Method for generating load data of elevator car of elevator system, elevator calculation unit and load estimation system
By using reinforcement learning models in the elevator system to estimate the load data of the elevator car with status data, the problem of not being able to access the load data in real time in the prior art is solved, efficient and accurate load estimation is achieved, and sensor complexity and cost are reduced.
Patent Information
- Application Number
- CN202280099728.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2022-09-12
- Publication Date
- 2025-04-01
AI Technical Summary
During remote monitoring or maintenance, existing elevator systems cannot access the load data of the elevator car in real time, and the sensor measurement requirements are high, resulting in complex and expensive sensor structures and manufacturing processes.
The reinforcement learning model is used to combine the status data to estimate the load data of the elevator car, and the loading event and motion cycle status data at the loading station are obtained through the elevator computing unit and sensor equipment, and the reinforcement learning model is trained using these data.
It realizes efficient and accurate estimation of elevator car load data in elevator system, reduces the complexity and cost of sensors, and improves the monitoring and control accuracy of elevator system.
Smart Images
Figure CN119744246A_ABST
Abstract
Description
Technical Field
[0001] The present invention generally relates to the technical field of elevator systems. In particular, the present invention relates to monitoring elevator systems. Background Art
[0002] An elevator system typically includes an elevator car and an elevator hoisting motor, which is arranged to drive the elevator car between multiple landings along the elevator shaft. The elevator system may also typically include one or more internal sensor devices for providing various operating data of the elevator system. The operating data may include, for example, load data of at least one elevator car. For example, the elevator system may include a load weighing device, which is arranged to the elevator car, for providing load data of the elevator car. However, there may be situations where the elevator control system of the elevator system cannot be accessed and therefore the load data of the elevator car is not available, such as in the case of remote monitoring or maintenance of a third-party elevator system. Alternatively or additionally, there may be situations where alternative ways of defining the load data of the elevator car may be required. High-quality sensor measurements play a vital role in elevator system monitoring. For example, the accuracy, resolution, repeatability and consistency of measurements during the life span can be considered to be critical to the quality of sensor measurements. These requirements may typically result in highly complex and tolerance error-sensitive sensor structures and sensor manufacturing processes, which in turn may result in high sensor prices and strict requirements for the installation and calibration process of the sensor.
[0003] Therefore, there is a need to develop further solutions for defining load data for elevator cars. Summary of the invention
[0004] A simplified overview is presented below to provide a basic understanding of some aspects of various invention embodiments. This overview is not a broad overview of the present invention. It is neither intended to identify key or important elements of the present invention nor to describe the scope of the present invention. The following overview only presents some concepts of the present invention in a simplified form as a preface to a more detailed description of exemplary embodiments of the present invention.
[0005] The object of the invention is to propose a method for generating load data of an elevator car of an elevator system, an elevator calculation unit, a load estimation system, a computer program and a computer-readable medium. Another object of the invention is to enhance the measurement of the load of the elevator car by a method for generating load data of an elevator car of an elevator system, an elevator calculation unit, a load estimation system, a computer program and a computer-readable medium.
[0006] The objects of the invention are achieved by a method, an elevator computing unit, a load estimation system, a computer program and a computer readable medium as defined by the respective independent claims.
[0007] According to a first aspect, a method for generating load data of an elevator car of an elevator system is provided, wherein the method comprises: obtaining status data including at least one load status parameter affected by the load of the elevator car, wherein the status data is obtained during a loading event of the elevator car at a loading landing or during a movement cycle of the elevator car between the loading landing and a destination landing; using the obtained status data as input data of a reinforcement learning model; processing the input data using the reinforcement learning model to generate output data including the load data of the elevator car, the output data representing an estimate of the load of the elevator car; and using the generated load data of the elevator car in the control of the elevator system and / or in the status monitoring of the elevator car.
[0008] The method may further include obtaining measured load data of the elevator car representing a measured load of the elevator car from an elevator drive unit after the elevator car departs from the loading platform, and training the reinforcement learning model using the obtained measured load data of the elevator car.
[0009] The method may also include providing the trained reinforcement learning model to an external entity for further development of the trained reinforcement learning model and / or for providing the trained reinforcement learning model to one or more other elevator systems having the same configuration and conditions as the elevator system.
[0010] The at least one loading condition parameter may include a rope elongation value, a hoist bed to hoist body distance value and / or a hoist inclination value.
[0011] The status data may also include at least one additional status parameter.
[0012] The at least one additional status parameter may comprise landing data, an ambient temperature of the elevator car, an ambient humidity of the elevator car and / or a number of starts of the elevator car.
[0013] A loading event may start with the opening of the doors of the elevator car and a loading event may end with the closing of the doors of the elevator car, the opening of the brakes or the activation of a torque control of the drive unit.
[0014] According to a second aspect, an elevator computing unit for generating load data of an elevator car of an elevator system is provided, wherein the elevator computing unit comprises: a processing unit comprising at least one processor; and a memory unit comprising at least one memory, wherein the at least one memory comprises computer program code; wherein the at least one memory and the computer program code are configured to, together with the at least one processor, cause the elevator computing unit to execute: obtaining condition data comprising at least one load condition parameter affected by the load of the elevator car, wherein the condition data is obtained during a loading event of the elevator car at a loading landing or during an elevator car movement cycle between a loading landing and a destination landing; using the obtained condition data as input data of a reinforcement learning model; processing the input data using the reinforcement learning model to generate output data comprising the load data of the elevator car, wherein the output data represents an estimate of the load of the elevator car; and utilizing the generated load data of the elevator car in control of the elevator system and / or in condition monitoring of the elevator car.
[0015] The elevator computing unit may also be configured to obtain measured load data of the elevator car representing a measured load of the elevator car from the elevator drive unit after the elevator car leaves the loading platform, and to train a reinforcement learning model using the obtained measured load data of the elevator car.
[0016] The elevator computing unit may be configured to provide the trained reinforcement learning model to an external entity for further development of the trained reinforcement learning model and / or for providing the trained reinforcement learning model to one or more other elevator systems having the same configuration and conditions as the elevator system.
[0017] The at least one loading condition parameter may include a rope elongation value, a hoist bed to hoist body distance value and / or a hoist inclination value.
[0018] The status data may also include at least one additional status parameter.
[0019] The at least one additional status parameter may comprise landing data, an ambient temperature of the elevator car, an ambient humidity of the elevator car and / or a number of starts of the elevator car.
[0020] A loading event may start with the opening of the doors of the elevator car and a loading event may end with the closing of the doors of the elevator car, the opening of the brakes or the activation of a torque control of the drive unit.
[0021] According to a third aspect, a load estimation system for generating load data of an elevator car of an elevator system is provided, wherein the load estimation system comprises: at least one sensor device, which is configured to provide condition data including at least one load condition parameter affected by the load of the elevator car, and an elevator calculation unit as described above.
[0022] According to a fourth aspect, a computer program product is provided, wherein the computer program product comprises instructions, which, when the program is executed by a computer, cause the computer to perform the method as described above.
[0023] According to a fifth aspect, a computer-readable medium is provided, wherein the computer-readable medium comprises instructions, which, when executed by a computer, cause the computer to perform the method as described above.
[0024] Various exemplary and non-limiting embodiments of the invention, both as to constructions and methods of operation, together with additional objects and advantages thereof, will be best understood from the following description of specific exemplary and non-limiting embodiments when read in connection with the accompanying drawings.
[0025] The verbs "comprise" and "include" are used in this document as open limitations, which neither exclude nor require the existence of unrecited features. Unless explicitly stated otherwise, the features recited in the dependent claims are mutually freely combinable. Furthermore, it should be understood that the use of "a" or "an" (i.e. the singular) throughout this document does not exclude a plurality. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] Embodiments of the invention are illustrated by way of example and not limitation in the figures of the accompanying drawings.
[0027] Figure 1 An example of an elevator system is schematically shown.
[0028] Figure 2A An example of a load estimation system is schematically shown.
[0029] Figure 2B Another example of a load estimation system is schematically shown.
[0030] Figure 3 An example of a method for generating load data for an elevator car of an elevator system is schematically shown.
[0031] Figure 4 An example of an example of generating load data by applying a reinforcement learning model is schematically shown.
[0032] Figure 5 Another example of the method is schematically shown.
[0033] Figure 6 An example of training a reinforcement learning model by using measured load data is schematically shown.
[0034] Figure 7 An example of components of an elevator computing unit is schematically shown. DETAILED DESCRIPTION
[0035] Figure 1 An example of an elevator system 100 is schematically shown. The elevator system 100 includes at least one elevator car 110 configured to travel along a respective elevator shaft 120 between a plurality of landings 125a-125n. Figure 1 The elevator system 100 of the example includes one elevator car 110 traveling along one elevator shaft 120, however the elevator system 100 may also include an elevator group, i.e., a group of two or more elevator cars 110, each traveling along a separate elevator shaft 120, configured to operate as a unit serving the same landing 125a-125n. The elevator system 100 also includes an elevator control unit, such as an elevator controller 130. The elevator control unit 130 may be configured to at least partially control the operation of the elevator system 100. The elevator control unit 130 may reside, for example, in a machine room 140 or in one of the landings 125a-125n of the elevator system 100. The elevator system 100 also includes a hoist 150, which includes a hoist motor 160 arranged to drive the elevator car 110 along the elevator shaft 120. The hoist 150 may also include one or more known hoist entities, such as one or more wheels, pulleys and / or brakes, etc., which are not described in detail for the sake of clarity. Figure 1 The hoist 150 may be arranged in a machine room 140, which is located above the elevator shaft 120. Figure 1 Alternatively, the hoist 150 may be located inside the elevator shaft 120 (e.g., in a machine room-less elevator system). The elevator control unit 130 may include a drive unit configured to control a hoist motor 160 to drive (i.e., move) the elevator car 110 along the elevator shaft 120. The elevator system 100 may also include one or more other known elevator-related entities, such as elevator ropes, user interface devices, safety circuits and devices, elevator door systems, etc., which are not shown in the figure for clarity. Figure 1 The elevator system 100 also includes a load estimation system 200 (for the sake of clarity, the load estimation system 200 is not shown in the figure). Figure 1 ).
[0036] Figure 2A and Figure 2BSchematically, a simple example of a load estimation system 200 for generating load data 430 of an elevator car 110 of an elevator system 100 is shown. The load estimation system 200 comprises at least one sensor device 210 for providing condition data 410 and an elevator computing unit 220. Figure 2A and Figure 2B In the example of FIG. 2 , one sensor device 210 is shown, but the load estimation system 200 may also include more than one sensor device 210. At least one sensor device 210 is communicatively coupled to the elevator computing unit 220. The communication between the at least one sensor device 210 and the elevator computing unit 220 may be based on one or more known communication technologies, wired or wireless. According to an example, the elevator computing unit 220 may be, for example, the elevator control unit 130 or included by the elevator control unit 130, such as Figure 2A According to another example, the elevator computing unit 220 may be, for example, Figure 2B The elevator computing unit 220 is an external computing unit as shown in the example of . In the context of a computing unit, the term "external" throughout this application means that the computing unit is external to the elevator system 100. The elevator computing unit 220 implemented as an external computing unit can be located on-site and / or off-site. The elevator computing unit 220 implemented as an external computing unit can, for example, include a server, a cloud-based computing unit, a remote computing unit, a computing circuit, and / or any other computing device or network of computing devices external to the elevator system 100. If the elevator computing unit 220 is implemented as an external computing unit, the elevator computing unit 220 can be communicatively coupled to the elevator control unit 130. The communication between the elevator computing unit 220 and the elevator control unit 130 can be based on one or more known communication technologies, wired or wireless.
[0037] Next, by referring to Figure 3 An example of a method for generating load data 430 for an elevator car 110 of an elevator system 100 is described. Figure 3 The method is schematically shown in a flow chart. Figure 3 However, the method may also be applied accordingly to generate load data 420 for more than one elevator car 110 .
[0038] At step 310, the elevator computing unit 220 obtains condition data 410, the condition data 410 including at least one load condition parameter affected by the load of the elevator car 110. As described above, the condition data 410 may be provided by at least one sensor device 210. The elevator computing unit 220 may obtain the condition data 410 from at least one sensor device 210. The condition data 410 is obtained during a loading event of the elevator car 110 at a loading landing. The loading landing may be any of the plurality of landings 125a-125n of the elevator system 100. The loading event may start with the opening of the door of the elevator car 110. The loading event may end with the closing of the door of the elevator car 110, the opening of the brake of the hoist 150, or the activation of the torque control of the drive unit. The elevator control unit 130 may, for example, provide information of the opening of the door of the elevator car 110, the closing of the door of the elevator car 110, the opening of the brake of the hoist 150, and / or the activation of the torque control to the drive unit from the elevator control unit 130. The loading event may include unloading the elevator car 110 and / or loading the elevator car 110 at a loading landing. Obtaining the condition data 410 during the loading event allows the load data 430 of the elevator car 110 to be defined before the elevator car 110 leaves the loading landing. Alternatively or additionally, obtaining the condition data 410 during the loading event allows the load weighing device of the elevator car 110 to be replaced by using the method for generating the load data 430 of the elevator car 110, or the method for generating the load data 430 of the elevator car 110 can be used with the load weighing device of the elevator car 110 to enhance and adjust the load measurement during the life of the elevator system 100. Replacement of the load weighing device enables cost reduction and quality improvement because there are fewer components in the elevator system 100. Alternatively, the condition data 410 is obtained during the elevator car movement cycle between the loading landing and the destination landing. The loading landing may be any of the plurality of landings 125a-125n of the elevator system 100, and the destination landing may be any other of the plurality of landings 125a-125n of the elevator system 100. Obtaining condition data during an elevator car movement cycle enables improved accuracy of the generated load data 430.
[0039] The load of the elevator car 110 affects not only itself, but also the effects of the load of the elevator car 110 can be seen in other components and / or parts of the elevator system 100. At least one loading condition parameter may include such a parameter that changes according to the load of the elevator car 110. For example, unloading and / or loading the elevator car 110 during a loading event, such as unloading one or more passengers and / or cargo from the elevator car 110 and / or loading one or more passengers and / or cargo to the elevator car 110, results in different condition parameters at the beginning of the loading event and the corresponding condition parameters at the end of the loading event. At least one loading condition parameter may include an elevator rope extension value, a hoist floor to hoist body distance value, and / or a hoist inclination value. The at least one sensor device 210 of the load estimation system 200 may include at least one sensor device for each loading condition parameter. For example, if the status data 410 includes an elevator rope extension value, a distance value from the elevator floor to the elevator body, and an elevator inclination value, then the at least one sensor device 210 of the load estimation system 200 may include at least one sensor device configured to provide a rope extension value, at least one sensor device configured to provide a distance value from the elevator floor to the elevator body, and at least one sensor device configured to provide an elevator inclination value.
[0040] The elevator rope elongation value may represent the elongation of the elevator rope. The elevator rope may be a hoisting rope (i.e., a suspension rope) configured to carry (i.e., suspend) the elevator car 110 such that the elevator car 110 is at one end of the hoisting rope and the counterweight is at the other end of the hoisting rope. The hoisting rope elongates depending on the load of the elevator car 110. If the condition data 410 is obtained during a loading event, the elevator rope elongation value may be a difference in the elongation of the elevator rope between the opening and closing of the door of the elevator car 110. The difference in the elongation of the elevator rope between the opening and closing of the door of the elevator car 110 may represent a difference, i.e., a difference between the elongation of the elevator rope when the door of the elevator car 110 is open and the elongation of the elevator rope when the door of the elevator car 110 is closed. The at least one sensor device configured to provide the elevator rope elongation value during the loading event may be, for example, a position sensor device. For example, if the condition data 410 is obtained during a loading event, the elevator rope value may be determined based on a position difference between a first position of the elevator car 110 provided by the position sensor device when the elevator door of the elevator car 110 is open and a second position of the elevator car 110 provided by the position sensor device when the door of the elevator car 110 is closed. Alternatively, if the condition data 410 is obtained during an elevator car movement cycle, the elevator rope elongation value may be an elevator rope elongation with a static load during the elevator car movement cycle. The at least one sensor device configured to provide an elevator rope elongation value during an elevator car movement cycle may, for example, be a visual monitoring device. For example, if the condition data 410 is obtained during an elevator car movement cycle, the visual monitoring device may include a light source and a line camera, wherein light is emitted by the light source toward the elevator rope, and the line camera detects peaks and valleys of the moving rope from opposite sides of the light source, and the elevator rope elongation value may be determined by the distance between the peaks or valleys.
[0041] The elevator floor is a platform on which the main body of the elevator 150 (i.e., the elevator body) is placed. The distance between the elevator floor and the elevator body varies according to the load of the elevator car 110. If the status data 410 is obtained during a loading event, the elevator floor to elevator body distance value may be the elevator floor to elevator body distance difference between the opening and closing of the door of the elevator car 110. The elevator floor to elevator body distance difference between the opening and closing of the door of the elevator car 110 represents the difference between the distance when the door of the elevator car 110 is open and the distance when the door of the elevator car 110 is closed, i.e., the difference value. Alternatively, if the status data 410 is obtained during an elevator car moving cycle, the elevator floor to elevator body distance value may be the distance of the elevator floor to the elevator body with a static load during the elevator car moving cycle. The at least one sensor device configured to provide the elevator floor to elevator body distance value may be, for example, a distance sensor device. Non-limiting examples of distance sensor devices may include, but are not limited to, laser distance sensor devices, capacitive distance sensor devices, or inductive displacement sensor devices, among others.
[0042] The hoist 150 may be tilted relative to gravity. For example, in the case of the machine room-less elevator system 100, when a load is applied to the elevator car 110, the hoist 150 may be tilted toward the guide rail. The hoist inclination varies according to the load of the elevator car 110. If the condition data 410 is obtained during a loading event, the hoist inclination value may be a hoist inclination difference between the opening and closing of the door of the elevator car 110. The hoist inclination difference between the opening and closing of the door of the elevator car 110 represents the difference between the hoist inclination when the door of the elevator car 110 is open and the hoist inclination when the door of the elevator car 110 is closed, i.e., the difference value. Alternatively, if the condition data 410 is obtained during an elevator car moving cycle, the hoist inclination value may be a hoist inclination with a static load during the elevator car moving cycle. The at least one sensor device configured to provide a hoist inclination value may be, for example, a tilt sensor device. Non-limiting examples of tilt sensor devices may include, but are not limited to, inclinometers, micro-electromechanical systems (MEMS) based sensor devices, fluid based sensor devices, potentiometers, and the like.
[0043] According to an example, the obtained condition data 410 may also include at least one additional condition parameter. The at least one additional condition parameter may be independent of the load of the elevator car 110. The at least one additional condition parameter may, for example, include landing data, the ambient temperature of the elevator car 110, the ambient humidity of the elevator car 110, and / or the number of starts of the elevator car 110. The landing data may include loading landing information representing a loading landing and / or destination landing information representing a destination landing. According to a non-limiting example, the obtained condition data 410 may include at least a combination of an elevator rope elongation value and loading landing information. This is an advantageous example combination of at least one loading condition parameter and at least one additional condition parameter, because the elevator rope elongation value may depend on the loading landing. The at least one sensor device 210 of the load estimation system 200 may include at least one sensor device for each additional condition parameter. Alternatively, at least some of the additional condition parameters may be provided with the same at least one sensor device, for example, the landing data and the number of starts of the elevator car 110. The at least one sensor device configured to provide the ambient temperature of the elevator car 110 may, for example, be a temperature sensor device. The at least one sensor device configured to provide the ambient humidity of the elevator car 110 may, for example, be a humidity sensor device. The at least one sensor device configured to provide the landing data and / or the number of starts of the elevator car 110 may, for example, include an elevator motion control unit and / or a cloud-based unit. The elevator motion control unit and the cloud-based computing unit may be communicatively coupled to the elevator computing unit 220. The communication between the elevator motion control unit and the elevator computing unit 220 may be based on one or more known communication technologies, wired or wireless. The communication between the cloud-based unit and the elevator computing unit 220 may be based on one or more known wired or wireless communication technologies.
[0044] In step 320, the elevator computing unit 220 uses the obtained situation data 410 as input data of the reinforcement learning model 420. The reinforcement learning model 420 may be stored in the memory unit 720 of the elevator computing unit 220. The reinforcement learning model 420 is a machine learning model based on rewarding desired behaviors and / or punishing undesired behaviors. The reinforcement learning model 420 with predefined initial model parameters may be provided to the elevator computing unit 220, and then the reinforcement learning model 420 may be trained during debugging of the method and / or during use of the method.
[0045] In step 330, the elevator computing unit 220 processes the input data (i.e., the status data 410) using the reinforcement learning model 420 to produce (i.e., generate) output data including the load data 430 of the elevator car 110, which represents an estimation of the load of the elevator car 110, i.e., a numerical estimation of the load of the elevator car 110. In other words, the elevator computing unit 220 is able to estimate the load data 430 of the elevator car 110 by applying the obtained status data to the reinforcement learning model 420. Figure 4 A simple example of generating load data 430 by applying a reinforcement learning model 420 is schematically shown, wherein the obtained condition data 410 is used as input data of the reinforcement learning model 420. Although an estimate of the load of the elevator car 110 can be generated by using the reinforcement learning model 420 when the condition data 410 includes one load condition parameter, the more load condition parameters and / or additional condition parameters the condition data includes, the more accurate the estimate of the load of the elevator car 110 generated. Using the reinforcement learning model 420 to generate the load data 430 of the elevator car 110 enables adaptation to changes in the configuration of the elevator system 100, changes in the environment of the elevator system 100, and / or aging of components of the elevator system 100.
[0046] At step 340, the elevator computing unit 220 uses the generated load data 430 of the elevator car 110 in the control of the elevator system 100 and / or the condition monitoring of the elevator car 100. The control of the elevator system 100 may, for example, include controlling the brakes of the hoist 150, the drive unit and / or the hoist motor 160. According to an example, the generated load data 430 of the elevator car 110 may be used by the drive unit to control the hoist motor 160. Using the generated load data of the elevator car 110 in the control of the elevator system 100 improves the elevator ride comfort, for example, by improving the smoothness of the elevator car 110 leaving the loading landing. For example, for comparison, the drive unit may provide the measured load data of the elevator car 110 after the elevator car 110 leaves the loading landing, but the measured load data is too late to improve the smoothness of the elevator car 110 leaving the loading landing. Using the generated load data 430 in the condition monitoring enables providing important information for condition monitoring, which in turn improves the safety of the elevator system 100.
[0047] According to an example, the method may further include training a reinforcement learning model 420 . Figure 5 An example of a method for generating load data 430 for an elevator car 110 is schematically shown, the method also comprising training a reinforcement learning model 420 .
[0048] At step 510, the elevator computing unit 220 obtains measured load data 610 of the elevator car 110, which represents the measured load of the elevator car 100. The measured load data 610 of the elevator car 110 may be obtained from the drive unit after the elevator car 110 leaves the loading landing.
[0049] At step 520 , the elevator computing unit 220 uses the obtained measured load data 610 of the elevator car 110 to train the reinforcement learning model 420 . Figure 6 A simple example of training the reinforcement learning model 420 by using the obtained measured load data 610 is schematically shown. As a result of training the reinforcement learning model 420, a trained reinforcement learning model 620 may be generated. The trained reinforcement learning model 620 may be stored in the memory unit 720 of the elevator computing unit 220. The trained reinforcement learning model 620 may replace the previously stored reinforcement learning model 420. In other words, after training, the trained reinforcement learning model 620 may replace the previously used reinforcement learning model 420 for generating the load data of the elevator car 110 at steps 320 and 330 of the method. The measured load data 610 accurately indicates the actual load of the elevator car 110. Therefore, using the measured load data 610 in the training of the reinforcement learning model 420 improves the accuracy of the trained reinforcement learning model 620.
[0050] According to an example, at step 530, the trained reinforcement learning model 620 may be further provided to an external entity for further developing the trained reinforcement learning model 620 and / or for providing the trained reinforcement learning model 620 to one or more other elevator systems having substantially the same configuration and conditions as the elevator system 100 including the elevator car 110. For example, the elevator computing unit 220 may provide the trained reinforcement learning model 620 to an external entity, wherein the trained reinforcement learning model 620 may be further developed to further enhance the trained reinforcement learning model 620, and after further developing the trained reinforcement learning model 620, it may be provided to one or more other elevator systems for generating load data of at least one elevator car of the one or more other elevator systems. Since the one or more other elevator systems have substantially the same configuration and conditions as the elevator system 100, the trained reinforcement learning model 620 and thus the above-described method may be compatible for use with the one or more other elevator systems. In the context of an entity, the term "external" throughout this application means that the entity is outside the elevator system 100. The external entity may be located on-site and / or off-site. The external entity may include, for example, a server, a cloud server, a remote server, a computing circuit, and / or any other computing device or network of computing devices external to the elevator system 100 .
[0051] Figure 7 An example of components of an elevator computing unit 220 is schematically shown. The elevator computing unit 220 may include a processing unit 710 having one or more processors, a memory unit 720 having one or more memories, a communication unit 730 having one or more communication devices, and a possible user interface (UI) unit 740. The mentioned elements may be communicatively coupled to each other via, for example, an internal bus. The memory unit 720 may store and maintain portions of a computer program (code) 725, a reinforcement learning model 420, a trained reinforcement learning model 620, obtained condition data 410, generated load data 430 of the elevator car 110, obtained measured load data 610, and any other data. The computer program 725 may include instructions that, when executed by the processing unit 710 of the elevator computing unit 220, may cause the processing unit 710 and, therefore, the elevator computing unit 220 to perform the desired tasks of the elevator computing unit 220, for example, one or more of the above-described method steps. Thus, the processing unit 710 may be arranged to access the memory unit 720 and retrieve any information from and store any information to the memory unit 720. For clarity, the processor herein refers to any unit suitable for processing information and controlling the operation of the elevator computing unit 220 and other tasks. The operation may also be implemented with a microcontroller solution with embedded software. Similarly, the memory unit 720 is not limited to a certain type of memory, but any memory type suitable for storing the described multiple pieces of information may be applied in the context of the present invention. The communication unit 730 provides one or more communication interfaces for communicating with any other unit (e.g., at least one sensor device 210, a drive unit, an external entity, one or more databases) or with any other unit. The user interface unit 740 may include one or more input / output (I / O) devices such as buttons, keyboards, touch screens, microphones, speakers, displays, etc., for receiving user input and outputting information. The computer program 725 may be a computer program product that may be included in a tangible non-volatile (non-transitory) computer readable medium carrying computer program code 725 embodied therein for use with a computer (ie, the elevator computing unit 220 ).
[0052] The specific examples provided in the description given above should not be construed as limiting the applicability and / or interpretation of the appended claims.The lists and groups of examples provided in the description given above are not exhaustive unless expressly stated otherwise.
Claims
1. A method for generating load data (430) of an elevator car (110) of an elevator system (100), the method comprising: obtaining (310) condition data (410), the condition data (410) comprising at least one load condition parameter affected by the load of the elevator car (110), wherein the condition data (410) is obtained during a loading event of the elevator car (110) at a loading landing or during an elevator car movement cycle between a loading landing and a destination landing; Using the status data obtained in (320) as input data for a reinforcement learning model (420); processing (330) the input data using the reinforcement learning model (420) to generate output data including load data (430) for the elevator car (110), the output data representing an estimate of the load of the elevator car (110); and The generated load data (430) of the elevator car (110) are used (340) in controlling the elevator system (100) and / or in condition monitoring of the elevator car (110).
2. The method according to claim 1, further comprising: obtaining (510) measured load data (610) of the elevator car (110) from an elevator drive unit after the elevator car (110) leaves the loading landing, the measured load data (610) representing a measured load of the elevator car (110), and The reinforcement learning model (420) is trained using (520) the obtained measured load data (610) of the elevator car (110).
3. The method of claim 2, further comprising providing (530) the trained reinforcement learning model (620) to an external entity to further develop the trained reinforcement learning model (620) and / or provide the trained reinforcement learning model (620) to one or more other elevator systems having the same configuration and conditions as the elevator system (100).
4. A method according to any one of the preceding claims, wherein: The at least one loading condition parameter comprises a rope elongation value, a hoist bottom plate to hoist body distance value and / or a hoist inclination value.
5. A method according to any one of the preceding claims, wherein: The status data (410) also includes at least one additional status parameter.
6. The method according to claim 5, wherein: The at least one additional status parameter comprises landing data, an ambient temperature of the elevator car (110), an ambient humidity of the elevator car (110) and / or a number of starts of the elevator car (110).
7. The method according to any of the preceding claims, wherein the loading event starts with the opening of a door of the elevator car (110), and wherein the loading event ends with the closing of the door of the elevator car (110), the opening of a brake, or the activation of a torque control of a drive unit.
8. An elevator calculation unit (220) for generating load data (430) of an elevator car (110) of an elevator system (100), the elevator calculation unit (220) comprising: A processing unit (710), comprising at least one processor; and a memory unit (720) comprising at least one memory, the at least one memory comprising a computer program code (725); The at least one memory and the computer program code (725) are configured to, together with the at least one processor, cause the elevator computing unit (220) to execute: obtaining condition data (410), the condition data (410) comprising at least one load condition parameter affected by the load of the elevator car (110), wherein the condition data (410) is obtained during a loading event of the elevator car (110) at a loading landing or during an elevator car movement cycle between a loading landing and a destination landing; Using the obtained status data (410) as input data of a reinforcement learning model (420); processing the input data using the reinforcement learning model (420) to generate output data including load data (430) for the elevator car (110), the output data representing an estimate of the load of the elevator car (110); and The generated load data (430) of the elevator car (110) are used in the control of the elevator system (100) and / or in the condition monitoring of the elevator car (110).
9. The elevator computing unit (220) according to claim 8, further configured to: obtaining measured load data (610) of the elevator car (110) from an elevator drive unit after the elevator car (110) leaves the loading landing, the measured load data (610) representing a measured load of the elevator car (110), and The reinforcement learning model (420) is trained using the obtained measured load data (610) of the elevator car (110).
10. The elevator computing unit (220) according to claim 9, wherein: The elevator computing unit (220) is configured to provide the trained reinforcement learning model (620) to an external entity to further develop the trained reinforcement learning model (620) and / or provide the trained reinforcement learning model to one or more other elevator systems having the same configuration and conditions as the elevator system (100).
11. The elevator computing unit (220) according to any one of claims 8 to 10, wherein: The at least one loading condition parameter comprises a rope elongation value, a hoist bottom plate to hoist body distance value and / or a hoist inclination value.
12. The elevator computing unit (220) according to any one of claims 8 to 11, wherein: The status data also includes at least one additional status parameter.
13. The elevator computing unit (220) according to claim 12, wherein: The at least one additional status parameter comprises landing data, an ambient temperature of the elevator car (110), an ambient humidity of the elevator car (110) and / or a number of starts of the elevator car (110).
14. The elevator computing unit (220) according to any one of claims 8 to 13, wherein: The loading event starts with the opening of a door of the elevator car (110), and wherein the loading event ends with the closing of the door of the elevator car (110), the opening of a brake, or the activation of a torque control of a drive unit.
15. A load estimation system (200) for generating load data (430) of an elevator car (110) of an elevator system (100), the load estimation system (200) comprising: at least one sensor device (210) configured to provide condition data (410), the condition data (410) comprising at least one loading condition parameter affected by a load of the elevator car (110), and An elevator computing unit (220) according to any one of claims 8 to 14.
16. A computer program product (725) comprising instructions which, when said program (725) is executed by a computer, cause said computer to perform the method according to any one of claims 1 to 7.
17. A computer-readable medium comprising instructions which, when executed by a computer, cause the computer to perform the method according to any one of claims 1 to 7.