Elevator control system and method based on intelligent building

Abstract

The invention discloses an elevator control system and method based on an intelligent building, relates to the technical field of elevator control, and solves the problems that people with low priorities may wait for a long time or cannot wait for an elevator because only user priorities are considered to distribute elevators; the number of people waiting for the elevator and the distance between the elevator request floor and the current floor where the elevator is located are not considered, so that the waiting time of some users is long, and the user experience feeling is poor. When the elevator is in a running state, generating an emergency degree coefficient and a distance coefficient according to elevator waiting data, elevator running data and elevator request floor data, and generating a basic priority coefficient according to the emergency degree coefficient and the distance coefficient; generating a final priority coefficient according to the elevator request floor data and the basic priority coefficient; the elevator is dispatched according to the final priority coefficient, the number of waiting personnel, the identity of the waiting personnel and the emergency state of the elevator are considered in the elevator stopping influence factors, the situation that the elevator cannot appear is solved through the final priority coefficient, and the experience feeling of a user is improved.

Description

Technical Field

[0001] The present application belongs to the technical field of elevator control, and specifically is an elevator control system and method based on intelligent buildings. Background Art

[0002] With the acceleration of urbanization, high-rise buildings are increasing day by day. As a vertical transportation tool, elevators play an important role in people's daily lives. However, traditional elevator control systems have problems such as low operating efficiency and long waiting time for passengers. In addition, with the rapid development of technologies such as the Internet of Things and big data, it is possible to upgrade the intelligent elevator control system.

[0003] The prior art (invention patent application with publication number CN115321288A) discloses an elevator scheduling method and system based on artificial intelligence, the method comprising obtaining the current time, wherein the current time period includes working time and rest time; when the current time is working time, executing a first control strategy to ensure the efficiency of the elevator; when the current time is rest time, executing a second control strategy to ensure the efficiency of the elevator.

[0004] The above-mentioned elevator scheduling method based on artificial intelligence adopts different scheduling methods by judging whether the current time is in working hours, and distinguishes the priorities of the people taking the elevator, and determines whether the elevator reaches the requested floor by priority; only considering the user priority to allocate elevators may cause people with low priority to wait for a long time, or even not get the elevator; not considering the number of people waiting for the elevator and the distance between the elevator request floor and the floor where the current elevator is located, causing some users to wait for a long time, resulting in a poor user experience, so the elevator control system and method still need to be further improved. Summary of the invention

[0005] The present application aims to solve at least one of the technical problems existing in the prior art; to this end, the present application proposes an elevator control system and method based on an intelligent building, which is used to solve the technical problems that only considering the priority of users to allocate elevators may cause people with low priority to wait for a long time or even not get the elevator; and not considering the number of people waiting for the elevator and the distance between the floor requested for the elevator and the floor where the current elevator is located, causing some users to wait for a long time, resulting in a poor user experience.

[0006] To achieve the above-mentioned purpose, the first aspect of the present application provides an elevator control system based on an intelligent building, comprising: a data acquisition module, a data analysis module and a database;

[0007] The data acquisition module is used to acquire elevator request floor data, elevator waiting data and elevator operation data in real time through data acquisition equipment;

[0008] The data analysis module: when the elevator is in operation: obtains elevator waiting data, elevator operation data and elevator request floor data, generates an urgency coefficient according to the elevator waiting data; generates a distance coefficient according to the elevator operation data and the elevator request floor data; generates a basic priority coefficient according to the urgency coefficient and the distance coefficient; generates a final priority coefficient for each request floor according to the request time in the elevator request floor data and the corresponding basic priority coefficient; performs elevator dispatch according to the elevator operation data and the final priority coefficient;

[0009] Otherwise, the current recorded time is input into the elevator request floor prediction model to output the future elevator predicted request floor number and its corresponding predicted request floor number; according to the predicted request floor number and its corresponding predicted request floor number, the idle elevator is dispatched to the corresponding request floor to wait.

[0010] The present application obtains elevator request floor data, elevator waiting data and elevator operation data; when the elevator is in operation: generates an urgency coefficient according to the elevator waiting data; generates a distance coefficient according to the elevator operation data and the elevator request floor data; generates a basic priority coefficient according to the urgency coefficient and the distance coefficient; generates a final priority coefficient for each request floor according to the request time and the corresponding basic priority coefficient in the elevator request floor data; performs elevator scheduling according to the elevator operation data and the final priority coefficient; otherwise, inputs the current recorded time into the elevator request floor prediction model to output the future elevator predicted request floor number and its corresponding predicted request floor number; schedules the idle elevator to the corresponding request floor for waiting according to the predicted request floor number and its corresponding predicted request floor number, considers the number of people waiting for the elevator, the identity of the waiting people and their anxiety in the factors affecting the elevator stop, so that the request floors with a large number of people waiting, a higher priority of the identity of the waiting people and anxious waiting people can be given priority in elevator stops, and solves the problem of users not being able to wait for the elevator through the final priority coefficient, thereby improving the user experience.

[0011] Furthermore, generating an urgency coefficient according to the elevator waiting data includes:

[0012] Extracting the waiting area image from the elevator waiting data, inputting the waiting area image into a personnel recognition model to obtain the priority coefficients XDij and anxiety coefficients JDij corresponding to a number of waiting personnel; the personnel recognition model is obtained by training an artificial intelligence model;

[0013] Extract the number of waiting personnel RSi of several requested floors in the elevator waiting data; calculate the coefficient of the number of waiting personnel LDXi of the requested floor numbered i by the formula LDXi=RSi / ∑(RSi);

[0014] The urgency coefficient JCXi of each requested floor is calculated by the formula;

[0015] JCXi=β×LDXi×∑(XDij×JDij); wherein j represents the number of the waiting person, and j=1, 2,…, mi, mi is the total number of waiting people corresponding to the requested floor numbered i, β is the adjustment factor, β∈(0,1).

[0016] Furthermore, the personnel recognition model is obtained by training an artificial intelligence model, including:

[0017] Obtaining a number of waiting area images and their corresponding waiting persons, as well as the priority coefficient and anxiety coefficient of each waiting person through a database; the priority coefficients include a management personnel priority coefficient, a registered personnel priority coefficient, and a non-registered personnel priority coefficient, the management personnel priority coefficient is greater than the registered personnel priority coefficient, and the registered personnel priority coefficient is greater than the non-registered personnel priority coefficient;

[0018] Integrate a number of waiting area images and their corresponding number of priority coefficients and anxiety coefficients into a number of training data, test data and verification data;

[0019] Select the artificial intelligence network model as the basic model;

[0020] Train the basic model with the training data, and adjust the learning rate or other hyperparameters on the validation data to obtain the pre-trained model;

[0021] By verifying the pre-trained model on the test data, we finally obtained a personnel recognition model whose input is the waiting area image and whose output is the priority coefficient and anxiety coefficient of each waiting person in the image.

[0022] Further, generating the distance coefficient according to the elevator operation data and the elevator requested floor data includes:

[0023] Obtain the elevator floor number YD in the elevator operation data, the requested floor number QLi in the elevator requested floor data, and the total number of floors LC of the building;

[0024] Calculate the distance coefficient JXi of the requested floor through the formula;

[0025] JXi=1-|YD-QLi| / LC; wherein i represents the number of the requested floor.

[0026] Furthermore, generating a basic priority coefficient according to the urgency coefficient and the distance coefficient includes:

[0027] Extract the elevator request direction in the elevator request floor data and the running direction in the elevator running data; when the elevator request direction is the same as the elevator running direction, set the direction coefficient QFXi to K1; otherwise, set the direction coefficient QFXi to K2; 1≥K1>K2>0;

[0028] Get the urgency coefficient JCXi and distance coefficient JXi of each requested floor;

[0029] Calculate the basic priority coefficient QLJXi of each requested floor number through the formula;

[0030] QLJXi=k×QFXi×exp(α1×JCXi+α2×JXi); wherein k is the adjustment factor, α1 and α2 are weight coefficients, α1 and α2∈(0,1); i represents the number of the requested floor.

[0031] Further, the generating of the final priority coefficient of each requested floor according to the request time and the corresponding basic priority coefficient in the elevator requested floor data includes:

[0032] Get the request time QLQSi and the current time DS of each requested floor whose request was not received;

[0033] Calculate the absolute value QCJi of the difference between the current time and the requested time;

[0034] Determine whether the absolute value of the difference is greater than the request time threshold QSY;

[0035] Yes, the additional coefficient QLEXi for each requested floor is calculated by the formula;

[0036] QLEXi=1 / (1+exp(-(QCJi-QSY)))-b;b is the deviation, and b∈(0,1);

[0037] No, do not perform the operation;

[0038] Get the basic priority coefficient QLJXi of each requested floor number;

[0039] The final priority coefficient QLZXi of each requested floor number is calculated by the formula;

[0040] QLZXi=c1×QLJXi+c2×QLEXi; wherein, i represents the number of the requested floor, c1 and c2 are adjustment factors, c1>0, c2>0.

[0041] Furthermore, the elevator scheduling according to the elevator operation data and the final priority coefficient includes:

[0042] Obtain the running direction, floor number, load weight, and elevator request direction and final priority coefficient of each requested floor in the elevator running data;

[0043] Sort the elevator request directions that are consistent with the running direction from large to small according to the final priority coefficient;

[0044] Dispatch the elevator to the requested floor corresponding to the requested floor number with the largest final priority coefficient in the running direction;

[0045] Determine whether the load weight is greater than the standard load weight;

[0046] Yes, sound the alarm;

[0047] No, when the absolute value of the difference between the elevator load and the standard load weight is less than the standard single-person weight; the elevator does not stop at the requested floor corresponding to other requested floor numbers in the running direction;

[0048] Otherwise, according to each final priority coefficient in the running direction, the requested floor number corresponding to the maximum final priority coefficient is selected for scheduling;

[0049] When the load weight in the elevator operation data is 0, the operation direction in the elevator operation data is adjusted, and the elevator is dispatched to the requested floor corresponding to the requested floor number with the largest final priority coefficient in the same operation direction.

[0050] Furthermore, the elevator requested floor prediction model is constructed by a time series prediction model, including:

[0051] Obtaining a number of historical building data and their corresponding recording time through a database; the historical building data includes historical requested floor numbers and their corresponding historical requested floor times; the recording time is whether it is a working day and various time periods within each day;

[0052] Performing data preprocessing on a number of recorded times and their corresponding historical building data to obtain a number of recorded times and their corresponding historical building data after preprocessing;

[0053] Integrate the historical building data and recording time with the same recording time into a time feature vector;

[0054] Integrate each time feature vector into several prediction sequences according to the order of recording time;

[0055] Divide several prediction sequences into training set, test set and validation set in chronological order;

[0056] Select the LSTM model as the base model;

[0057] Train the basic model with the training set, and adjust the learning rate or other hyperparameters on the validation set to obtain the pre-trained model;

[0058] By verifying the pre-trained model on the test set, we finally get the elevator request floor prediction model which inputs the record time and outputs the predicted requested floor number and its predicted requested floor number.

[0059] Furthermore, the data preprocessing includes: data cleaning and data normalization.

[0060] Further, dispatching the idle elevator to the corresponding requested floor for waiting according to the predicted requested floor number and its corresponding predicted requested floor number includes:

[0061] Get the predicted requested floor number and its corresponding predicted requested floor frequency;

[0062] The predicted requested floor numbers and their corresponding predicted requested floor times are sorted from largest to smallest according to the predicted requested floor times;

[0063] If there are several identical maximum predicted request floor times, the corresponding predicted request floor numbers are obtained, and the idle elevator is dispatched to the request floor corresponding to the maximum predicted request floor number to wait;

[0064] Otherwise, the idle elevator is dispatched to the requested floor corresponding to the predicted requested floor number with the largest number of predicted requested floors to wait.

[0065] The present application obtains the predicted requested floor number and its corresponding predicted requested floor times, and dispatches the elevator to the requested floor corresponding to the predicted requested floor number with the largest predicted requested floor times. Considering the situation where multiple maximum predicted requested floor times are the same, the elevator is dispatched to the requested floor corresponding to the maximum predicted requested floor number corresponding to multiple maximum predicted request floor times. The elevator is dispatched to the designated requested floor based on the prediction results, thereby reducing the waiting time for users when using the elevator.

[0066] On the other hand, the present application provides an elevator control method based on an intelligent building, comprising the following steps:

[0067] S1: Real-time acquisition of elevator request floor data, elevator waiting data and elevator operation data;

[0068] S2: Extract the elevator status in the elevator operation data and determine whether the elevator status is in operation; if yes, go to S3; if no, go to S6;

[0069] S3: Obtain elevator waiting data, elevator operation data and elevator requested floor data, generate an urgency coefficient according to the elevator waiting data; generate a distance coefficient according to the elevator operation data and the elevator requested floor data;

[0070] S4: Generate a basic priority coefficient according to the urgency coefficient and the distance coefficient; generate a final priority coefficient for each requested floor according to the request time in the elevator requested floor data and the corresponding basic priority coefficient;

[0071] S5: Elevator dispatching based on elevator operation data and final priority coefficient;

[0072] S6: input the current recording time into the elevator request floor prediction model to output the future elevator predicted request floor number and its corresponding predicted request floor number;

[0073] S7: Dispatch the idle elevator to the corresponding requested floor for waiting according to the predicted requested floor number and its corresponding predicted requested floor number.

[0074] Compared with the prior art, the beneficial effects of this application are:

[0075] 1. The present application obtains elevator request floor data, elevator waiting data and elevator operation data; when the elevator is in operation: generates an urgency coefficient according to the elevator waiting data; generates a distance coefficient according to the elevator operation data and the elevator request floor data; generates a basic priority coefficient according to the urgency coefficient and the distance coefficient; generates a final priority coefficient for each request floor according to the request time and the corresponding basic priority coefficient in the elevator request floor data; performs elevator scheduling according to the elevator operation data and the final priority coefficient; otherwise, inputs the current recorded time into the elevator request floor prediction model to output the future elevator predicted request floor number and its corresponding predicted request floor number; schedules the idle elevator to the corresponding request floor for waiting according to the predicted request floor number and its corresponding predicted request floor number, considers the number of people waiting for the elevator, the identity of the waiting people and the anxiety of the waiting people in the factors affecting the elevator stop, so that the request floor with a large number of people waiting, a high priority of the identity of the waiting people and an anxious waiting people can be given priority in the elevator stop, solves the problem of users not being able to wait for the elevator through the final priority coefficient, and improves the user experience.

[0076] 2. The present application obtains the predicted requested floor number and its corresponding predicted requested floor times, and dispatches the elevator to the requested floor corresponding to the predicted requested floor number with the largest predicted requested floor times. Considering the situation where multiple maximum predicted requested floor times are the same, the elevator is dispatched to the requested floor corresponding to the maximum predicted requested floor number corresponding to multiple maximum predicted requested floor times. The elevator is dispatched to the designated requested floor based on the prediction results, thereby reducing the waiting time for users when using the elevator. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] In order to more clearly illustrate the embodiments of the present application 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 application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0078] Figure 1 This is a schematic diagram of the principle of an elevator control system based on a smart building in this application;

[0079] Figure 2 The elevator dispatching flow chart for the predicted requested floor of this application;

[0080] Figure 3 This is a flow chart of an elevator control method based on intelligent buildings in this application. DETAILED DESCRIPTION

[0081] The technical solution of the present application will be described clearly and completely below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0082] See also Figure 1 , the first aspect of the present application provides an elevator control system based on a smart building, including: a data acquisition module, a data analysis module and a database;

[0083] Data acquisition module: The data acquisition device is used to obtain the elevator request floor data, elevator waiting data and elevator operation data in real time; the data acquisition device includes a high-definition camera and corresponding sensors; the elevator request floor data includes the request time, the elevator request direction and the elevator request signal; the elevator request time refers to the time when the elevator button is pressed for the first time, the elevator request direction includes up or down, and the elevator request signal refers to the signal sent when the elevator button is pressed; the elevator waiting data includes the number of waiting people and the image of the waiting area; the elevator operation data includes the running direction, elevator status, load weight and floor number, and the elevator status includes the running status and idle status;

[0084] Data analysis module: when the elevator is in operation: obtain the elevator waiting data, elevator operation data and elevator request floor data, and generate an emergency coefficient according to the elevator waiting data; the emergency coefficient is the importance of each floor setting considering the situation of waiting personnel on each floor; generate a distance coefficient according to the elevator operation data and the elevator request floor data; the distance coefficient refers to the influence of the distance between the current elevator and the floor that issued the elevator request; generate a basic priority coefficient according to the emergency coefficient and the distance coefficient; the basic priority coefficient is the importance of each floor setting considering the situation of waiting personnel on each floor and the distance between the current elevator and each floor; generate the final priority coefficient of each requested floor according to the request time in the elevator request floor data and the corresponding basic priority coefficient; the final priority coefficient is the importance of each floor setting considering the situation of waiting personnel on each floor and the waiting time of waiting personnel on each floor, as well as the distance between the current elevator and each floor; perform elevator scheduling according to the elevator operation data and the final priority coefficient;

[0085] Otherwise, the current recorded time is input into the elevator request floor prediction model to output the future elevator predicted request floor number and its corresponding predicted request floor number; the recorded time includes the time period of the current time and whether it is a weekday, etc.; according to the predicted request floor number and its corresponding predicted request floor number, the idle elevator is dispatched to the corresponding request floor to wait.

[0086] This embodiment obtains elevator request floor data, elevator waiting data and elevator operation data; when the elevator is in operation: an urgency coefficient is generated according to the elevator waiting data; a distance coefficient is generated according to the elevator operation data and the elevator request floor data; a basic priority coefficient is generated according to the urgency coefficient and the distance coefficient; a final priority coefficient of each request floor is generated according to the request time in the elevator request floor data and the corresponding basic priority coefficient; elevator scheduling is performed according to the elevator operation data and the final priority coefficient; otherwise, the current recorded time is input into the elevator request floor prediction model to output the future elevator predicted request floor number and its corresponding predicted request floor number; according to the predicted request floor number and its corresponding predicted request floor number, an idle elevator is scheduled to the corresponding request floor for waiting, and the number of people waiting for the elevator, the identity of the waiting people and the anxiety of the waiting people are considered in the factors affecting the elevator stop, so that the request floors with a large number of waiting people, a higher priority of the identity of the waiting people and the anxious waiting people can be given priority in the elevator stop, and the situation that users cannot wait for the elevator is solved through the final priority coefficient, thereby improving the user experience.

[0087] Generate an urgency factor based on elevator waiting data, including:

[0088] Extracting the waiting area image from the elevator waiting data, inputting the waiting area image into a personnel recognition model to obtain the priority coefficients XDij and anxiety coefficients JDij corresponding to a number of waiting personnel; the personnel recognition model is obtained by training an artificial intelligence model;

[0089] Extract the number of waiting personnel RSi of several request floors in the elevator waiting data; calculate the waiting personnel coefficient LDXi of the request floor numbered i by the formula LDXi=RSi / ∑(RSi); as the number of waiting personnel on the floor increases, the total number of waiting personnel in the entire building increases, and the waiting personnel coefficient of the floor will also increase accordingly; if and only if there is only one floor in the building that sends an elevator request signal, the waiting personnel coefficient of the floor remains unchanged;

[0090] The urgency coefficient JCXi of each requested floor is calculated by the formula;

[0091] JCXi=β×LDXi×∑(XDij×JDij); wherein, j represents the number of the waiting personnel, and j=1, 2, …, mi, mi is the total number of waiting personnel corresponding to the requested floor numbered i; β is an adjustment factor, β∈(0, 1) and β is set according to experience, and the adjustment factor β is set so that the urgency coefficient JCXi∈(0, 1); as the number of waiting personnel in the requested floor increases, the number of high-priority waiting personnel increases, and the anxiety level of the waiting personnel increases, the urgency of using the elevator on this floor increases, and the urgency coefficient of this floor will also increase accordingly.

[0092] Furthermore, the personnel recognition model is obtained by training an artificial intelligence model, including:

[0093] A number of waiting area images and their corresponding waiting persons, as well as the priority coefficient and anxiety coefficient of each waiting person are obtained through a database; the priority coefficient includes a management personnel priority coefficient, a registered personnel priority coefficient, and an unregistered personnel priority coefficient, the management personnel priority coefficient is greater than the registered personnel priority coefficient, the registered personnel priority coefficient is greater than the unregistered personnel priority coefficient, the management personnel priority coefficient is the priority coefficient corresponding to the personnel in the management personnel database, the registered personnel priority coefficient is the priority coefficient corresponding to the personnel in the registered personnel database, and the unregistered personnel is the priority coefficient corresponding to the personnel who are not in the management personnel database and the registered personnel database; the anxiety coefficient is the anxiety coefficient set by the expression of each person;

[0094] Integrate a number of waiting area images and their corresponding priority coefficients and anxiety coefficients into a number of training data, test data and verification data; the training data, test data and verification data are divided according to 7:2:1;

[0095] Select the artificial intelligence network model as the basic model;

[0096] Train the basic model with the training data, and adjust the learning rate or other hyperparameters on the validation data to obtain the pre-trained model;

[0097] By verifying the pre-trained model on the test data, we finally obtained a personnel recognition model whose input is the waiting area image and whose output is the priority coefficient and anxiety coefficient of each waiting person in the image.

[0098] Generate distance coefficients based on elevator operation data and elevator request floor data, including:

[0099] Obtain the elevator floor number YD in the elevator operation data, the requested floor number QLi in the elevator requested floor data, and the total number of floors LC of the building;

[0100] Calculate the distance coefficient JXi of the requested floor through the formula;

[0101] JXi=1-|YD-QLi| / LC; once the total number of floors of a building is determined, it will basically not change. As the distance between the requested floor and the floor where the elevator is located increases, the distance coefficient decreases accordingly; where i represents the number of the requested floor.

[0102] Generate basic priority factors based on urgency factor and distance factor, including:

[0103] Extract the elevator request direction in the elevator request floor data and the running direction in the elevator running data; when the elevator request direction is the same as the elevator running direction, set the direction coefficient QFXi to K1; otherwise, set the direction coefficient QFXi to K2; where 1≥K1>K2>0; and K1 and K2 are set based on experience;

[0104] Get the urgency coefficient JCXi and distance coefficient JXi of each requested floor;

[0105] Calculate the basic priority coefficient QLJXi of each requested floor number through the formula;

[0106] QLJXi=k×QFXi×exp(α1×JCXi+α2×JXi); as the number of waiting people on the requested floor increases, the number of high-priority people among the waiting people increases, and the waiting people become anxious, the floor urgently needs to use the elevator, which increases the urgency coefficient of the requested floor, and then the basic priority coefficient of the requested floor increases accordingly; as the distance between the requested floor and the floor where the elevator is located decreases, the distance coefficient increases, the elevator will give priority to the nearest requested floor, and then the basic priority coefficient of the requested floor will also increase accordingly; k is the adjustment factor, 1>k>0, and k is set according to experience; the adjustment factor k is set so that the basic priority coefficient QLJXi∈(0,1); α1 and α2 are weight coefficients, α1 and α2∈(0,1), and α1 and α2 are set according to experience; i represents the number of the requested floor.

[0107] The final priority coefficient of each requested floor is generated according to the request time and the corresponding basic priority coefficient in the elevator request floor data, including:

[0108] Get the request time QLQSi and the current time DS of each requested floor whose request was not received;

[0109] Calculate the absolute value QCJi of the difference between the current time and the requested time;

[0110] Determine whether the absolute value of the difference is greater than the request time threshold QSY; QSY is set based on experience;

[0111] Yes, the additional coefficient QLEXi for each requested floor is calculated by the formula;

[0112] QLEXi=1 / (1+exp(-(QCJi-QSY)))-b;When the absolute value of the difference is greater than the request time threshold, the formula is used, so (QCJi-QSY)>0. As the current time passes, there is no elevator stopping at the requested floor that issues the elevator request signal. Therefore, it is necessary to add an additional coefficient to the basic priority coefficient of the requested floor to obtain the final priority coefficient. The additional priority coefficient increases with the increase of the absolute value of the difference; b is the deviation, and b∈(0,1), which is set according to the absolute value of the difference and the request time threshold. When the absolute value of the difference is equal to the request time threshold, the additional coefficient should be 0, then b is 0.5;

[0113] No, do not perform the operation;

[0114] Get the basic priority coefficient QLJXi of each requested floor number;

[0115] The final priority coefficient QLZXi of each requested floor number is calculated by the formula;

[0116] QLZXi=c1×QLJXi+c2×QLEXi; the final priority coefficient of the requested floor is determined by its corresponding basic priority coefficient and additional coefficient, and increases with the increase of the basic priority coefficient and the additional priority coefficient; wherein, i represents the number of the requested floor; c1 and c2 are adjustment factors, c1>0, c2>0, and c1 and c2 are set according to experience, and the adjustment factors c1 and c2 are set to make the basic priority coefficient QLJXi and the additional coefficient QLEXi within the same range.

[0117] Elevator dispatching is performed based on elevator operation data and final priority coefficients, including:

[0118] Get the running direction, floor number, load weight, and elevator request direction and final priority coefficient of each requested floor in the elevator running data; the running direction includes up or down; the load weight is the actual weight borne by the elevator. When there are no people or objects in the elevator, the load weight is 0;

[0119] Sort the elevator request directions that are consistent with the running direction from large to small according to the final priority coefficient;

[0120] Dispatch the elevator to the requested floor corresponding to the requested floor number with the largest final priority coefficient in the running direction;

[0121] Determine whether the load weight is greater than the standard load weight; the standard load weight is the maximum weight that the elevator can withstand in a safe state;

[0122] Yes, sound the alarm;

[0123] No, when the absolute value of the difference between the elevator load and the standard load weight is less than the standard single-person weight; the elevator does not stop at the requested floor corresponding to other requested floor numbers in the running direction; the standard single-person weight is the standard weight of an adult;

[0124] Otherwise, according to each final priority coefficient in the running direction, the requested floor number corresponding to the maximum final priority coefficient is selected for scheduling;

[0125] When the load weight in the elevator operation data is 0, the operation direction in the elevator operation data is adjusted, and the elevator is dispatched to the requested floor corresponding to the requested floor number with the largest final priority coefficient in the same operation direction.

[0126] The elevator request floor prediction model is constructed through the time series prediction model, including:

[0127] Obtaining a number of historical building data and their corresponding recording time through the database; the historical building data includes the historical requested floor number and its corresponding historical requested floor number; the recording time is whether it is a working day and the various time periods within each day;

[0128] Performing data preprocessing on a number of recorded times and their corresponding historical building data to obtain a number of recorded times and their corresponding historical building data after preprocessing;

[0129] Integrate the historical building data and recording time with the same recording time into a time feature vector;

[0130] Integrate each time feature vector into several prediction sequences according to the order of recording time;

[0131] Divide several prediction sequences into training set, test set and validation set in chronological order; the ratio between training set, test set and validation set is 7:2:1;

[0132] Select the LSTM model as the base model;

[0133] Train the basic model with the training set, and adjust the learning rate or other hyperparameters on the validation set to obtain the pre-trained model;

[0134] By verifying the pre-trained model on the test set, we finally get the elevator request floor prediction model which inputs the record time and outputs the predicted requested floor number and its predicted requested floor number.

[0135] Data preprocessing includes: data cleaning and data normalization;

[0136] Data cleaning: directly remove the missing values ​​in the acquired historical building data and its corresponding recording time to obtain normal data;

[0137] Normalization operation: Normalize the normal data using the maximum and minimum normalization method to obtain the preprocessed historical building data and its corresponding recording time.

[0138] See also Figure 2 , according to the predicted requested floor number and its corresponding predicted requested floor number, the idle elevator is dispatched to the corresponding requested floor to wait, including:

[0139] Get the predicted requested floor number and its corresponding predicted requested floor times; the predicted requested floor number refers to the floor number of the predicted building that requires an elevator; the predicted requested floor times refers to the elevator request signal issued within the predicted time on the predicted requested floor;

[0140] The predicted requested floor numbers and their corresponding predicted requested floor times are sorted from largest to smallest according to the predicted requested floor times;

[0141] If there are several identical maximum predicted request floor times, the corresponding predicted request floor numbers are obtained, and the idle elevator is dispatched to the request floor corresponding to the maximum predicted request floor number to wait;

[0142] Otherwise, the idle elevator is dispatched to the requested floor corresponding to the predicted requested floor number with the largest number of predicted requested floors to wait.

[0143] This embodiment obtains the predicted request floor number and its corresponding predicted request floor number, and dispatches the elevator to the request floor corresponding to the predicted request floor number with the largest predicted request floor number. Considering the situation where multiple maximum predicted request floor numbers are the same, the elevator is dispatched to the request floor corresponding to the maximum predicted request floor number corresponding to multiple maximum predicted request floor numbers. The elevator is dispatched to the specified request floor based on the prediction result, reducing the waiting time for users when using the elevator.

[0144] See also Figure 3 In another aspect, an embodiment of the present application provides an elevator control method based on an intelligent building, comprising the following steps:

[0145] S1: Real-time acquisition of elevator request floor data, elevator waiting data and elevator operation data;

[0146] S2: Extract the elevator status in the elevator operation data and determine whether the elevator status is in operation; if yes, go to S3; if no, go to S6;

[0147] S3: Obtain elevator waiting data, elevator operation data and elevator requested floor data, generate an urgency coefficient according to the elevator waiting data; generate a distance coefficient according to the elevator operation data and the elevator requested floor data;

[0148] S4: Generate a basic priority coefficient according to the urgency coefficient and the distance coefficient; generate a final priority coefficient for each requested floor according to the request time in the elevator requested floor data and the corresponding basic priority coefficient;

[0149] S5: Elevator dispatching based on elevator operation data and final priority coefficient;

[0150] S6: input the current recording time into the elevator request floor prediction model to output the future elevator predicted request floor number and its corresponding predicted request floor number;

[0151] S7: Dispatch the idle elevator to the corresponding requested floor for waiting according to the predicted requested floor number and its corresponding predicted requested floor number.

[0152] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0153] The working principle of this application is as follows: by obtaining elevator request floor data, elevator waiting data and elevator operation data; when the elevator is in operation: generating an urgency coefficient according to the elevator waiting data; generating a distance coefficient according to the elevator operation data and the elevator request floor data; generating a basic priority coefficient according to the urgency coefficient and the distance coefficient; generating a final priority coefficient for each request floor according to the request time and the corresponding basic priority coefficient in the elevator request floor data; performing elevator scheduling according to the elevator operation data and the final priority coefficient; otherwise, inputting the current recorded time into the elevator request floor prediction model to output the future elevator predicted request floor number and its corresponding predicted request floor number; and The corresponding predicted requested floor times will dispatch idle elevators to the corresponding requested floors for waiting, and the number of people waiting for the elevator and their identities will be taken into account in the factors affecting the elevator stop, so that the requested floors with a large number of waiting people, a higher priority of the waiting people and anxious waiting people can be given priority in elevator stops. The final priority coefficient solves the problem of users not being able to wait for the elevator, improves the user experience, and avoids the situation where only considering user priority to allocate elevators may cause people with low priority to wait for a long time, or even not be able to wait for the elevator; the number of people waiting for the elevator and the distance between the elevator request floor and the floor where the current elevator is located are not considered, which makes some users wait for a long time, resulting in a poor user experience.

[0154] The above embodiments are only used to illustrate the technical method of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, a person of ordinary skill in the art should understand that the technical method of the present application may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present application.

Claims

1. An elevator control system based on intelligent buildings, characterized in that: include: Data collection module, data analysis module and database; The data acquisition module is used to acquire elevator request floor data, elevator waiting data and elevator operation data in real time through data acquisition equipment; The data analysis module: when the elevator is in operation: obtains elevator waiting data, elevator operation data and elevator requested floor data, generates an urgency coefficient according to the elevator waiting data; generates a distance coefficient according to the elevator operation data and the elevator requested floor data; Generate a basic priority coefficient according to the urgency coefficient and the distance coefficient; generate a final priority coefficient for each requested floor according to the request time and the corresponding basic priority coefficient in the elevator request floor data; Elevator dispatching based on elevator operation data and final priority coefficient; Otherwise, the current recorded time is input into the elevator request floor prediction model to output the future elevator predicted request floor number and its corresponding predicted request floor number; according to the predicted request floor number and its corresponding predicted request floor number, the idle elevator is dispatched to the corresponding request floor to wait.

2. The elevator control system based on intelligent building according to claim 1, characterized in that: The generating of the urgency coefficient according to the elevator waiting data comprises: Extracting the waiting area image from the elevator waiting data, inputting the waiting area image into a personnel recognition model to obtain the priority coefficients XDij and anxiety coefficients JDij corresponding to a number of waiting personnel; the personnel recognition model is obtained by training an artificial intelligence model; Extract the number of waiting personnel RSi of several requested floors in the elevator waiting data; calculate the waiting personnel number coefficient LDXi of the requested floor numbered i by the formula LDXi=RSi / ∑(RSi); The urgency coefficient JCXi of each requested floor is calculated by the formula; JCXi=β×LDXi×∑(XDij×JDij); wherein j represents the number of the waiting person, and j=1, 2,…, mi, mi is the total number of waiting people corresponding to the requested floor numbered i, β is the adjustment factor, β∈(0,1).

3. The elevator control system based on intelligent building according to claim 2 is characterized in that: The personnel recognition model is obtained by training an artificial intelligence model, including: Obtaining a number of waiting area images and their corresponding waiting persons, as well as the priority coefficient and anxiety coefficient of each waiting person through a database; the priority coefficients include a management personnel priority coefficient, a registered personnel priority coefficient, and a non-registered personnel priority coefficient, the management personnel priority coefficient is greater than the registered personnel priority coefficient, and the registered personnel priority coefficient is greater than the non-registered personnel priority coefficient; Integrate a number of waiting area images and their corresponding number of priority coefficients and anxiety coefficients into a number of training data, test data and verification data; Select the artificial intelligence network model as the basic model; Train the basic model with the training data, and adjust the learning rate or other hyperparameters on the validation data to obtain the pre-trained model; By verifying the pre-trained model on the test data, we finally obtained a personnel recognition model whose input is the waiting area image and whose output is the priority coefficient and anxiety coefficient of each waiting person in the image.

4. The elevator control system based on intelligent building according to claim 1, characterized in that: The generating of the distance coefficient according to the elevator operation data and the elevator requested floor data comprises: Obtain the elevator floor number YD in the elevator operation data, the requested floor number QLi in the elevator requested floor data, and the total number of floors LC of the building; Calculate the distance coefficient JXi of the requested floor through the formula; JXi=1-|YD-QLi| / LC; wherein i represents the number of the requested floor.

5. The elevator control system based on intelligent building according to claim 1, characterized in that: The generating of the basic priority coefficient according to the urgency coefficient and the distance coefficient comprises: Extract the elevator request direction in the elevator request floor data and the running direction in the elevator running data; when the elevator request direction is the same as the elevator running direction, set the direction coefficient QFXi to K1; otherwise, set the direction coefficient QFXi to K2; 1≥K1>K2>0; Get the urgency coefficient JCXi and distance coefficient JXi of each requested floor; Calculate the basic priority coefficient QLJXi of each requested floor number through the formula; QLJXi=k×QFXi×exp(α1×JCXi+α2×JXi); wherein k is the adjustment factor, α1 and α2 are weight coefficients, α1 and α2∈(0,1); i represents the number of the requested floor.

6. The elevator control system based on intelligent building according to claim 1, characterized in that: The method of generating the final priority coefficient of each requested floor according to the request time and the corresponding basic priority coefficient in the elevator requested floor data includes: Get the request time QLQSi and the current time DS of each requested floor whose request was not received; Calculate the absolute value QCJi of the difference between the current time and the requested time; Determine whether the absolute value of the difference is greater than the request time threshold QSY; Yes, the additional coefficient QLEXi for each requested floor is calculated by the formula; QLEXi=1 / (1+exp(-(QCJi-QSY)))-b;b is the deviation, and b∈(0,1); No, do not perform the operation; Get the basic priority coefficient QLJXi of each requested floor number; The final priority coefficient QLZXi of each requested floor number is calculated by the formula; QLZXi=c1×QLJXi+c2×QLEXi; wherein, i represents the number of the requested floor, c1 and c2 are adjustment factors, c1>0, c2>0.

7. The elevator control system based on intelligent building according to claim 1, characterized in that: The elevator dispatching according to the elevator operation data and the final priority coefficient includes: Obtain the running direction, floor number, load weight, and elevator request direction and final priority coefficient of each requested floor in the elevator running data; Sort the elevator request directions that are consistent with the running direction from large to small according to the final priority coefficient; Dispatch the elevator to the requested floor corresponding to the requested floor number with the largest final priority coefficient in the running direction; Determine whether the load weight is greater than the standard load weight; Yes, sound the alarm; No, when the absolute value of the difference between the elevator load and the standard load weight is less than the standard single-person weight; the elevator does not stop at the requested floor corresponding to other requested floor numbers in the running direction; Otherwise, according to each final priority coefficient in the running direction, the requested floor number corresponding to the maximum final priority coefficient is selected for scheduling; When the load weight in the elevator operation data is 0, the operation direction in the elevator operation data is adjusted, and the elevator is dispatched to the requested floor corresponding to the requested floor number with the largest final priority coefficient in the same operation direction.

8. The elevator control system based on intelligent building according to claim 1, characterized in that: The elevator request floor prediction model is constructed by a time series prediction model, including: Obtaining a number of historical building data and their corresponding recording time through a database; the historical building data includes historical requested floor numbers and their corresponding historical requested floor times; the recording time is whether it is a working day and various time periods within each day; Performing data preprocessing on a number of recorded times and their corresponding historical building data to obtain a number of recorded times and their corresponding historical building data after preprocessing; Integrate the historical building data and recording time with the same recording time into a time feature vector; Integrate each time feature vector into several prediction sequences according to the order of recording time; Divide several prediction sequences into training set, test set and validation set in chronological order; Select the LSTM model as the base model; Train the basic model with the training set, and adjust the learning rate or other hyperparameters on the validation set to obtain the pre-trained model; By verifying the pre-trained model on the test set, we finally get the elevator request floor prediction model which inputs the record time and outputs the predicted requested floor number and its predicted requested floor number.

9. The elevator control system based on intelligent building according to claim 1, characterized in that: The method of dispatching an idle elevator to a corresponding requested floor for waiting according to the predicted requested floor number and its corresponding predicted requested floor number includes: Get the predicted requested floor number and its corresponding predicted requested floor frequency; The predicted requested floor numbers and their corresponding predicted requested floor times are sorted from largest to smallest according to the predicted requested floor times; If there are several identical maximum predicted request floor times, the corresponding predicted request floor numbers are obtained, and the idle elevator is dispatched to the request floor corresponding to the maximum predicted request floor number to wait; Otherwise, the idle elevator is dispatched to the requested floor corresponding to the predicted requested floor number with the largest number of predicted requested floors to wait.

10. An elevator control method based on an intelligent building, applied to an elevator control system based on an intelligent building as claimed in any one of claims 1 to 9, characterized in that: include: S1: Real-time acquisition of elevator request floor data, elevator waiting data and elevator operation data; S2: extracting the elevator status in the elevator operation data and determining whether the elevator status is in operation; If yes, then go to S3; If not, then go to S6; S3: Obtain elevator waiting data, elevator operation data and elevator requested floor data, generate an urgency coefficient according to the elevator waiting data; generate a distance coefficient according to the elevator operation data and the elevator requested floor data; S4: Generate a basic priority coefficient according to the urgency coefficient and the distance coefficient; generate a final priority coefficient for each requested floor according to the request time in the elevator requested floor data and the corresponding basic priority coefficient; S5: Elevator dispatching based on elevator operation data and final priority coefficient; S6: input the current building data and the corresponding recording time into the elevator request floor prediction model to output the future elevator predicted request floor number and its corresponding predicted request floor number; S7: Dispatch the idle elevator to the corresponding requested floor for waiting according to the predicted requested floor number and its corresponding predicted requested floor number.

Images