Priority acquisition system for elevator dispatching

Abstract

The invention discloses a priority acquisition system for elevator scheduling, and relates to the technical field of elevator scheduling control, the system comprises a data acquisition module, a time sequence analysis module, an emotion recognition engine, a personalized elevator taking preference learning module, a scheduling decision module and a communication and execution module; according to the data acquisition module, a high-definition camera, a microphone array and a sensor are arranged in an elevator car, facial expressions and voice data of passengers are analyzed in real time through an emotion recognition engine, the comprehensive emotion value of the passengers is quantified, corresponding adjustment is conducted according to the emotion state, and meanwhile, the emotion recognition engine is used for recognizing the emotion of the passengers. Through the personalized elevator taking preference learning module, the elevator taking habits and preferences of the passengers are deeply excavated, exclusive preference portraits are constructed, the intelligent level of elevator dispatching is improved, elevator services can be individually adjusted according to the requirements and emotional states of different passengers, and therefore the elevator taking experience and satisfaction degree of the passengers are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of elevator dispatching control, in particular to a priority acquisition system for elevator dispatching. Background Art

[0002] With the acceleration of urbanization, high-rise buildings have sprung up like mushrooms after a rain. As the main tool for vertical transportation, the operating efficiency and passenger experience of elevators have become the focus of attention from all walks of life. In order to improve the service quality and operating efficiency of elevators, elevator dispatching and control technology based on time series has emerged. This technology collects and analyzes elevator operation data, passenger behavior data and environmental parameters to achieve precise dispatching and control of elevator operation.

[0003] Traditional elevator dispatching systems mostly rely on simple floor call and response mechanisms, lacking in-depth understanding and intelligent analysis of passenger needs, emotional states, and the elevator operating environment. This leads to congestion and long waiting times during peak hours, seriously affecting passengers' travel experience. At the same time, traditional systems have failed to make full use of big data and artificial intelligence technologies to achieve optimized dispatching and intelligent control of elevator operations, thus limiting the further improvement of elevator service efficiency.

[0004] Therefore, the development of a priority acquisition system for elevator scheduling can realize intelligent scheduling and control of elevator operation, thereby effectively improving the elevator service quality and operating efficiency, and providing passengers with a more comfortable and convenient travel experience. Summary of the invention

[0005] The purpose of the present invention is to make up for the shortcomings of the prior art and to provide a priority acquisition system for elevator scheduling. It can realize comprehensive perception and intelligent analysis of passenger needs, emotional state and elevator operating environment by introducing emotion recognition, personalized elevator preference learning and adaptation technology. It can significantly improve the operating efficiency of the elevator and the passenger's elevator experience by accurately predicting passenger flow, identifying passenger emotions, learning and adapting personalized elevator preferences.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a priority acquisition system for elevator scheduling, the system comprising: a data acquisition module, a time series analysis module, an emotion recognition engine, a personalized elevator preference learning module, a scheduling decision module, and a communication and execution module;

[0007] The data collection module: by arranging cameras, microphone arrays and sensors in the elevator car, collects passenger facial images, voice information, the number of people in the car and the data of entry and exit status, and stores them with timestamps. At the same time, it connects with the building access control system, office software, and hospital registration system to obtain travel-related information, and creates a database to store relevant data;

[0008] The time series analysis module extracts passenger flow data from the database, divides the time period into 15-minute periods according to the floor dimension, cleans abnormal values, fills in missing values, performs standardization, archives by date type, constructs a structured data set, uses the passenger flow prediction formula combined with the current real-time passenger flow situation to predict the passenger flow on each floor in each period, and generates a passenger flow heat map;

[0009] The emotion recognition engine: connects to the data collection module in real time, uses facial key point positioning and expression feature extraction technology to process the collected image data, performs noise reduction, voiceprint separation and rhythm feature analysis on the voice data, inputs the processed data into the emotion quantification fusion formula, and calculates the passenger's comprehensive emotion value E T The calculation formula is: T =μ×E I +ν×E V , where E I It is the passenger emotion value obtained based on image analysis. After identifying the corresponding emotion category through the key features of facial expressions, combined with the expression duration T I Seconds, expression clarity C I The weighted correction is obtained, and the calculation formula is: E I =(emotion category basic value)×T I ×C I , E V It is the emotional value obtained based on speech analysis, which is calculated by the amplitude of voice intonation change A V , Speech speed V , the frequency of occurrence of speech emotion keywords K V Calculated, assuming the corresponding value of standard intonation, speaking speed, and no keywords is 0.5, the calculation formula is: E V =(A V -0.5)+(S V -0.5)+K V , μ and ν are the weights corresponding to the passenger emotion values ​​obtained based on image analysis and speech analysis. The calculated passenger comprehensive emotion value E V Compare and divide the comprehensive emotional state based on the comprehensive emotional threshold and output it;

[0010] The personalized elevator preference learning module: relies on the constructed database to extract elevator preference data, divides passenger groups according to their elevator requirements through feature classification formulas, deeply explores the unique travel habits and preference characteristics of different passenger groups and individuals, and constructs exclusive preference portraits. When a passenger issues an elevator call, the exclusive preference portrait is retrieved;

[0011] The dispatch decision module integrates passenger flow prediction results, passenger emotion recognition information, and personalized elevator preference data, introduces an elevator selection priority calculation formula to calculate the priority of elevator use, and generates operation control instructions covering elevator speed adjustment, start and stop control, and door opening and closing timing setting;

[0012] The communication and execution module: The dispatch decision center communicates dispatch instructions through a two-way data exchange link between the wireless communication technology and each elevator controller, adjusts the elevator running speed, start and stop actions, and door opening and closing status, and synchronously and real-time feedback on the elevator operating conditions. In the event of an elevator failure or abnormal operating condition, the system automatically and seamlessly switches to a backup dispatch plan to ensure the operation of the elevator.

[0013] Furthermore, the sensors used in the data acquisition module are temperature sensors, humidity sensors, noise sensors, pressure sensors and infrared sensor strips.

[0014] Furthermore, the emotion recognition engine uses the passenger flow prediction formula combined with the current real-time passenger flow to predict the passenger flow of each floor in each period. Let F t,f represents the predicted passenger flow at future time t and floor f, F h,f is the average number of passengers in the same period of history, D t is the current date type coefficient, G t-1,f is the actual passenger flow change on floor f at the previous moment t-1, α, β, γ are weight coefficients, and the formula is:

[0015] Furthermore, the emotion recognition engine obtains the basic value of the emotion category based on the rule-based expression feature judgment:

[0016] For anxiety, the rule is: when the frown level exceeds a certain threshold T f And the degree of eyes widening exceeds the threshold T e , judged as anxious emotion, the basic value of the emotion category is -1, let the frown degree be represented by f, the eye widening degree be represented by e, f>T f And e>T e , then the emotion category is anxiety, and the base value is -1;

[0017] For calm emotions, the rules are set as follows: facial muscles are relaxed, mouth is naturally closed, and eyes are in normal state. Let m represent the degree of facial muscle relaxation, b represent the state of mouth, g represent the state of eyes, and the corresponding threshold T m 、T b 、T g , m>T m And b>T b And g>T g , the emotional category is calm and the base value is 0;

[0018] For the pleasant emotion, the rule is set as: the degree of the upward curvature of the corners of the mouth exceeds a certain threshold T s , and the degree of squinting of the eyes exceeds a certain threshold T g′ , let the degree of the upward curvature of the corners of the mouth be represented by s, and the degree of squinting of the eyes be represented by g ′ s > T s and g ′ > T g ′ The emotional category is pleasant and the base value is 1.

[0019] Furthermore, for the comprehensive emotional state threshold corresponding in the emotional recognition engine, E T < -0.5 is determined as anxiety, -0.5 ≤ E T ≤ 0.5 is calm, E T > 0.5 is pleasant.

[0020] Furthermore, in the personalized elevator riding preference learning module, the passenger groups are divided according to the passenger elevator riding requirements through the feature classification formula. Let P i represent the i-th passenger, D ij be the proportion of the frequency of the i-th passenger going to the j-th floor, O i be the comprehensive car operation habit score, G i represent the encoding based on the travel purpose, C ij be the association strength between the travel purpose and the floor, and calculate the comprehensive eigenvalue F i of the passenger P i , and the formula is: F i = α × ∑ j (D ij × C ij ) + β × O i + γ × G i , where α, β, and γ are weight coefficients, and the classification thresholds T1, T2,..., T i are set, and i is determined according to the historical data distribution and scene characteristics. The classification rule is: F i < T1, classify the passenger P i into the first category, T1 ≤ F i < T2, classify into the second category, T i-1 ≤ F i < T i , classify into the i-th category.

[0021] Furthermore, for the construction of the exclusive elevator riding preference portrait in the personalized elevator riding preference learning module, a preference vector is constructed for each passenger where p1 is the floor preference weight, p2 is the environment preference, p3 represents the service duration preference, and let the matching degree M of the elevator e to the passenger demand e, and the elevator current floor L e Distance from calling floor L h Distance d e , remaining space in the car S e , Operation stability R e Related, combined with the weight calculation of the corresponding dimension of the preference vector, the formula is: Where D max is the maximum floor span of the building, S max is the maximum area of ​​the car, w1, w2, w3 are the weights of each factor.

[0022] Furthermore, the calculation of the passenger preference vector in the personalized elevator preference learning module is:

[0023] Floor preference weight p1: Analyze the passenger's historical elevator records and count the frequency of their visits to each floor f i , let the total number of elevator rides be F total , the formula is: After normalization, the sum of the weights of each floor is 1. The formula is:

[0024] The calculation formula of environmental preference weight p2 is: p2 = (1-T0) × k T +(1-H0)×k H +(1-N0)×k N , where T0 is the relative comfortable temperature, H0 is the relative comfortable humidity, and N0 is the relative comfortable sound. The calculation formula is: T min , T max They are respectively the temperature comfort range [T min ,T max ], H min ,H max They are humidity comfort range [H min ,H max ], N tol The upper limit of noise tolerance is T, the temperature sensor monitoring value, H, the humidity sensor monitoring value, N, the noise sensor monitoring value, k T , k H , k N They are the weight coefficients corresponding to temperature, humidity and sound;

[0025] Service time preference weight p3: The total time that the passenger presses the door opening delay button is recorded as t delay , the total number of elevator rides is N, and the average reference time for the reasonable door opening delay for a single elevator ride is t ref , the service time preference weight p3 calculation formula is:

[0026] Furthermore, the elevator selection priority calculation formula is introduced into the scheduling decision to calculate the elevator use priority Q e , the calculation formula is: Q e =U t,f ×I E ×M e , where U t,f Indicates the passenger flow emergency coefficient, which is calculated according to the passenger flow prediction formula. t,f The average number of people P in the elevator car avg Comparative calculation, the formula is: I E Represents the emotional intervention coefficient, based on the comprehensive emotional value E output by the emotion recognition engine T Get E T <-0.5, I E =1.5; -0.5≤E T ≤0.5I E =1; E T >0.5, I E =0.8,M e Indicates the matching degree of the elevator to the passenger demand. Press Q e Dispatch elevators according to high and low priority and generate operation control instructions.

[0027] Compared with the prior art, the priority acquisition system for elevator scheduling has the following beneficial effects:

[0028] 1. The present invention utilizes an emotion recognition engine to analyze passengers' facial expressions and voice data in real time, quantifies passengers' comprehensive emotion values, and makes corresponding adjustments based on their emotional states. At the same time, through a personalized elevator preference learning module, it deeply explores passengers' elevator riding habits and preferences and constructs exclusive preference portraits. This not only improves the intelligence level of elevator dispatching, but also enables elevator services to be personalized according to the needs and emotional states of different passengers, thereby significantly enhancing passengers' elevator riding experience and satisfaction.

[0029] 2. The present invention uses a time series analysis module, combined with historical passenger flow data and real-time passenger flow conditions, to accurately predict the passenger flow conditions on each floor in each period and generate a passenger flow heat map. On this basis, the scheduling decision module comprehensively considers the passenger flow prediction results, passenger emotion recognition information and personalized elevator preference data, introduces the elevator selection priority calculation formula, and intelligently generates operation control instructions, making elevator scheduling more efficient and flexible, and can effectively cope with passenger flow pressure during peak hours, reduce passenger waiting time, and improve elevator operation efficiency. At the same time, through real-time feedback of elevator operating conditions and seamless switching of backup scheduling plans, the safe and stable operation of the elevator is guaranteed, and the passengers' elevator experience and safety are improved.

[0030] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] 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 prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0032] Figure 1 A flow chart of a priority acquisition system for elevator dispatching;

[0033] Figure 2 Architectural association diagram of a system for obtaining priorities for elevator scheduling. DETAILED DESCRIPTION

[0034] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. 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.

[0035] Embodiment 1:

[0036] Office building elevator dispatching

[0037] High-definition cameras, microphone arrays, temperature sensors, humidity sensors, noise sensors, pressure sensors and infrared sensor strips are installed in the elevator cars of office buildings. The cameras and microphone arrays collect facial images and voice information of passengers, and the sensors collect environmental data in the car. At the same time, they are connected to the building's access control system to obtain employees' entry and exit information, and to office software to obtain employees' meeting arrangements and travel-related information. All this data is time-stamped and stored in the database.

[0038] The time series analysis module extracts passenger flow data from the database, divides the time period into 15-minute periods according to the floor dimension, cleans abnormal values ​​(such as erroneous data caused by sensor failure), fills in missing values ​​(such as data not collected during a period due to network problems), performs standardization and unified dimension, and archives them by date type (weekdays, weekends), constructs a structured data set, and uses the passenger flow prediction formula combined with the current real-time passenger flow situation to predict the passenger flow on each floor in each period. The formula is: Among them, Ft,f represents the predicted passenger flow at future time t and floor f, F h ,f is the average passenger flow in the same period of history, D t is the current date type coefficient, G t-1,f is the actual passenger flow change on floor f at the previous moment t-1, α, β, γ are weight coefficients, for example, based on the average passenger flow F from 9 to 10 am in the same period in history h,f , the actual passenger flow change at the previous moment G t-1,f And the average number of palace flow in the first 6 periods Predict passenger flow on different floors at future times and generate passenger flow heat maps.

[0039] The emotion recognition engine connects to the data collection module in real time. Based on the collected image data, it uses facial key point positioning and expression feature extraction technology to obtain the basic value of the emotion category through rule-based expression feature judgment (such as frowning degree exceeding the threshold T f And the degree of eyes widening exceeds the threshold T e It is judged as anxiety, with a base value of -1; facial muscles are relaxed, mouth is naturally closed, and eyes are normal, which is judged as calmness, with a base value of 0; the degree of mouth corners rising exceeds the threshold T s And the degree of eye squinting exceeds the threshold T' g It is judged as a happy emotion, with a base value of 1), and combined with the duration of the expression T I and expression clarity C I The passenger emotion value E obtained by weighted correction based on image analysis I .The calculation formula is: E I =(emotion category basic value)×T I ×C I , noise reduction, voiceprint separation and rhythm feature analysis are performed on the voice data, and the voice intonation change amplitude A V , speaking speed S V , the frequency of occurrence of speech emotion keywords K V Calculate the emotion value E based on speech analysis V , the calculation formula is: E V =(A V -0.5)+(S V -0.5)+K V , and then calculate the formula E according to the comprehensive emotion value of the passenger T =μ×E I +ν×E V Get the comprehensive emotion value, and compare it with the comprehensive emotion threshold to divide the comprehensive emotion state (such as the comprehensive emotion value E T <-0.5 is considered anxiety, -0.5≤E T ≤0.5 is calm, E T>0.5 is pleasant).

[0040] The personalized elevator preference learning module relies on the constructed database to extract the elevator preference data and calculates the comprehensive feature value F of the passenger through the feature classification formula. i , the formula is: F i =α×∑ j (D ij ×C ij )+β×O i +γ×G i , divide the passenger groups according to the comprehensive characteristic values, (such as setting classification thresholds T1 and T2, the comprehensive characteristic values ​​less than T1 are classified into one category, and different threshold ranges are classified into different categories), where P i represents the i-th passenger, D ij is the frequency ratio of the i-th passenger going to the j-th floor, O i is the comprehensive car operation habit score, G i Indicates the code based on the travel purpose, C ij is the strength of association between travel purpose and floor, α, β, γ are weight coefficients, for example, calculate the frequency ratio D of employees going to each floor ij , Comprehensive car operation habit score O i And based on the travel purpose code G i (For example, going to the conference room floor may correspond to a travel purpose related to the meeting), determine the comprehensive characteristic value of the employee, and deeply explore the unique travel habits and preference characteristics of different passenger groups and individuals based on each division to build an exclusive preference portrait. For example, if an employee often goes to high floors and likes a quiet environment, the system will build a preference portrait for him with a higher floor preference weight and a quiet environment preference.

[0041] The dispatch decision module integrates passenger flow prediction results, passenger emotion recognition information, and personalized elevator preference data, and introduces the passenger flow emergency coefficient U t,f (e.g., based on the comparison between the predicted passenger flow and the car capacity, the emergency coefficient increases when the passenger flow is close to the car capacity), emotional intervention coefficient I E (Determined according to the passenger's emotional state, the coefficient corresponding to anxious passengers is higher, when E T <-0.5, I E =1.5; -0.5≤E T ≤0.5I E =1; E T >0.5, I E =0.8), and the elevator selection priority calculation formula Q is formed comprehensively e =U t,f ×I E ×M e Calculate the priority of elevator use, where M eThe elevator's matching degree to passenger demand and the elevator's current floor L e Distance from calling floor L h Distance d e , remaining space in the car S e , Operation stability R e Related, the calculation formula is: For example, for the emergency coefficient U t,f High and anxious passengers (emotional intervention coefficient I E =1.5), give priority to dispatching passengers with a matching degree M e High (such as the distance to the elevator call floor is d e Small, large remaining space in the car e , stable operation e ) and generates operation control instructions covering elevator running speed adjustment (such as increasing the running speed), start and stop control (such as reducing the number of stops), and door opening and closing timing setting (such as extending the door opening time to facilitate passengers to enter and exit).

[0042] The communication and execution module transmits dispatch instructions to each elevator controller through wireless communication technology, adjusts the elevator's operating speed, start and stop actions, and door opening and closing status, and simultaneously provides real-time feedback on the elevator's operating conditions. If the elevator fails or operates abnormally, the system automatically and seamlessly switches to the backup dispatch plan. Relying on redundant design and intelligent diagnosis and recovery mechanism, it ensures the normal operation of the elevator. If an elevator suddenly fails, the system immediately reallocates the tasks originally assigned to the elevator to other normal elevators, and notifies maintenance personnel to deal with the faulty elevator in a timely manner.

[0043] To sum up, in office buildings, this elevator dispatching system collects data through the collaborative work of multiple devices and systems, uses time series analysis to accurately predict passenger flow, uses emotion recognition and preference learning modules to gain insight into passenger needs, and the dispatching decision module scientifically calculates priorities and issues instructions. The communication execution module ensures operation, effectively improving elevator operation efficiency, reducing waiting time, improving comfort and safety, and greatly optimizing the elevator experience of office staff, fully reflecting its important value in office scenarios.

[0044] Embodiment 2:

[0045] Hospital elevator dispatch

[0046] Data collection and analysis: High-definition cameras, microphone arrays, temperature sensors, humidity sensors, noise sensors, pressure sensors and infrared sensor strips are arranged in the elevator car of the hospital. The cameras and microphone arrays collect facial images and voice information of patients and their families. The sensors collect environmental data in the car, such as temperature T, humidity H, and noise N. At the same time, they connect with the hospital registration system to obtain information about the patient's registration department and consultation time. These data are timestamped and stored in the database. The time series analysis module extracts passenger flow data from the database, divides the time period into 15 minutes according to the floor dimension, cleans abnormal values ​​(such as erroneous data caused by sensor failure), fills in missing values ​​(such as data not collected during a period of time due to network problems), performs standardized processing and unifies the dimensions, and classifies and archives them according to date types (weekdays, weekends, holidays, because the distribution of passenger flow in hospitals on different dates is different), constructs a structured data set, and uses the passenger flow prediction formula combined with the current real-time passenger flow situation to predict the passenger flow on each floor in each period. The formula is: Among them, F t,f represents the predicted passenger flow at future time t and floor f, F h,f is the average number of passengers in the same period of history, D t is the current date type coefficient, G t-1,f is the actual passenger flow change on floor f at the previous moment t-1, α, β, γ are weight coefficients, for example, based on the average passenger flow F of each floor (such as the outpatient hall floor, different department floors, and inpatient floor) at 9-10 am in the same period in history h,f , the actual passenger flow change at the previous moment G t-1,f And the average number of passengers in the first 6 periods Predict passenger flow on different floors in the future and generate passenger flow heat maps so that hospitals can prepare for the allocation of elevator resources in advance.

[0047] The emotion recognition engine connects to the data collection module in real time. Based on the collected image data, it uses facial key point positioning and expression feature extraction technology to obtain the basic value of the emotion category through rule-based expression feature judgment (such as frowning degree exceeding the threshold T f And the degree of eyes widening exceeds the threshold T e It is judged as anxiety, with a base value of -1; facial muscles are relaxed, mouth is naturally closed, and eyes are normal, which is judged as calmness, with a base value of 0; the degree of mouth corners rising exceeds the threshold T s And the degree of eye squinting exceeds the threshold T' g It is judged as a happy emotion, with a base value of 1), and combined with the duration of the expression T I and expression clarity C I The passenger emotion value E obtained by weighted correction based on image analysis I , the calculation formula is: E I=(emotion category basic value)×T I ×C I , noise reduction, voiceprint separation and rhythm feature analysis are performed on the voice data, and the voice intonation change amplitude A V , speaking speed S V , the frequency of occurrence of speech emotion keywords K V Calculate the emotion value E based on speech analysis V , the calculation formula is: E V =(A V -0.5)+(S V -0.5)+K V , and then the comprehensive emotion value is obtained according to the calculation formula of the comprehensive emotion value of the passenger: E T =μ×E I +ν×E V and compared with the comprehensive emotion threshold to divide the comprehensive emotion state (such as the comprehensive emotion value E T <-0.5 is considered anxiety, -0.5≤E T ≤0.5 is calm, E T >0.5 is pleasant).

[0048] According to the frequency of patients going to different floors of different departments, their operating habits in the car (such as whether a stretcher is needed), and the purpose of travel (seeing a doctor, visiting a patient), the passenger groups are divided by the feature classification formula: i =α×∑ j (D ij ×C ij )+β×O i +γ×G i , build a dedicated preference profile, where P i represents the i-th passenger (patient or family member), D ij is the frequency ratio of the i-th passenger going to the j-th floor, O i is the comprehensive car operation habit score, G i Indicates the code based on the travel purpose, C ij α, β, and γ are weight coefficients for the strength of association between travel purpose and floor. For example, family members who often go to the obstetrics and gynecology floor may pay more attention to the stability of the elevator when taking the elevator. The system builds a corresponding preference portrait for them, while patients and their families who often go to the emergency floor may need faster-response elevator services, and the system will also reflect this feature in the portrait.

[0049] The dispatch decision module comprehensively considers passenger flow forecast, emotional state and personalized preferences. For patients with limited mobility or serious illness, the appropriate elevator is arranged first based on their floor and registration time, and the elevator operation mode is adjusted according to the emotional state. The passenger flow emergency coefficient U is introduced. t,f(e.g., based on the comparison between the predicted passenger flow and the car capacity, the emergency coefficient increases when the passenger flow is close to the car capacity), emotional intervention coefficient I E (Determined according to the passenger's emotional state, the coefficient corresponding to anxious patients is higher, when E T <-0.5, I E =1.5; -0.5≤E T ≤0.5I E =1; E T >0.5, I E =0.8), and the elevator selection priority calculation formula is formed to calculate the priority of elevator use. The formula is: Q e =U t,f ×I E ×M e , where M e The elevator's matching degree to passenger demand and the elevator's current floor L e Distance from calling floor L h Distance d e , remaining space in the car S e , Operation stability R e Related, the calculation formula is:

[0050] For example, for the floor where the patient who needs emergency surgery is located, the system detects the passenger flow emergency coefficient U t,f Higher, and the patient's family may be in a state of extreme anxiety (emotional intervention coefficient I E =1.5), at this time, the system will give priority to the elevator that matches the passenger demand M e The highest elevator, assuming that an elevator e1 is currently on floor L e1 Distance from the patient's elevator calling floor L h Closer (ie d e 1 is smaller), the remaining space in the car is S e1 Large enough to accommodate a stretcher, operating stability R e1 Good, and according to the preference profile of the patient group on this floor, the parameters of this elevator match the patient's needs M e1 If the floor's speed is higher, the system will dispatch elevator e1 to that floor and generate operation control instructions covering elevator speed adjustment (such as increasing the running speed), start and stop control (such as reducing the number of stops, going directly to the target floor or giving priority to stopping at relevant key floors), and door opening and closing timing settings (such as extending the door opening time to facilitate patients' entry and exit, especially those with limited mobility).

[0051] At the same time, the environment in the elevator is adjusted according to the calculation formula of the environmental preference weight p2, which is: p2 = (1-T0) × k T +(1-H0)×k H +(1-N0)×kN ,in T is the temperature sensor monitoring value, H is the humidity sensor monitoring value, N is the noise sensor monitoring value, T min、 T max is the comfortable temperature range, H min、 H max is the comfortable humidity range, N tol is the upper limit of noise tolerance, k T、 k H、 k N They are the weight coefficients corresponding to temperature, humidity and sound. If the temperature T in the car deviates from the comfortable temperature range, the system can adjust the elevator ventilation equipment appropriately. If the noise N exceeds the upper limit of noise tolerance, the system can issue a prompt to reduce the noise.

[0052] The communication and execution module transmits dispatch instructions to each elevator controller through wireless communication technology, adjusts the elevator's operating speed, start and stop actions, and door opening and closing status, and simultaneously provides real-time feedback on the elevator's operating conditions. If the elevator fails or operates abnormally, the system automatically and seamlessly switches to the backup dispatch plan, relying on redundant design and intelligent diagnosis and recovery mechanism to ensure the normal operation of the elevator. If an elevator suddenly fails, the system will immediately reallocate the task originally assigned to the elevator to other normal elevators, and notify maintenance personnel to deal with the faulty elevator in a timely manner to ensure that the patient's medical journey is not greatly affected.

[0053] To sum up, in the hospital scenario, the system relies on the connection between the car equipment and the registration system to collect data, time series analysis to cope with changes in passenger flow, emotion recognition to assist in caring for patients, preference profiling to meet special needs, comprehensive considerations in scheduling decisions to ensure service priority, and communication execution to ensure operational reliability. It significantly improves the quality and efficiency of hospital elevator services, provides convenient and comfortable protection for patients to seek medical treatment, and is of great significance to optimizing hospital operations and medical services.

[0054] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A priority acquisition system for elevator dispatching, characterized in that: The system includes: data acquisition module, time series analysis module, emotion recognition engine, personalized elevator preference learning module, scheduling decision module and communication and execution module; The data collection module: by arranging cameras, microphone arrays and sensors in the elevator car, collects passenger facial images, voice information, the number of people in the car and the data of entry and exit status, and stores them with timestamps. At the same time, it connects with the building access control system, office software, and hospital registration system to obtain travel-related information, and creates a database to store relevant data; The time series analysis module extracts passenger flow data from the database, divides the time period into 15-minute periods according to the floor dimension, cleans abnormal values, fills in missing values, performs standardization, archives by date type, constructs a structured data set, uses the passenger flow prediction formula combined with the current real-time passenger flow situation to predict the passenger flow on each floor in each period, and generates a passenger flow heat map; The emotion recognition engine: connects to the data collection module in real time, uses facial key point positioning and expression feature extraction technology to process the collected image data, performs noise reduction, voiceprint separation and rhythm feature analysis on the voice data, inputs the processed data into the emotion quantification fusion formula, and calculates the passenger's comprehensive emotion value E T The calculation formula is: T =μ×E I +ν×E V , where E I It is the passenger emotion value obtained based on image analysis. After identifying the corresponding emotion category through the key features of facial expressions, combined with the expression duration T I Seconds, expression clarity C I The weighted correction is obtained, and the calculation formula is: E I =(emotion category basic value)×T I ×C I , E V It is the emotional value obtained based on speech analysis, which is calculated by the amplitude of voice intonation change A V , Speech speed V , the frequency of occurrence of speech emotion keywords K V Calculated, assuming the corresponding value of standard intonation, speaking speed, and no keywords is 0.5, the calculation formula is: E V =(A V -0.5)+(S V -0.5)+K V , μ and ν are the weights corresponding to the passenger emotion values ​​obtained based on image analysis and speech analysis. The calculated passenger comprehensive emotion value E V Compare and divide the comprehensive emotional state based on the comprehensive emotional threshold and output it; The personalized elevator preference learning module: relies on the constructed database to extract elevator preference data, divides passenger groups according to their elevator requirements through feature classification formulas, deeply explores the unique travel habits and preference characteristics of different passenger groups and individuals, and constructs exclusive preference portraits. When a passenger issues an elevator call, the exclusive preference portrait is retrieved; The dispatch decision module integrates passenger flow prediction results, passenger emotion recognition information, and personalized elevator preference data, introduces an elevator selection priority calculation formula to calculate the priority of elevator use, and generates operation control instructions covering elevator speed adjustment, start and stop control, and door opening and closing timing setting; The communication and execution module: The dispatch decision center communicates dispatch instructions through a two-way data exchange link between the wireless communication technology and each elevator controller, adjusts the elevator running speed, start and stop actions, and door opening and closing status, and synchronously and real-time feedback on the elevator operating conditions. In the event of an elevator failure or abnormal operating condition, the system automatically and seamlessly switches to a backup dispatch plan to ensure the operation of the elevator.

2. The priority acquisition system for elevator scheduling according to claim 1, characterized in that: The sensors used in the data acquisition module are temperature sensors, humidity sensors, noise sensors, pressure sensors and infrared sensing strips.

3. The priority acquisition system for elevator scheduling according to claim 1, characterized in that: The emotion recognition engine uses the passenger flow prediction formula combined with the current real-time passenger flow situation to predict the passenger flow of each floor in each period, assuming F t,f represents the predicted passenger flow at future time t and floor f, F h,f is the average number of passengers in the same period of history, D t is the current date type coefficient, G t-1,f is the actual passenger flow change on floor f at the previous moment t-1, α, β, γ are weight coefficients, and the formula is:

4. The priority acquisition system for elevator scheduling according to claim 1, characterized in that: The emotion recognition engine obtains the basic value of the emotion category by judging the expression features based on the rules: For anxiety, the rule is: when the frown level exceeds a certain threshold T f And the degree of eyes widening exceeds the threshold T e , judged as anxious emotion, the basic value of the emotion category is -1, let the frown degree be represented by f, the eye widening degree be represented by e, f>T f And e>T e , then the emotion category is anxiety, and the base value is -1; For calm emotions, the rules are set as follows: facial muscles are relaxed, mouth is naturally closed, and eyes are in normal state. Let m represent the degree of facial muscle relaxation, b represent the state of mouth, g represent the state of eyes, and the corresponding threshold T m , T b , T g , m>T m And b>T b And g>T g , the emotional category is calm and the base value is 0; For happy emotions, the rule is: the degree of mouth corner lifting exceeds a certain threshold T s , and the degree of eye squinting exceeds a certain threshold T g′ , let s represent the degree of mouth corner upturn and g represent the degree of eye narrowing ′ It means that s>T s And g ′ >T g ′ The emotion category is Pleasure and the base value is 1.

5. The priority acquisition system for elevator scheduling according to claim 1, characterized in that: The corresponding comprehensive emotional state threshold in the emotion recognition engine, E T <-0.5 is considered anxiety, -0.5≤E T ≤0.5 is calm, E T >0.5 is pleasant.

6. The priority acquisition system for elevator scheduling according to claim 1, characterized in that: In the personalized elevator riding preference learning module, the passenger groups are divided according to the passenger elevator riding requirements through the feature classification formula. Let P i represent the i-th passenger, D ij be the proportion of the frequency of the i-th passenger going to the j-th floor, O i be the comprehensive car operation habit score, G i represent the encoding based on the travel purpose, C ij be the correlation strength between the travel purpose and the floor. Calculate the comprehensive feature value F i of the passenger P i , and the formula is: F i =α×∑ j (D ij ×C ij )+β×O i +γ×G i , where α, β, and γ are weight coefficients. Set classification thresholds T1, T2,..., T i , and i is determined according to the historical data distribution and scene characteristics. The classification rule is: F i <T1, classify the passenger P i into the first category, T1≤F i <T2, classify it into the second category, T i-1 ≤F i <T i , classify it into the i-th category.

7. The priority acquisition system for elevator scheduling according to claim 1, characterized in that: The construction of the exclusive elevator preference portrait in the personalized elevator preference learning module constructs a preference vector for each passenger Among them, p1 is the floor preference weight, p2 is the environment preference, and p3 represents the service time preference. Suppose the elevator e matches the passenger demand M e , and the elevator current floor L e Distance from calling floor L h Distance d e , remaining space in the car S e , Operation stability R e Related, combined with the weight calculation of the corresponding dimension of the preference vector, the formula is: Where D max is the maximum floor span of the building, S max is the maximum area of ​​the car, w1, w2, w3 are the weights of each factor.

8. The priority acquisition system for elevator scheduling according to claim 1, characterized in that: Calculation of passenger preference vector in the personalized elevator preference learning module: Floor preference weight p1: Analyze the passenger's historical elevator records and count the frequency of their visits to each floor f i , let the total number of elevator rides be F total , the formula is: After normalization, the sum of the weights of each floor is 1. The formula is: The calculation formula of environmental preference weight p2 is: p2 = (1-T0) × k T +(1-H0)×k H +(1-N0)×k N , where T0 is the relative comfortable temperature, H0 is the relative comfortable humidity, and N0 is the relative comfortable sound. The calculation formula is: T min , T max They are respectively the temperature comfort range [T min ,T max ], H min ,H max They are humidity comfort range [H min ,H max ], N tol The upper limit of noise tolerance is T, the temperature sensor monitoring value, H, the humidity sensor monitoring value, N, the noise sensor monitoring value, k T , k H , k N They are the weight coefficients corresponding to temperature, humidity and sound; Service time preference weight p3: The total time that the passenger presses the door opening delay button is recorded as t delay , the total number of elevator rides is N, and the average reference time for the reasonable door opening delay for a single elevator ride is t ref , the service time preference weight p3 calculation formula is:

9. The priority acquisition system for elevator scheduling according to claim 1, characterized in that: The elevator selection priority calculation formula is introduced in the scheduling decision to calculate the priority Q of the elevator use e , the calculation formula is: Q e =U t,f ×I E ×M e , where U t,f Indicates the passenger flow emergency coefficient, which is calculated according to the passenger flow prediction formula. t,f The average number of people P in the elevator car avg Comparative calculation, the formula is: I E Represents the emotional intervention coefficient, based on the comprehensive emotional value E output by the emotion recognition engine T Get E T <-0.5, I E =1.5; -0.5≤E T ≤0.5I E =1; E T >0.5, I E =0.8,M e Indicates the matching degree of the elevator to the passenger demand. Press Q e Dispatch elevators according to high and low priority and generate operation control instructions.

Images