the energy consumption of the system is at a minimum. For example, the call time from the placing of a call to the arrival of an elevator, the total travel time, or the running time, which only detects the time consumed in an elevator, can be used as service time.

In one application according to patent application WO 02/066356, two magnitudes of different types, which cannot be measured in the same unit, are optimized, namely the waiting time and the energy consumption. In order to enable these quantities to be measured in units of one another and to be compared with one another, the route (route) R of the elevator is selected such that the cost items

C＝W_TT_N(R)+W_EE_N(R)。 (1)

And (4) minimizing. T is_N(R) is the sum of the normalization processes of the call times of the alternative route R, and likewise E_N(R) is the energy consumption of the normalization process due to the alternative route R. W_TAnd W_EIs a weighting factor of the aforementioned cost term, such that:

0≤W_T≤1，W_E＝1-W_T。 (2)

publication US6857506 describes another call allocation method of an elevator system. In the method, an energy consumption file is formed, which describes the energy consumption of each possible elevator trip (trip) between two floors. Thus, the elements of the energy consumption file use the departure floor, the arrival floor, and the car load as variables. When the energy consumed for each trip between two floors is known, the routing of the elevator car can be calculated for the active call, minimizing the total energy consumption of the system.

Publication FI 115130 also relates to the control of an elevator group. In this method, a desired target value may also be set for a certain service time, such as the average waiting time of passengers. In this case, the purpose is: the energy consumption is minimized so that the available model of the elevator system is utilized in the optimization. By means of this model, the expected service time can be predicted. The system also includes a PID regulator that utilizes the predicted service time, and thus, the cost function can be more efficiently optimized. A number of alternative routes according to smaller energy consumption can be obtained from the optimiser from which a suitable solution is selected according to the target value of the desired service time.

The common features of the above described prior art solutions are: the elevator cars are routed so that the amount of change in system potential energy due to transporting passengers in the elevation direction is minimized. In this connection, the system comprises all the mass points moving in the vertical direction, in other words the elevator car with counterweight and the elevator passengers.

One problem with the prior art is that: in the minimization of the energy consumption according to the prior art, only the mass to be moved in the system and the length of the journey (journey), i.e. the height difference between the departure floor and the arrival floor, are taken into account. When considering the optimization of energy consumption, the optimization can be performed more accurately by including an energy term related to the elevator speed.

Object of the Invention

The object of the invention is to disclose an allocation method for an elevator car, in which both the potential energy and the kinetic energy prevailing in the elevator system are taken into account, so that the energy consumed by the system is minimized.

Disclosure of Invention

Some inventive embodiments are presented in the drawings of the descriptive section of the present application. The inventive content of the application is defined differently than in the claims presented below. The inventive content may also consist of several separate inventions, especially if the inventions are considered in the light of expressions or implicit sub-tasks or from the point of view of advantages or categories of advantages achieved. In this case, some of the attributes contained in the claims below may be superfluous from the point of view of separate inventive concepts. The features of the various embodiments can be applied in combination with other embodiments within the scope of the basic inventive concept.

The invention discloses a method for controlling the elevators belonging to the same elevator group on the basis of calls issued, wherein the elevators comprise one or more elevator cars disposed in the same elevator shaft, and in which method a genetic algorithm is used. In the genetic algorithm at least one allocation option is formed, i.e. a chromosome, wherein the chromosome comprises call data and elevator data for each active landing call or destination call (destination call), and the data, i.e. genes, together determine the elevator car serving each landing call or destination call. Furthermore, in the algorithm, the value of the cost function for each chromosome is determined. Thereafter, at least one chromosome is formulated for at least one gene. Thereafter, the value of the cost function for each formulated chromosome is determined. The formulation of chromosomes is repeated until the end criteria are met. Based on the value of the cost function, the best chromosome is selected and the elevator car is assigned to the call issued according to the gene of the selected best chromosome. The invention is characterized in that: in the method information about the type of run, i.e. a run type gene, wherein the run type determines the speed profile, is associated with a chromosome, which is associated with each call data and elevator data, according to which the elevator having the elevator car travels between the departure floor and the call-giving floor defined by the call gene associated with the run type gene.

In one embodiment of the invention the kinetic energy of the elevator per elevator trip is determined by means of a speed profile determined by the car load and the type of operation of the elevator. The total energy consumed by the elevator system is selected as a cost function or a part of a cost function, whereby the term applied to the kinetic energy of the elevator is included in the cost function. The global minimum of the total energy consumption of the elevator system found is selected as the end criterion.

According to another embodiment of the invention, the chromosomes are formulated as a next generation of genetic algorithms by selection, crossover and/or mutation.

According to another embodiment of the invention, the end criterion is fulfilled when a predetermined value of the cost function, the number of generations, the running time of the algorithm or a sufficient homogeneity of the population is achieved. Here, uniformity means a state in which the same chromosome is transferred from one generation to the next when a successive multi-generation genetic algorithm is formed.

According to another embodiment of the present invention, the cost function of the chromosome is defined to include an energy consumption term and a service time term, and both are weighted by a preset weighting coefficient. Furthermore, the elevator model and the current state of the elevator system can be used as an aid for calculating the cost function.

According to another embodiment of the invention at least one of the group of items consisting of energy consumption of the elevator system, passenger waiting time, passenger travel time, passenger running time is selected as an item of the cost function. In addition, in calculating the energy consumption, the kinetic energy of the elevator, the potential energy stored in the elevator, the energy consumed due to friction and other losses, and the energy regenerated to the electricity input system are taken into account.

According to another embodiment of the invention at least one constraint is defined for the speed profile of each elevator according to the run type from a set of magnitudes, which comprises the maximum speed, the maximum acceleration and the maximum jerk (jerk) of the elevator, wherein the maximum jerk is defined as the change in acceleration per unit time.

According to another embodiment of the invention, a direction gene of the chromosome is defined for each stationary elevator.

According to another embodiment of the present invention, the operation types are defined as "normal", "slight deceleration", "significant deceleration", "slight acceleration", and "significant acceleration". In this case the run type is defined by setting the maximum speed used by the elevator, so that the run type "normal" represents the nominal travel speed of the elevator, and in other run types the travel speed of the elevator deviates from the nominal value by an amount which is a percentage from a preset value.

In addition to the above-described control method, the inventive concept of the invention also comprises a control system of a similar elevator system, wherein the aforementioned phases of the method are performed by a GA optimizer.

The inventive concept of the present invention also comprises a computer program defined to perform the different phases of the method described above when run.

As can be summarized from the foregoing, the potential energy related to the system mass point, as well as the kinetic energy related to the moving and rotating parts, can be taken into account when calculating the optimal (i.e. given the minimum energy consumption) routing of the elevator car. In addition, wear caused by friction can also be detected. Thus, the most fundamental advantages of the invention are: kinetic energy is included in the optimization evaluation, whereas prior art techniques only consider potential energy and frictional losses. When the running speed profile is included in the optimization, the GA optimizer can use more alternative routes where the energy consumption is kept low. Since the kinetic energy depends on the square of the speed, relatively small changes in car speed can have a considerable effect on the kinetic energy that is predominant in the system and thus on the energy economy of the entire system. For example, when the maximum speed of the car changes by +/-20%, the kinetic energy of the car is obtained and is changed within the range of-36-44%.

In this case, when the genetic algorithm has more available alternative routes, of which the alternative routes satisfying the desired criteria are selected, it is clear that the optimization in the case according to the invention works better on average, and in this case the elevator system consumes less energy than in the prior art technique without reducing the level of service provided to its users.

Drawings

Fig. 1 shows a method according to the invention for scheduling the trips of an elevator car, in which a so-called genetic algorithm is utilized; and

fig. 2 shows the structure of a chromosome of a genetic algorithm according to the present invention.

Detailed Description

The present invention discloses a method for allocating elevator cars based on active calls, which utilizes prior art genetic algorithms and a novel chromosome structure according to the present invention. The basic concept of the invention is as follows: in addition to potential energy, kinetic energy related to system mass and system velocity is contained in the test.

It is conceivable that a single elevator in use comprises three mass members, i.e. an empty elevator car, an elevator counterweight and passengers in the elevator car. If the passenger capacity of the elevator car is CC (in kg), then in this case the counterweight can be calibrated so that:

if the losses are caused by friction in the system, the mechanical energy consumed and the part used by the return system (energy) can be determined from the potential energy, kinetic energy and energy losses in the run of the elevator from one floor to another:

E_F＝F_μΔh

in equation (4), Δ h is the distance from the departure floor to the arrival floor, m_pIs the internal load (m) of the car_sEffective inertial mass (inertia) for all masses moving linearly and rotationallymass)， The maximum speed that can be achieved during operation. F_μIs the total effective friction force exerted on all moving parts and the traction devices of the car. Modern elevator machinery can be operated with a certain efficiency coefficient eta_RReturning potential energy and kinetic energy of the system to the power source, wherein:

0≤η_R≤1 (5)

in older mechanisms, energy is directed to the load resistance, in which case the efficiency coefficient η_RIs zero.

The movement of the elevator car is controlled in a control system which, in modern elevators, operates by means of a closed feedback loop. The aim is to control the elevator car so that the movement of the elevator car is comfortably smooth, in other words the car does not jerk, i.e. in this case da/dt (change in acceleration per unit time) is kept practically at zero or very small. The energy required in the travel of the elevator comes from a power source and is consumed in the form of electrical and mechanical losses. When the elevator car moves, energy is stored in the form of elevator kinetic energy, which constitutes the combined kinetic energy of the elevator car and other moving masses (such as the counterweight) comprised in one elevator. Thus, the detection of energy and speed can also be extended to other solutions, where at least one elevator shaft of the building contains a double-or multi-deck elevator. When the car stops after an elevator trip, depending on the direction of movement of the car and the car load, the potential energy prevailing in the system changes at the beginning of the elevator trip and part of the kinetic energy prevailing during this trip returns to the system in the form of potential energy. One example that can be considered is a fully loaded, upwardly moving elevator car. During the deceleration phase of the car, a part of the kinetic energy of the car is dissipated in the form of losses caused by friction, a part is changed into potential energy, and the remaining energy can be returned to the power supply via the power transmission system. Since the control system, its feedback connections and part of the power transmission system complicate the situation, it is difficult to derive a general formula for the energy consumed during an elevator trip.

In the following, the relative difference in weight of the energy terms in the actual elevator system is evaluated. Here, a simplification is made according to which three different kinds of energy (kinetic energy, potential energy and losses) are assumed to be independent of each other. In this case, each of these energy terms may be detected independently.

With respect to the power input, the flow of energy may be detected from the perspective of potential energy as follows:

and (6)

E_PSP＝E_P·η_R|E_P＜0

Wherein eta is_MFor the efficiency coefficient, η, of the power transmission apparatus during operation_RIs the efficiency factor of energy regeneration (energy used by the return to the power source).

Correspondingly, the net kinetic energy is the difference between the input energy required for the highest speed of the car and the regenerated energy:

the loss due to friction can be obtained by

The most interesting energy forms that are the dominant part in elevator systems from the point of view of landing call allocation and from the point of view of elevator car routing are potential energy and kinetic energy. By defining different alternative routes for the elevator car, the direction of movement and the load of the car can be changed. At the same time, the amount of energy that changes from one form to another also varies with the schedule. The control system of the elevator is able to influence the parameters relevant to the run, the most essential of which are the speed, acceleration and so-called jerk (which is defined above as the change in acceleration per unit time) of the elevator car. Since the speed of movement of the car is influenced by the control, and thus the maximum speed is also influenced (e.g. by adjusting the time period, the force is exerted on the car), the kinetic energy described in equation (7) is also directly influenced in this way.

The magnitude of the range of variation in kinetic energy and potential energy when the detected objects are an empty elevator car and a most loaded elevator car (full of people) is detected by way of an example below. If it is assumed that the mass of an empty elevator car with respect to its maximum load (maximum number of passengers) CC is:

m_car＝2.5·CC (9)

then the following can be obtained for the mass of the counterweight according to equation (3):

m_cw＝3·CC (10)

when the combined mass of other moving mass points (such as trailing cables) is assumed to be:

m_r＝0.5·CC (11)

the resulting total mass to be moved is then:

m_S＝m_P+6·CC (12)

if it is further assumed that the control system is able to change the maximum speed of the elevator car by ± 20% of the nominal value, then using the preceding formula the range of kinetic energy variation between the case of a full elevator car and an empty elevator car can be calculated:

in a corresponding manner, an estimate of the magnitude of the range of potential energy variation is obtained:

from equations (13) and (14), the relationship between the ranges of variation can be calculated:

it must be noted that the control system does not need to support regeneration if it is desired to optimize energy consumption by detecting potential energy only. In a non-renewable (energy) system, the chance of reducing energy consumption is halved compared to a renewable (energy) system, and in the above example, the range of potential energy variation of the non-renewable (energy) system is:

one prior-art method of allocating elevators on the basis of calls is to use Genetic Algorithms (GA), especially in large elevator systems. Genetic algorithms are described in e.g. patent publication FI 112856. The rules of operation of the genetic algorithm are also shown in the example of fig. 1, which is described as follows:

genetic algorithms do not absolutely guarantee that an optimal routing is found, but in practical applications the results obtained are very close to this optimal routing. In the genetic algorithm the travel routes of the elevators 10 in the system are coded 15 into different chromosomes 11, wherein the position of one gene 17 defines the active call, to which the value of the gene 17 of the elevator (elevator a or elevator B) is assigned. In the example of fig. 1, there are four active calls, assumed to be floor 2, floor 4, floor 5, and floor 6. A special direction gene can also be defined for a stationary elevator (e.g. on floor 3), the value of which can be "up" or "down" describing the starting direction of the elevator in question. The system starts the movement from, for example, a randomly selected alternative route to which different genetic processes such as crossing, mutation and selection 18 are applied. Crossing means that two alternative routes are randomly integrated into one new alternative route. In mutation, the value of the chromosomal gene is randomly changed. Through these genetic routines, a new set of chromosomes is formed one generation at a time, while the viability (viability) of the obtained chromosomes is tested for further processing purposes. For example, viability may mean reducing (underscut) the value of a certain waiting time or reducing the desired energy consumption value. In the definition of viability, models 12 and 13 applied to the elevator can be used in order to obtain good advantages. In the example of fig. 1, viability is measured by calculating the value of a so-called cost function 14. In this example, the function to be computed is the sum of the call times (times). Each alternative route 11 is directed to the calculation of a cost function 14, whereby the value C of the resulting cost function is also encoded as a cost gene 19 in the chromosome. The chromosome results given by the algorithm are sometimes convergent, in other words, for example, the chromosome giving the smallest C is finally selected for further processing. Finally, from the last set of chromosomes for processing, the best or most suitable in terms of viability is selected. The routing of the elevators is controlled on the basis of the genes of the selected chromosome and the passengers are allocated to the elevators. The genetic algorithm runs continuously according to this rule because when the elevator system learns a new active call, the chromosome 11 has to be defined again and the above-mentioned operation 18 has to be applied in the new chromosome 11.

The present invention utilizes genetic algorithms, but the new concept regarding the use of run types is attached to the aforementioned methods. This means that: for example, a number of different speed classes can be defined for the elevators, in which the elevators move. As a preferred embodiment, three different speed levels may be defined: "fast", "normal" or "slow". As another preferred embodiment, five different speed levels may be defined: "very slow", "slower", "normal", "faster" and "very fast". By defining the speed classes such that: the travel speed is set for the part of the elevator trip to be traveled according to the standard speed of the elevator, in other words the maximum travel speed of the elevator is limited to this set value of speed for the entire elevator trip. The rating may be defined as a deviation from its nominal value (from normal), e.g. + -. 10% or. + -. 20%. Furthermore, the acceleration (a) of the elevator car and the jerk (Δ a/Δ t) of the elevator car describing the change in acceleration per unit time can be included in the detection, and the same type of classes can also be defined for these quantities relating to kinetic energy and passenger comfort. Thus, in a curve of a grade, such as the highest speed of the elevator car, the maximum acceleration and the maximum allowed jerk can be defined, or on the other hand, only one or two of the aforementioned magnitudes can be included in the curve.

The invention is characterized in that: in connection with each landing call, a new type of additional gene is formed in chromosome 11 of the GA system according to fig. 1. Such chromosomes are depicted by way of example in fig. 2. Herein, such a novel gene is referred to as an operation speed gene or an operation type gene of an elevator. In the example of fig. 2, there are three different alternatives for the operating speed of the car: normal operating speed, decelerated operating speed, and accelerated operating speed. The operator of the elevator system can set these speeds in the control system. For example, in the foregoing manner, the slow and fast operating modes may be lower or higher than the normal operating speed by a desired percentage amount.

From the set of chromosomes according to fig. 1, the top chromosome 11 was selected for fig. 2. Four active calls are seen on chromosome 11 for which elevator a 20 is allocated to the up call issued on floor 2, elevator a 22 to the down call issued on floor 4, elevator B24 to the up call issued on floor 5 and elevator B26 to the down call issued on floor 6. In the embodiment of fig. 1, the chromosome 11 includes a direction gene 16 related to the stationary elevator on the floor 3. In the case of FIG. 2, the cost gene 19 comprises two parts: waiting time WT of passenger and energy consumption E of the route_R. However, the cost gene is not limited to these quantities, but the terms of the calculated cost function may be other quantities to be optimized. In the cost function, the different magnitudes can also be weighted with desired weighting coefficients.

Therefore, the invention is characterized in that: a run type gene is associated with each call gene 20, 22, 24, 26 that determines how to drive to the floor according to each landing call. It can be seen that the level of the highest speed of these trips between floors is associated with each call gene. The following assumptions: the chromosome according to fig. 2 is chosen completely so that it is used by the control of the elevator system as the final result given by the genetic algorithm for the optimized alternative route. For example, elevator a responds to the first call 20 at a slow maximum speed (according to the run type gene 21), which ensures smooth and jerk-free travel of the car. When the elevator is aggregating to a landing call 20 at floor 2, elevator a continues to move towards floor 4 to aggregate a call 22. For this second run between floors, the elevator takes its nominal value as its maximum speed, since the run type gene 23 gives "normal" for this run speed. In a corresponding manner, elevator B thereafter travels at a high maximum speed away from floor 3 towards floor 5 (run type gene 25 "fast"). Thereafter, elevator B travels between floor 5 and floor 6 at the highest fast speed, as indicated by the fourth gene pair 26, 27 of the chromosome.

By means of the run type classification, the speed profile of the elevator and thus of each elevator car is obtained for use by the genetic algorithm. Thus, more alternative routes for assigning cars to each active call are obtained. Since the maximum speed can vary from floor to floor level, not only the travel time used for different alternative routes changes, but also the amount of kinetic energy stored in the elevator in connection with the route. In accordance with the above description, the form of energy is converted between energy given by the power supply, kinetic energy and potential energy, some of which is dissipated in losses and some of which is returned to the system for use. In this case, it must be logically assumed that: for example, in an alternative route giving the least energy consumption, the maximum speed used by the elevator car is not kept constant when comparing the elevator trips between two floors with each other. It is therefore clear that there are more alternative routes available to the genetic algorithm, from which the route giving the lowest energy consumption is selected accurately. For example, during exit traffic (exit traffic) times, when passengers attempt to descend from floors to an exit floor at street level, it is more advantageous from an energy point of view to gather as many passengers as possible to the elevator car. By means of the speed classes the loading of the elevator cars can be improved when empty elevators return from the exit floors to the floors more slowly than normal speed to carry people leaving the building. In this case the elevator arrival will last longer, and more passengers will be waiting for the elevator than usual. The result is: the car is already loaded quickly in the upper part of the building, which promotes regeneration of the potential energy associated with the passengers. This is the case, since there are more alternative routes, it can be assumed that: the extension according to the invention gives a better optimized routing regardless of whether it is time or energy used that is relevant to the optimization objective.

Although the foregoing example only addresses the allocation of hall calls, the present invention is not limited to this conventional call system. The invention can also be applied in so-called destination floor call systems (destination floor call systems), in which a user has given his/her destination floor call already in the corridor of the departure floor. The destination floor call system may also be referred to as a destination call system. When using such a system, there is no longer a need to give a separate call (call) in the elevator car. In this case, an elevator car is assigned to each passenger, not individually to an up call or a down call as in the conventional system.

The basic idea of the invention can further be applied in double-deck and multi-deck elevator systems. In these elevator systems, two or more elevator cars are located one above the other in the same elevator shaft, respectively, so that the elevator cars arranged in one shaft together constitute a movable stationary unit, i.e. an elevator. In this case the distance between the two elevator cars is dimensioned to be the same as the distance between the two floors. In this allocation method, the chromosome contains each individual elevator car and, if necessary, the direction gene of the particular elevator. When evaluating each chromosome (i.e. the alternative scheduling) in the optimizer, in the model of the elevator system, the "detached" elevator cars in the chromosome are first added to those elevators to which they belong. Thereafter, the cost function is calculated from the route indicated by the chromosome using the elevator located in the elevator shaft with all its cars and hoisting gear. These costs may be of the same magnitude as those mentioned above, in other words: call time, waiting time or change in potential or kinetic energy of the elevator (i.e. the entire hoisting mechanism, not just one car). Finally, the allocation method indicates for each given call the elevator car that is most suitable among all the elevator cars.

The invention is not limited to the embodiments described above, but many variations are possible within the scope of the inventive concept defined by the claims below.

Claims (20)

1. Method for controlling multiple elevators belonging to an elevator group on the basis of calls issued, wherein the elevators comprise one or more elevator cars (10) disposed in the same elevator shaft, in which method a genetic algorithm is used, according to which algorithm:

forming at least one allocation option, i.e. a chromosome (11), the chromosome (11) containing call data and elevator data for each active landing call or destination call, which data, i.e. genes (17), together determine the elevator car (10) serving the landing call or destination call;

for each chromosome (11), determining a value (14, 19) of a cost function;

forming at least one chromosome (11) with respect to at least one gene (17);

determining, for each chromosome of the formed at least one chromosome, a value of a cost function (14, 19);

repeatedly forming chromosomes (11) until an exit criterion is met;

selecting the best chromosome on the basis of the value (19) of the cost function; and

allocating an elevator car (10) in response to the issued call according to the selected gene of the best chromosome,

characterized in that the method further comprises the following stages:

information about the type of operation, i.e. an operation type gene (21, 23, 25, 27), is associated with the chromosome (11), the chromosome (11) being associated with each call data and each elevator data, in which operation type gene the operation type determines a speed profile, according to which the elevator having the elevator car (10) travels between the departure floor and the call-giving floor defined by a call gene (20, 22, 24, 26), which call gene (20, 22, 24, 26) is associated with the operation type gene (21, 23, 25, 27).

2. The method according to claim 1, characterized in that the method further comprises the following stages:

determining the kinetic energy of the elevator for each elevator trip from a speed profile determined from the car load and the run type of the elevator;

selecting the total energy consumed by the elevator system as a cost function or as a part (14) of the cost function, so that the cost function comprises a term applied to the kinetic energy of the elevator; and

a global minimum of the total energy consumption of the elevator system is found to be selected as the exit criterion.

3. The method according to claim 1, characterized in that the method further comprises the following stages:

the chromosomes (11) are formulated into the next generation of genetic algorithms by selection, crossover and/or mutation (18).

4. The method of claim 1, wherein: the exit criteria are met when a predetermined value of the cost function, the number of generations, the processing time of the algorithm, or a sufficient uniformity of the number of people is reached.

5. The method according to claim 1, characterized in that the method further comprises the following stages:

defining a cost function (19) of the chromosome (11) such that the cost function includes an energy consumption term and a service time term, and weighting both by preset weighting coefficients; and

in the calculation (14) of the cost function, the elevator model (12, 13) and the current state of the elevator system are used as aids.

6. The method according to claim 1, characterized in that the method further comprises the following stages:

selecting at least one item from the group consisting of energy consumption of the elevator system, passenger waiting time, passenger travel time and passenger running time as an item of the cost function (14); and

when calculating the energy consumption, the kinetic energy of the elevator, the potential energy stored in the elevator, the energy consumed in friction and the energy regenerated to the electricity transmission system are taken into account.

7. The method according to claim 1, characterized in that the method further comprises the following stages:

at least one constraint is defined for the elevator run speed profile according to each run type from a set of quantities, which comprises the maximum speed, the maximum acceleration and the maximum jerk of the elevator, and the jerk is defined as the change in acceleration per time unit.

8. The method according to claim 1, characterized in that the method further comprises the following stages:

an orientation gene (16) of the chromosome (11) is defined for each stationary elevator.

9. Method according to any one of claims 1 to 8, characterized in that the method further comprises the following phases:

"normal" (23), "slight deceleration," "significant deceleration," "slight acceleration," and "significant acceleration" are defined as the operation modes.

10. The method according to claim 9, characterized in that the method further comprises the following stages:

the run type is defined by defining the maximum speed used by the elevator so that: in the run type "normal" (23), the nominal travel speed of the elevator is used, while in the other run types the travel speed of the elevator deviates from the nominal value by an amount which is in the form of a percentage from a preset value.

11. Elevator group control system for controlling a plurality of elevators belonging to an elevator group on the basis of calls issued, wherein an elevator comprises one or more elevator cars (10) in the same elevator shaft, and the control system comprises a genetic algorithm GA optimizer utilizing a genetic algorithm, which GA optimizer is designed to:

forming at least one allocation option, i.e. a chromosome (11), the chromosome (11) containing call data and elevator data for each active landing call or destination call, which data, i.e. genes (17), together determine the elevator car (10) serving the landing call or destination call;

for each chromosome (11), determining a value (14, 19) of a cost function;

formulating at least one chromosome (11) with respect to at least one gene (17);

determining, for each of the at least one chromosome formulated, a value of a cost function (14, 19);

repeating the preparation of chromosomes until an exit criterion is met;

selecting the best chromosome based on the value (19) of the cost function; and

allocating an elevator car (10) in response to the issued call according to the selected gene of the best chromosome,

characterized in that the GA optimizer is further designed to:

information about the type of operation, i.e. an operation type gene (21, 23, 25, 27), is associated with a chromosome (11), in which operation type a speed profile is determined, which chromosome (11) is associated with each call data and each elevator data, according to which speed profile an elevator having an elevator car (10) travels between a departure floor and a call-giving floor defined by the call gene (20, 22, 24, 26), which call gene (20, 22, 24, 26) is associated with the operation type gene (21, 23, 25, 27).

12. Control system according to claim 11, characterized in that the GA optimizer is further arranged:

determining the kinetic energy of the elevator for each elevator trip from a speed profile determined from the car load and the run type of the elevator;

selecting the total energy consumed by the elevator system as a cost function or a part (14, 19) of the cost function, so that the cost function (14, 19) comprises a term applied to the kinetic energy of the elevator; and

a global minimum of the total energy consumption of the elevator system is found to be selected as the exit criterion.

13. Control system according to claim 11, characterized in that the GA optimizer is further arranged:

the chromosomes (11) are formulated into the next generation of genetic algorithms by selection, crossover and/or mutation (18).

14. The control system of claim 11, wherein: in the GA optimizer, the exit criterion is fulfilled when a predetermined value (19) of the cost function, the number of generations, the processing time of the algorithm or a sufficient homogeneity of the population is obtained.

15. Control system according to claim 11, characterized in that the GA optimizer is further arranged:

defining a cost function (14, 19) of the chromosome (11) such that the cost function comprises an energy consumption term and a service time term, and weighting both with preset weighting coefficients; and

in the calculation (14) of the cost function, the elevator model (12, 13) and the current state of the elevator system are used as aids.

16. Control system according to claim 11, characterized in that the GA optimizer is further arranged:

selecting at least one item from the group consisting of energy consumption of the elevator system, passenger waiting time, passenger travel time and passenger running time as an item of the cost function (14); and

when calculating the energy consumption, the kinetic energy of the elevator, the potential energy stored in the elevator, the energy consumed in friction and the energy regenerated to the electricity transmission system are taken into account.

17. Control system according to claim 11, characterized in that the GA optimizer is further arranged:

at least one constraint is defined for the elevator run speed profile according to each run type from a set of quantities, which comprises the maximum speed, the maximum acceleration and the maximum jerk of the elevator, and the jerk is defined as the change in acceleration per time unit.

18. Control system according to claim 11, characterized in that the GA optimizer is further arranged:

an orientation gene (16) of the chromosome (11) is defined for each stationary elevator.

19. Control system according to any of claims 11-18, characterized in that the GA optimizer is further arranged:

"normal" (23), "slight deceleration," "significant deceleration," "slight acceleration," and "significant acceleration" are defined as the operation modes.

20. Control system according to claim 19, characterized in that the GA optimizer is further arranged:

the run type is defined by defining the maximum speed used by the elevator so that: in the run type "normal" (23), the nominal travel speed of the elevator is used, while in the other run types the travel speed of the elevator deviates from the nominal value by an amount which is in the form of a percentage from a preset value.