Elevator group scheduling method, electronic equipment and computer readable storage medium

Abstract

The invention discloses an elevator group scheduling method, electronic equipment and a computer readable storage medium. The method comprises the steps that an elevator calling request initiated to an elevator group is detected; the elevator group comprises a plurality of elevators; determining a target request category corresponding to the elevator calling request in a plurality of candidate request categories according to elevator calling data associated with the elevator calling request; determining a target model corresponding to the target request category in a plurality of candidate models; based on the running state data of each elevator, the target model is used for calculating the predicted execution duration of running of each elevator in response to the elevator calling request; and the elevator group is dispatched according to the predicted execution duration of each elevator. According to the technical scheme, the technical problem that an existing elevator system is low in overall operation efficiency is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of elevator control, and in particular to a dispatching method for an elevator group, an electronic device, and a computer-readable storage medium. Background Art

[0002] Elevator systems are a core component of vertical transportation and play an increasingly important role in architectural design, especially in office buildings with dense traffic and concentrated elevator usage peaks. Elevator traffic has the characteristics of fuzziness, nonlinearity, susceptibility to interference, incomplete information, and multiple goals, making elevator traffic planning extremely complex. Although traditional elevator parallel and group control technologies can improve transportation capacity to a certain extent, reduce repeated responses to the same elevator demand, and enable elevators to respond to closer calls, thereby shortening waiting time. However, during periods of high traffic and peak usage, these methods may make it difficult for passengers on the middle and lower floors to take the elevator, and the elevators stop frequently, reducing overall operating efficiency. Therefore, how to improve the overall operating efficiency of the elevator system has become an urgent problem to be solved. Summary of the invention

[0003] The main purpose of the present invention is to provide an elevator group scheduling method, electronic equipment, and computer-readable storage medium, aiming to solve the technical problem of low overall operating efficiency of existing elevator systems.

[0004] To achieve the above object, an embodiment of the present invention provides a dispatching method for an elevator group, comprising:

[0005] An elevator call request initiated to an elevator group is detected; the elevator group includes a plurality of elevators;

[0006] determining, according to the elevator call data associated with the elevator call request, a target request category corresponding to the elevator call request from a plurality of candidate request categories;

[0007] Determining a target model corresponding to the target request category among multiple candidate models;

[0008] Based on the operation status data of each elevator, using the target model to calculate the predicted execution time of each elevator in response to the elevator call request;

[0009] The elevator group is scheduled according to the predicted execution time of each elevator.

[0010] Optionally, the predicting, based on the operation status data of each elevator, using the target model to predict the predicted execution time of each elevator in response to the elevator call request, includes:

[0011] Obtain the operating status data of each elevator;

[0012] For each elevator, construct a response feature vector of the elevator according to the elevator call data and the operation status data; the response feature vector is used to represent the theoretical execution time of the elevator in response to the elevator call request;

[0013] For each elevator, the response feature vector of the elevator is input into the target model to obtain the predicted execution time.

[0014] Optionally, the multiple candidate models correspond one-to-one to the multiple candidate request categories, and each of the candidate models is trained using associated data of historical elevator call requests belonging to the corresponding candidate request category, and the associated data of the historical elevator call requests include one or more of the following data obtained according to the historical elevator call requests: the historical elevator call data, the operating status data of the responding elevator, and the actual execution time of the responding elevator; the responding elevator is the elevator that actually responds to the historical elevator call request.

[0015] Optionally, the weight of each variable in the objective function of the multiple candidate models is independently set, the objective function includes at least one variable, and the at least one variable includes at least one variable related to the actual execution time of the historical elevator call request.

[0016] Optionally, before determining the target request category corresponding to the elevator call request from a plurality of candidate request categories according to the elevator call data associated with the elevator call request, the method further includes:

[0017] Acquire a plurality of historical elevator call data of the elevator group, wherein the plurality of historical elevator call data correspond one by one to a plurality of historical elevator call requests initiated to the elevator group;

[0018] According to the similarity between every two of the plurality of historical elevator call data, the plurality of historical elevator call requests are divided into a preset number of candidate request categories.

[0019] Optionally, the dividing the plurality of historical elevator call requests into a preset number of candidate request categories according to the similarity between every two of the plurality of historical elevator call data includes:

[0020] For each of the historical elevator call requests, constructing a historical elevator call feature vector of the historical elevator call request according to the corresponding historical elevator call data;

[0021] The plurality of historical elevator call requests are clustered into a preset number of candidate request categories according to the distance between any two of the historical elevator call feature vectors.

[0022] Optionally, determining, according to the elevator call data associated with the elevator call request, a target request category corresponding to the elevator call request from a plurality of candidate request categories includes:

[0023] Acquiring reference call data for each candidate request category; the reference call data for any candidate request category being used as a reference sample of call data associated with a call request belonging to the candidate request category;

[0024] calculating the similarity between each of the reference elevator call data and the elevator call data;

[0025] The candidate request category corresponding to the reference elevator call data with the highest similarity is determined to obtain the target request category.

[0026] Optionally, before obtaining the benchmark elevator call data of each candidate request category, the method further includes:

[0027] Acquiring historical call data corresponding to historical call requests belonging to each of the candidate request categories;

[0028] The statistical center sample of the historical elevator call data corresponding to each of the candidate request categories is calculated to obtain the benchmark elevator call data of each of the candidate request categories.

[0029] An embodiment of the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the elevator group scheduling method provided in any embodiment of the present invention.

[0030] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein the program is executed by a processor to implement the elevator group scheduling method provided in any embodiment of the present invention.

[0031] An embodiment of the present invention further provides a computer program product, including a computer program, wherein the computer program is executed by a processor to implement the elevator group scheduling method provided in any embodiment of the present invention.

[0032] An embodiment of the present invention further provides an elevator system, comprising a plurality of elevators, and also comprising an electronic device provided by an embodiment of the present invention, wherein the electronic device is used to execute the elevator group scheduling method provided by an embodiment of the present invention.

[0033] In one or more technical solutions provided by the above-mentioned embodiments of the present invention, after detecting an elevator call request initiated to an elevator group, a target request category corresponding to the elevator call request is determined from multiple candidate request categories based on the elevator call data associated with the elevator call request, and a target model corresponding to the target request category can be determined from multiple candidate models, and then based on the operating status data of each elevator, the target model is used to calculate the predicted execution time of each elevator in response to the elevator call request, and the elevator group is scheduled according to the predicted execution time of each elevator. By dividing the elevator call requests into different request categories and selecting the corresponding target model to predict the execution time of each elevator, a more accurate prediction result can be obtained, and the user's elevator call request can be responded to more quickly, thereby improving the operating efficiency of the elevator group. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0035] Figure 1 A schematic diagram of a process for scheduling an elevator group provided by an embodiment of the present invention Figure 1 ;

[0036] Figure 2 A schematic diagram of a process for scheduling an elevator group provided by an embodiment of the present invention Figure 2 ;

[0037] Figure 3 A schematic diagram of the structure of a neural network model in an elevator group scheduling method provided by an embodiment of the present invention;

[0038] Figure 4 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention.

[0039] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] Those skilled in the art will readily appreciate other embodiments of the invention after considering the specification and practicing the invention disclosed herein. The invention is intended to cover any variations, uses or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be regarded as exemplary only, and the inventive concept of the invention is indicated by the claims.

[0042] The terms used in the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms of "a", "said" and "the" used in the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items. Depending on the context, the word "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determining".

[0043] It should also be noted that the terms "comprises", "includes" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, product or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, product or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, product or device including the element.

[0044] In addition, in the present invention, descriptions such as "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0045] In addition, the technical solutions between the various embodiments of the present invention can be combined with each other, but it must be based on the fact that ordinary technicians in the field can implement it. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0046] An embodiment of the present invention provides a dispatching method for an elevator group, referring to Figure 1 , the dispatching method of the elevator group comprises the following steps:

[0047] Step 101, detecting an elevator call request initiated to an elevator group; the elevator group includes a plurality of elevators;

[0048] Step 102, determining a target request category corresponding to the elevator call request from a plurality of candidate request categories according to the elevator call data associated with the elevator call request;

[0049] Step 103, determining a target model corresponding to the target request category from multiple candidate models;

[0050] Step 104, based on the operation status data of each elevator, using the target model to calculate the predicted execution time of each elevator in response to the elevator call request;

[0051] Step 105, scheduling the elevator group according to the predicted execution time of each elevator.

[0052] The elevator group may include multiple elevators. The elevator described in the embodiment of the present application is an electric drive device for transporting passengers or goods vertically or obliquely. The types of elevators include vertical passenger elevators, freight elevators for freight transportation, vehicle transportation equipment in multi-story parking lots, and escalators and moving walkways. Multiple elevators in the elevator group can be scheduled and operated by a control system. A call elevator group control device can be set at each service layer of the elevator group. There can be one or more call elevator group control devices on each layer. For example, a call elevator group control device can be set next to each elevator door. The same or different operation panels can be set on each call elevator group control device. For example, next to an elevator that cannot provide upward operation service, the operation panel of the call elevator group control device can only be provided with a downward operation button. A user using the call elevator group control device can call an elevator that can provide service for the current layer by operating the call elevator group control device at any layer. The service layer refers to the layer where the elevator can provide service. Optionally, the service layers of different elevators in the elevator group may be the same or different. The layer that issues a call elevator request can call the elevator that can provide service for the current layer. In some embodiments, buttons for running up and / or running down may be provided on the operation panel of the elevator group control device to provide the user with the elevator call function of requesting to run up and / or requesting to run down. In other embodiments, a button for setting a specific destination floor may be provided on the operation panel of the elevator group control device to provide the user with the elevator call function of requesting to run to a specific floor. When any user registers an elevator call request through the elevator group control device and sets the up / down / destination floor, the executor of the embodiment of the present application detects the elevator call request initiated to the elevator group. In some embodiments, the executor of the embodiment of the present application may be a control system capable of controlling the elevator group, and obtains the elevator call request through a communication connection with each elevator group control device.

[0053] According to the elevator call data associated with the elevator call request, the target request category corresponding to the elevator call request can be determined from multiple candidate request categories. The elevator call data associated with the elevator call request is the demand information data used to describe the elevator call request, including but not limited to the direction of the request, the calling floor (i.e., the current floor), the destination floor (i.e., the floor the user wants to go to), the time of initiating the elevator call request (specifically including but not limited to the specific time of the day or the time period, the day of the week, the day of the month, etc.), and the identity of the passenger initiating the elevator call request (for example, the passenger initiates the elevator call request by swiping the card, and the passenger identity is determined based on the card identification). Classification is performed based on the elevator call data, and the category to which the elevator call request belongs is determined from multiple pre-provided candidate request categories. In some embodiments, at least one dimension of the elevator call data can be divided into multiple ranges, for example, the elevator call time is divided into different time periods of the day, so as to divide the category of the elevator call request based on the specific range of the elevator call data of this dimension. It can be understood that the specific category can be determined in combination with other dimensional data. In other embodiments, historical elevator call data of multiple historical elevator call requests can be obtained to train a classification model based on artificial intelligence, and then the trained classification model is used to determine the category of the current elevator call request.

[0054] After determining the target request category to which the elevator call request belongs, the target model corresponding to the target request category is selected to predict the operation of each elevator in response to the elevator call request. Different candidate request categories correspond to different candidate models. Each candidate model can be a trained artificial intelligence model. In some embodiments, each candidate model can be a different type of artificial intelligence model based on different frameworks. In some embodiments, each candidate model can be based on the same framework but have different specific structures, such as different numbers of network layers of the neural network model. In some other embodiments, each candidate model can be based on the same framework and structure, but have different weights / parameters after training. Optionally, the network structure of the candidate model can choose to use a fuzzy neural network, which combines fuzzy theory with neural networks, and maps through activation functions in the neural network to realize fuzzy membership calculation and fuzzy reasoning of the input value, and output a predicted value with fuzziness and historical data rules.

[0055] Specifically, elevator request data samples belonging to each candidate request category can be used to train candidate models of the corresponding category. An elevator request data sample can correspond to a historical elevator request. The elevator request data sample can include monitoring data of a responding elevator, where the responding elevator is the elevator that actually responds to the historical elevator request. The monitoring data includes but is not limited to elevator call data, operating status data and actual execution time of the historical elevator request. The elevator call data is data obtained when a user initiates an elevator call request, including but not limited to the time of initiating the elevator call request, the calling floor, the destination floor, and the user identity (for example, the user calls the elevator by swiping a card, etc., and the user identity is determined based on the identification of the swiped card). The operating status data is the operating parameters of the elevator monitored when the elevator call request is initiated. It can be understood that initiating the elevator call request can refer to initiating the elevator call request. The actual execution time is related to the time of arriving at the departure floor and the time of arriving at the destination floor after the elevator responds to the elevator call request. It can be understood that when the elevator request data sample corresponds to a historical elevator call request, the above-mentioned elevator refers to the monitoring data of the elevator that actually responds to the elevator call request in the elevator group. After obtaining the above monitoring data, an input vector can be constructed based on the elevator call data and the operating status data, and an output vector can be constructed based on the actual execution time. The model can be trained with the actual execution time as the output training target, so that the trained model can be used to predict the execution time of the elevator responding to the elevator call request.

[0056] After selecting the target model corresponding to the target request category to which the elevator call request belongs, the target model is used to predict the predicted execution time of each elevator in response to the current elevator call request based on the operation status data of each elevator. The operation status data refers to the parameters collected within a preset period before and after receiving the elevator call request to represent the operation status of the elevator, including but not limited to whether the elevator is currently running or idle, the current number of elevator stops / operating position / number of floors ahead, whether the elevator door is open, whether the elevator is available (for example, if it is under maintenance, the elevator is not available), call data received in the elevator car, call data obtained at any serviceable floor outside the elevator car, call data of the elevator call request to be assigned, etc. Specifically, for each elevator, the operation status data of the elevator is used to construct a feature vector, and the feature vector is used to input into the target model. Optionally, the feature vector input into the target model can also be constructed in combination with the elevator call data. The elevator call data includes but is not limited to the call floor, destination floor, call time, etc. in the elevator call data.

[0057] After constructing the feature vector for each elevator, the feature vector of each elevator is input into the target model, so that the target model can output the predicted execution time for the elevator according to the feature vector of the elevator input this time. It can be understood that when the feature vector is input into the target model, each elevator can be relatively independent, for example, multiple threads can be initiated, each thread can use the target model to calculate the feature vector of an elevator independently, and any thread can calculate the feature vectors of different elevators in sequence. In other embodiments, the structure of the target model can also be configured to receive the feature vectors of multiple elevators at the input layer at the same time. Accordingly, when training the target model, the input and output training data can be constructed according to the corresponding structure. The predicted execution time of any elevator refers to the time that the elevator responds to the elevator call request of the elevator to be scheduled and allocated, for example, the waiting time of passengers when the elevator runs to the calling layer. It can be understood that in the application scenario where the elevator group control device is equipped with the function of calling the destination layer, the predicted execution time can also include the passenger elevator ride time from the calling layer to the destination layer, and the total time of the passenger waiting time and the passenger elevator ride time. In a specific application example, the predicted execution time output by the target model = w1×passenger waiting time + w2×passenger riding time, w1 and w2 are the distribution weights of the two time lengths respectively. The weight values ​​can be determined in advance or obtained by training the target model.

[0058] After the predicted execution time is calculated for each elevator, the corresponding elevator can be assigned to the elevator call request based on the prediction result, and the assigned elevator can be scheduled to stop at the calling floor. Optionally, if the destination floor can be set when the elevator call request is initiated, after the assigned elevator is determined, a dispatch instruction can be issued to the assigned elevator to control it to stop at both the calling floor and the destination floor. In other embodiments, if the destination floor is not set when the elevator call request is initiated, after the assigned elevator is determined, the dispatch instruction issued to the assigned elevator may include controlling it to stop at the calling floor, and after the called passenger takes the elevator, the destination floor specified by the passenger is received through the instruction receiving device in the car, and then the elevator stops at the destination floor according to the passenger's instruction in the elevator.

[0059] In one or more technical solutions provided by the above-mentioned embodiments of the present invention, after detecting an elevator call request initiated to an elevator group, a target request category corresponding to the elevator call request is determined from multiple candidate request categories based on the elevator call data associated with the elevator call request, and a target model corresponding to the target request category can be determined from multiple candidate models, and then based on the operating status data of each elevator, the target model is used to calculate the predicted execution time of each elevator in response to the elevator call request, and the elevator group is scheduled according to the predicted execution time of each elevator. By dividing the elevator call requests into different request categories and selecting the corresponding target model to predict the execution time of each elevator, a more accurate prediction result can be obtained, and the user's elevator call request can be responded to more quickly, thereby improving the operating efficiency of the elevator group.

[0060] In some optional technical solutions of this embodiment, step 104 uses the target model to predict the predicted execution time of each elevator in response to the elevator call request based on the operation status data of each elevator, which may specifically include executing the following steps:

[0061] Obtain the operating status data of each elevator;

[0062] For each elevator, a response feature vector of the elevator is constructed based on the elevator call data and the operation status data; the response feature vector is used to represent the theoretical execution time of the elevator in response to the elevator call request;

[0063] For each elevator, the response feature vector of the elevator is input into the target model to obtain the predicted execution time.

[0064] The running status data of each elevator can be automatically uploaded by each elevator, and can be communicated with the executor of the embodiment of the present invention based on a pre-configured communication method, or obtained through monitoring sensors pre-configured on each elevator, which is not limited by the embodiment of the present invention. After the running status data is obtained, a response feature vector input to the target model can be constructed based on the elevator call data and the running status data. For example, the theoretical execution time represented by the response feature vector includes but is not limited to: the waiting time of the current elevator call request, the elevator ride time of the current elevator call request, the waiting time increment of the allocated elevator call request, the elevator ride time increment of the allocated elevator call request, the elevator ride time increment of the passengers in the elevator, and the elevator single round trip operation time increment. When calculating the above theoretical execution time, the following data can be specifically combined for calculation, at least one of the following elevator running status data: the elevator running direction, the current position, whether the door is open, the number of floors to be stopped, and the call floor and destination floor in the elevator call data.

[0065] Since the prediction result output by the target model is the predicted execution time of the elevator responding to the elevator call request, setting the feature vector input to the target model as the response feature vector related to the theoretical execution time can make the input vector and the output vector closely related and match in the feature dimension, thereby making the decision of the target model more interpretable, obtaining more accurate prediction results, and improving the training efficiency of the target model.

[0066] In some optional technical solutions of this embodiment, multiple candidate models correspond to multiple candidate request categories one by one, and each candidate model is trained using the associated data of historical elevator call requests belonging to the corresponding candidate request category, and the associated data of the historical elevator call requests include one or more of the following data obtained according to the historical elevator call requests: historical elevator call data, operating status data of the responding elevator, and actual execution time of the responding elevator, wherein the responding elevator is the elevator that actually responds to the historical elevator call request, and the responding elevator is the elevator that is actually assigned to the historical elevator call request at that time, is dispatched to the calling floor, and runs to the destination floor. Specifically, before executing step 103 to determine the target model corresponding to the target request category in multiple candidate models, the method may also include the following steps:

[0067] For each candidate request category, obtain the historical elevator call data, the operation status data of the responding elevator and the actual execution time;

[0068] For each candidate request category, a candidate model corresponding to the candidate request category is trained according to the historical call data of each historical call request, the operating status data of the responding elevator, and the actual execution time.

[0069] The above technical solution customizes training models for different candidate request categories, so that each trained candidate model can more accurately capture the unique patterns and features of various elevator call requests. This targeted training method allows the model to learn the data characteristics of specific request categories. Through refined model training, the accuracy of the prediction results of each candidate model is optimized. Using more accurate prediction results to allocate elevators can improve the overall operating efficiency of the elevator system.

[0070] Furthermore, in the above-mentioned step of training the candidate models corresponding to the candidate request categories, in some optional technical solutions, the weights of the variables in the objective function of the candidate models can be set for the candidate models corresponding to each candidate request category, and the objective function includes at least one variable, and the at least one variable includes at least one variable related to the actual execution time of the historical elevator call request. In other words, the objective function weights of the multiple candidate models are set relatively independently, and the objective function weights include the weights of the variables in the objective function of the candidate models, and the variables in the objective function are related to the actual execution time of the historical elevator call request.

[0071] The objective function (also called loss function or cost function) is a function used to measure the difference between the model output and the actual target. In some embodiments, the variables in the objective function may include the passenger waiting time and the passenger riding time. The objective function may be w1×passenger waiting time+w2×passenger riding time, where w1 and w2 are the weights of the two time lengths. For each candidate request category, the weight values ​​w1 and w2 in the objective function may be set independently. A specific example is provided as follows:

[0072] Candidate request categories may include the following four categories: morning peak, lunch peak, evening peak, and non-peak. For requests in the morning peak category, the weight w1 is set to the highest value among multiple categories and exceeds the preset value of w2. The weight w1 of the lunch peak and evening peak is set to be lower than the weight w1 of the morning peak and close to or equal to w2. The weight w1 of the non-peak category is the lowest among multiple categories and may be less than the weight w2. The weight w2 of the non-peak category is the highest among multiple categories. The principle is that the time to take the elevator is usually less than the waiting time, and people need to reach the destination floor more quickly during the morning peak. Therefore, setting the waiting time to have a larger weight can take into account the impact of the waiting time on the objective function results more. During other peak hours, people's demand for quickly reaching the destination floor is reduced, which can reduce the impact of the waiting time on the objective function results. During non-peak hours, people may pursue the experience of taking the elevator more. Therefore, for non-peak hours, the weight of the elevator time can be appropriately increased.

[0073] Setting different objective function weights for different candidate request categories can make the model more in line with the specific needs and priorities of each candidate request category in the actual application scenario, thereby improving the accuracy of prediction and the efficiency of scheduling. For example, during the morning peak period, by increasing the weight of the waiting time for the elevator, the model can more effectively reduce the waiting time for the elevator and quickly respond to the passenger flow demand during the peak period; and during the non-peak period, increasing the weight of the elevator ride time can better optimize the passenger's elevator experience. This differentiated weight setting enables the model to flexibly adapt to the demand characteristics of different elevator call request categories and achieve more refined and personalized elevator scheduling management.

[0074] In some other optional technical solutions of this embodiment, before executing step 102 to determine the target request category corresponding to the elevator request from multiple candidate request categories based on the elevator call data associated with the elevator call request, the multiple candidate request categories can be determined by executing the following steps: obtaining multiple historical elevator call data of the elevator group, the multiple historical elevator call data corresponding one by one to the multiple historical elevator call requests initiated to the elevator group, and dividing the multiple historical elevator call requests into a preset number of candidate request categories according to the similarity between each two historical elevator call data in the multiple historical elevator call data. By collecting the elevator call data of the historical elevator call requests in the elevator group, the historical elevator call requests can be clustered into a specified number of categories based on the similarity between the elevator call data, and by specifying the number of categories, the classification results can meet the required discrimination. Clustering is used to divide an unlabeled data set into classes consisting of multiple similar samples. Optionally, the clustering method can choose to use a hierarchical clustering method.

[0075] The historical call data may be a vector including at least one dimension of data, and the similarity between any two historical call data may be measured by a distance between the two vectors, a cosine similarity, a correlation coefficient, or other measurement values. Further, the above-mentioned step of dividing the multiple historical call requests into a preset number of candidate request categories according to the similarity between each two historical call data in the multiple historical call data may specifically include the following steps: for each historical call request, constructing a historical call feature vector of the historical call request according to the corresponding historical call data; and clustering the multiple historical call requests into a preset number of candidate request categories according to the distance between any two historical call feature vectors. It can be understood that in the constructed historical call feature vector, the data of each dimension may include the original data in the historical call data, or may include the processed data obtained by calculating and processing the historical call data. For any two historical call requests, the similarity is expressed by the vector distance between the historical call data, which can intuitively quantify the similarity between any two historical call requests, and provide a clear mathematical basis for cluster analysis. Classification by clustering method helps to discover the association between historical call data, and then naturally group the historical call requests.

[0076] In some other optional technical solutions of this embodiment, the above step 102 determines the target request category corresponding to the elevator call request from multiple candidate request categories according to the elevator call data associated with the elevator call request, which may specifically include executing the following steps:

[0077] Obtaining reference call data for each candidate request category; the reference call data for any candidate request category is used as a reference sample of call data associated with a call request belonging to the candidate request category;

[0078] Calculate the similarity between each reference elevator call data and the elevator call data;

[0079] The candidate request category corresponding to the benchmark elevator call data with the highest similarity is determined to obtain the target request category.

[0080] For each candidate request category, elevator call data used as a benchmark in each candidate request category can be provided in advance. In this way, by calculating the similarity between the benchmark elevator call data and the elevator call data of the current elevator call request, it can be determined which candidate request category's benchmark the current elevator request is more similar to, and the candidate request category with the highest similarity is selected as the target request category corresponding to the current elevator call request.

[0081] Optionally, each candidate request category can be a classification determined based on experience or manually, and accordingly, the benchmark call data of each candidate request category can also be manually assigned based on experience; or, the benchmark call data can be obtained by selecting the data with the most call data repetitions or the data located at the median / mean from the historical call data belonging to each candidate request category. Optionally, before executing the above step of obtaining the benchmark call data of each candidate request category, the historical call data corresponding to the historical call requests belonging to each candidate request category can be obtained, and the statistical center sample of the historical call data corresponding to each candidate request category can be calculated to obtain the benchmark call data of each candidate request category. The statistical center sample can be the mean, median, mode, centroid, etc. In the case where the historical call data includes more than two data, one historical call data can be regarded as a vector including more than two dimensions, and the statistical center samples of multiple historical call data can be calculated by calculating the geometric center of multiple vectors. By using the statistical center samples of the historical elevator call data corresponding to each candidate request category as the benchmark elevator call data of each candidate request category, the typical characteristics of each candidate request category can be captured, which helps to improve the accuracy of classifying elevator call requests.

[0082] Figure 2The present invention provides a flowchart of a method for dispatching an elevator group in a specific application example. In the application scenario of the example, the elevator system supports group control of the destination layer, and the destination layer information of the passenger can be obtained by a calling device outside the elevator before the passenger takes the elevator. At present, the dispatching method for the elevator group includes: a control algorithm based on an expert system, a control algorithm based on fuzzy logic, and a control algorithm based on a neural network. Among them, the control algorithm based on the expert system and the control algorithm based on fuzzy logic, the rules or weights are only determined by experts based on experience. Due to the different demand preferences of passengers in different time periods, for example, during the morning rush hour, passengers care less about the crowdedness of the elevator and are more concerned about being able to take the elevator, while during the rush hour after get off work, the tolerance of passengers for crowdedness decreases. If a completely consistent allocation algorithm is used for the above two time periods, it may cause low satisfaction of some passengers. Therefore, it is unreasonable to adopt an allocation algorithm with the same weight for different building configurations and different time periods for fixed parameters.

[0083] refer to Figure 2 The technical solution of the elevator group scheduling method provided in this example can be summarized as follows: after preprocessing the historical elevator call data, it is divided into different subsets by a hierarchical clustering method, and the historical elevator call data of each subset is used to train the fuzzy neural network model to obtain a fuzzy neural network model suitable for different subsets. The independent variables of the model take into account many factors such as the waiting time and the riding time of passengers in different states, such as new elevator call passengers, waiting passengers, and passengers on the elevator. The model is used to predict the waiting time and the riding time of passengers in different states taking each elevator in the elevator group. When predicting the current elevator call request, the classification of the elevator call request of the newly added elevator call passenger is determined according to the distance between the elevator call data of the newly added elevator call passenger and the various request categories, and the corresponding neural network model is selected to obtain the operating status data of each elevator, and a feature vector for input to the neural network model is generated for each elevator. After using the neural network model to calculate the prediction result for each elevator, the elevator with the smallest prediction value is selected and assigned to the current newly added elevator call request.

[0084] Specifically, refer to Figure 2 The elevator group scheduling method provided in this example includes the following steps:

[0085] Step 201, through the data recording function of the group controller set on each service floor of the elevator system, the elevator call data, operation status data, actual waiting time and actual riding time (actual execution time) at the installation site of the elevator system are collected, and the above data can be processed and saved through the cloud and / or local devices. The elevator call data may include the calling floor, the destination floor, and the elevator call time. The elevator operation status data may include the current floor of the assigned elevator, the next stop floor, the running direction, the door status, the elevator call information outside the elevator, the elevator call information inside the elevator, the elevator call information to be assigned (i.e., the elevator call request information of the elevator that has not yet been assigned), etc. The actual waiting time and the actual riding time are the information monitored by the elevator that actually responds to the elevator call request in the elevator group after operation.

[0086] Step 202, pre-process the above historical elevator call data to construct a feature vector based on which the elevator call requests are classified. The feature vector based on which the classification is based may include variables of the following dimensions: running direction, elevator call type (entry type, exit type, inter-floor type), day of the week, etc.

[0087] In one example, construct a data set S = {X ij},1≤i≤N,1≤j≤K+3, each historical elevator call request is regarded as a sample X i , the data set includes N samples, each sample includes K+3 variables (or data dimensions), where the first K variables in each sample are the original data collected, and the (K+1)th to (K+3)th variables are data generated based on the original data. For example, the direction X of the elevator is determined based on the calling floor and the destination floor. i,K+1 , determine the day of the week X based on the time i,K+2 , determine the call type X according to the calling floor and the destination floor i,K+3 Belongs to outgoing type, incoming type or inter-layer type. If it is an upward trip from the entrance layer, it is incoming type data. If the destination layer is a downward trip from the entrance layer, it is outgoing data. The others are inter-layer type data.

[0088] Step 203: cluster the above feature vectors using a hierarchical clustering method.

[0089] Define two samples and The distance between them is:

[0090]

[0091] The class distance between two classes S1 and S2 is defined as the maximum distance between the sample points in the two classes, that is, the class distance between class S1 and class S2 is:

[0092]

[0093] In the initial state, each sample point is set as a class, and the two classes with the smallest class distance are merged in turn until the final number of classes is consistent with the set value. The final division is specified to be 4 classes, and 4 data subsets are obtained. By performing descriptive analysis on the 4 data subsets, the characteristics corresponding to each group of data can be preliminarily analyzed according to the proportion of elevator call types, etc.

[0094] Step 204, use 4 groups of data subsets to perform model training respectively, and the models trained in each subset adopt fuzzy neural network. Construct independent variables and dependent variables for each data sample in each class to obtain training data for the training model. The independent variables are constructed based on the operating status data of the historical response elevator at the time of initiation of the elevator call request. Specifically, according to the operating status data of the allocated elevator, the theoretical waiting time Z1 and the elevator riding time Z2 of the current elevator-calling passenger, the waiting time increment Z3 and the elevator riding time increment Z4 of the elevator-calling crowd, the elevator riding time increment Z5 of the current elevator crowd, and the time increment Z6 of a single round trip of the elevator are calculated to obtain 6 independent variables. An example fuzzy neural network structure is shown in Figure 3 As shown, there are 5 layers in total. The first layer is the input layer, which inputs the independent variables Z1, Z2, Z3, Z4, Z5, and Z6. The second layer is the membership function layer. Specifically, the Gaussian function can be selected as the membership function, u ii Represents a membership function:

[0095]

[0096] where u ij is the output value of the membership function layer node, z ij is the input value, is the parameter of the membership function, and m is the number of fuzzy levels. The third layer is the rule layer, a j represents the jth rule, namely:

[0097] a j =Πu ij

[0098] The fourth layer is the normalization layer, which outputs the value a for the rule j Normalize, that is:

[0099]

[0100] The fifth layer is the output layer, which performs weighted output of the normalized values, namely:

[0101]

[0102] The dependent variable is constructed based on the weighted sum of the actual waiting time and the time to take the elevator. The dependent variable Y = W1 × the actual waiting time of the passenger + W2 × the actual time to take the elevator. For data sets that tend to be in the morning peak mode, the actual waiting time in the dependent variable is relatively large. For data sets that tend to be in the off-get off work peak mode, the actual time to take the elevator in the dependent variable is increased. For other modes, the proportion of waiting time and time to take the elevator is consistent. When training the model, the difference between the model output y and the actual constructed dependent variable Y can be used as feedback to adjust the parameters within the model.

[0103] Step 205: deploy the trained model to the elevator system.

[0104] Step 206: receiving a new elevator call request.

[0105] Step 207, calculating the similarity between the elevator call data of the newly added elevator call request and each data subset, and selecting the model trained by the subset with the largest similarity.

[0106] Calculate the data set S separately k |k=1..4 class centers are denoted as For newly added elevator call request X p , calculate X p With class center The distance d(X p ,S k ), select the data with the smallest distance Corresponding model FNN k .

[0107] Step 208, for each elevator in the elevator group, according to the current running status data of the elevator, the theoretical values ​​of the durations such as Z1 to Z6 are calculated as independent variables, substituted into the model for prediction, and the predicted values ​​of each elevator are predicted respectively.

[0108] Step 209, allocate elevators according to the predicted values, and send dispatch instructions to the corresponding elevators to respond to the elevator call request. The elevator with the smallest predicted value is allocated to the newly added elevator call request, or the predicted value is negated to obtain the matching degree between the elevator and the newly added elevator call request, and the elevator with the largest matching degree is selected to be allocated to the elevator call request.

[0109] In other application scenarios, for elevator systems that do not support destination floor group control and only support ordinary group control, since the elevator call data of the elevator call request lacks information about new passengers calling the elevator and the destination floor information of the passengers waiting for the elevator, the independent variables can be reduced. For example, the independent variables can include: the waiting time of new passengers calling the elevator, the incremental waiting time of the people waiting for the elevator at the calling floor, the incremental riding time of passengers in the elevator car, and the incremental time of a single round trip operation of the elevator, which can then be substituted into the model for training and prediction.

[0110] In different data mode scenarios, using specific models for prediction can meet passenger needs more specifically and improve passenger satisfaction. Fuzzy neural networks combine the advantages of fuzzy logic and neural networks, can handle uncertainty and nonlinear problems, and have adaptive learning capabilities. Neural network models can learn the laws of historical data, introduce fuzzy rules, and mine fuzzy rules from data, making the model interpretable. Models trained with the same type of data can predict the best elevator for new passengers, taking into account the overall efficiency of new passengers and other passengers.

[0111] The present invention also provides an embodiment of a dispatching device for an elevator group, which includes an execution module that can be used to execute each step in the dispatching method for an elevator group provided in the embodiment of the present invention. For details, please refer to the relevant description in the embodiment of the dispatching method for the above-mentioned elevator group, which will not be repeated here.

[0112] Figure 4 The block diagram of an electronic device used to implement the elevator group scheduling method provided in the embodiment of the present application. Figure 4 As shown, the electronic device includes: a memory 401 and a processor 402, and the memory 401 stores a computer program that can be run on the processor 402. When the processor 402 executes the computer program, the method in the above embodiment is implemented. The number of the memory 401 and the processor 402 can be one or more. In a specific implementation, the electronic device may also include a communication interface 403 for communicating with external devices and performing data exchange transmission.

[0113] In specific implementation, if the memory 401, the processor 402 and the communication interface 403 are implemented independently, the memory 401, the processor 402 and the communication interface 403 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0114] Optionally, in a specific implementation, if the memory 401, the processor 402 and the communication interface 403 are integrated on a chip, the memory 401, the processor 402 and the communication interface 403 can communicate with each other through an internal interface.

[0115] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein the program is executed by a processor to implement the elevator group scheduling method provided in any embodiment of the present invention.

[0116] An embodiment of the present invention further provides a computer program product, including a computer program, wherein the computer program is executed by a processor to implement the elevator group scheduling method provided in any embodiment of the present invention. Similar to a computer-readable storage medium, the computer program can also be executed by an electronic device or an electronic device that communicates with the electronic device, which will not be described in detail.

[0117] An embodiment of the present invention further provides an elevator system, comprising a plurality of elevators, and also comprising an electronic device provided by an embodiment of the present invention, wherein the electronic device is used to execute the elevator group scheduling method provided by an embodiment of the present invention.

[0118] The above descriptions are only optional embodiments of the present invention, and are not intended to limit the patent scope of the present invention. All modifications, equivalent substitutions, improvements, etc. made using the contents of the present invention specification and drawings under the concept of the present invention, or directly / indirectly applied in other related technical fields, should be included in the patent protection scope of the present invention.

Claims

1. A dispatching method for an elevator group, characterized in that: The method comprises: An elevator call request initiated to an elevator group is detected; the elevator group includes a plurality of elevators; determining, according to the elevator call data associated with the elevator call request, a target request category corresponding to the elevator call request from a plurality of candidate request categories; Determining a target model corresponding to the target request category among multiple candidate models; Based on the operation status data of each elevator, using the target model to calculate the predicted execution time of each elevator in response to the elevator call request; The elevator group is scheduled according to the predicted execution time of each elevator.

2. The method according to claim 1, characterized in that The predicting of the predicted execution time of each elevator in response to the elevator call request based on the operation status data of each elevator using the target model includes: Obtain the operating status data of each elevator; For each elevator, construct a response feature vector of the elevator according to the elevator call data and the operation status data; the response feature vector is used to represent the theoretical execution time of the elevator in response to the elevator call request; For each elevator, the response feature vector of the elevator is input into the target model to obtain the predicted execution time.

3. The method according to claim 1, characterized in that The multiple candidate models correspond one-to-one to the multiple candidate request categories, and each of the candidate models is trained using the associated data of historical elevator call requests belonging to the corresponding candidate request category. The associated data of the historical elevator call requests include one or more of the following data obtained according to the historical elevator call requests: historical elevator call data, operating status data of the responding elevator, and actual execution time of the responding elevator; the responding elevator is the elevator that actually responds to the historical elevator call request.

4. The method according to claim 3, characterized in that The weight of each variable in the objective function of the multiple candidate models is independently set, and the objective function includes at least one variable, and the at least one variable includes at least one variable related to the actual execution time of the historical elevator call request.

5. The method according to claim 1, characterized in that Before determining, based on the elevator call data associated with the elevator call request, a target request category corresponding to the elevator call request from a plurality of candidate request categories, the method further includes: Acquire a plurality of historical elevator call data of the elevator group, wherein the plurality of historical elevator call data correspond one by one to a plurality of historical elevator call requests initiated to the elevator group; According to the similarity between every two of the plurality of historical elevator call data, the plurality of historical elevator call requests are divided into a preset number of the candidate request categories.

6. The method according to claim 5, characterized in that The dividing the plurality of historical elevator call requests into a preset number of candidate request categories according to the similarity between every two of the plurality of historical elevator call data comprises: For each of the historical elevator call requests, constructing a historical elevator call feature vector of the historical elevator call request according to the corresponding historical elevator call data; The plurality of historical elevator call requests are clustered into a preset number of the candidate request categories according to the distance between any two of the historical elevator call feature vectors.

7. The method according to claim 1, characterized in that The step of determining, based on the elevator call data associated with the elevator call request, a target request category corresponding to the elevator call request from a plurality of candidate request categories comprises: Obtaining reference call data of each candidate request category; using any reference call data of the candidate request category as a reference sample of call data associated with a call request belonging to the candidate request category; calculating the similarity between each reference call data and the call data; The candidate request category corresponding to the reference elevator call data with the highest similarity is determined to obtain the target request category.

8. The method according to claim 7, characterized in that Before obtaining the benchmark elevator call data of each candidate request category, the method further includes: Acquiring historical call data corresponding to historical call requests belonging to each of the candidate request categories; The statistical center sample of the historical elevator call data corresponding to each of the candidate request categories is calculated to obtain the benchmark elevator call data of each of the candidate request categories.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: The processor executes the program to implement the method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the method according to any one of claims 1 to 8.

11. An elevator system, characterized in that: The device comprises a plurality of elevators and also comprises the electronic device according to claim 9, wherein the electronic device is used to execute the method according to any one of claims 1-8.

Images