Elevator group control energy-saving dispatching method and system based on load prediction

Abstract

The invention relates to the technical field of elevator dispatching optimization, in particular to an elevator group control energy-saving dispatching method and system based on load prediction.The method comprises the steps that the actual passenger flow with the same time property as the current moment in elevator historical operation data is analyzed, and a first predictive factor is constructed; the method comprises the following steps: acquiring a pedestrian flow load, taking the pedestrian flow load as a reference prediction value, analyzing the similarity between the reference prediction value and all historical operation data, distributing contribution degree weight to the historical operation data of each day, constructing a second prediction factor, obtaining a pedestrian flow prediction value, and finally correcting the pedestrian flow prediction value in real time according to the difference between the pedestrian flow prediction value and the actual pedestrian flow load of the current day; and constructing an adaptive inertia weight of a particle swarm algorithm based on the human traffic prediction correction value, and searching an optimal solution to perform response scheduling. The invention aims to improve the dispatching efficiency of the whole elevator group.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of elevator dispatching optimization, in particular to an elevator group control energy-saving dispatching method and system based on load prediction. BACKGROUND

[0002] Elevators are key facilities for vertical transportation. With the development of the construction industry, the height and area of high-rise buildings in cities are becoming larger and larger. Large-scale elevator groups are usually equipped in buildings to meet the needs and efficiency of people going up and down. With the continuous development of various sensor technologies and intelligent control devices, the elevator control system is no longer limited to single elevator control. The demand for group control elevator systems in buildings is increasing. In order to improve the operation efficiency of elevators in the entire building, how to efficiently judge the demand for elevators to intelligently dispatch the elevator group, and design a more intelligent and efficient energy-saving elevator group control dispatching method is an important problem.

[0003] Most of the current elevator group control dispatching methods are based on traditional dispatching algorithms or intelligent heuristic optimization algorithms to calculate the comprehensive evaluation value of the elevator responding to the actual call request according to multiple performance indicators of the elevator. Then, the elevators are dispatched to complete the demand for elevators at the actual call request according to the comprehensive evaluation value of each elevator. However, this elevator group control dispatching method does not take into account the actual passenger flow load data of each floor at different time periods, which leads to poor elevator floor selection during the morning and evening passenger flow load peaks, long waiting time for users, and low elevator dispatching efficiency. On the other hand, when facing actual call requests, large elevator group control systems will dispatch multiple elevators to work. When the passenger flow is small, it is inevitable to cause the empty running of multiple elevators, which increases the use frequency and start-stop times of elevators, resulting in a large waste of energy and affecting the efficiency of elevator dispatching. SUMMARY

[0004] In view of the above, it is necessary to provide an elevator group control energy-saving dispatching method and system based on load prediction to solve the above problems.

[0005] The first aspect of the present application provides an elevator group control energy-saving dispatching method based on load prediction, which comprises: Using the historical operation data of the elevator group, the periodic distribution characteristics of the actual passenger flow of each floor at the current time period and the similarity degree with the distribution of the actual passenger flow in the historical operation data are analyzed to determine the passenger flow prediction value of each floor at the current time period. Each reference correction time period of each time period is preset, the overall change characteristics of the difference between the passenger flow prediction value and the actual passenger flow in all reference correction time periods of the current time period are compared, the passenger flow prediction error of each floor at the current time period is determined, the passenger flow prediction value is corrected, and the passenger flow prediction correction value of each floor at the current time period is obtained. Based on the predicted passenger flow and the actual load capacity of the elevator, the elevator's load availability is determined. Based on the predicted passenger flow for all floors in the current time period, the fixed inertia weights of the particle swarm optimization algorithm are adjusted to find the optimal solution for response scheduling.

[0006] The process of determining the predicted pedestrian flow for each floor during the current time period is as follows: Based on the periodic distribution characteristics, determine the first prediction factor for each floor in the current time period; Analyze the similarity between the first predictor and the distribution of actual pedestrian traffic in historical operational data to determine the second predictor for each floor in the current time period; For each time period and each floor, the first and second predictive factors are weighted respectively, and the weighted sum and rounded value is used as the predicted value of pedestrian flow.

[0007] Specifically, determining the first prediction factor for each floor in the current time period involves: In the historical data corresponding to the current moment, obtain the average actual pedestrian flow of each floor in the same time period for all floors with the same week number as the current moment, and use it as the first predictor for each floor in the current time period.

[0008] Specifically, the second prediction factor for determining each floor in the current time period is as follows: Encode all the first predictor factors obtained on the same day to obtain the encoding matrix; Encode all actual passenger flow for each day in the historical operational data of the day, and obtain each reference coding matrix corresponding to the current moment; Calculate the similarity between the coding matrix of the day and each of its reference coding matrices. Use the similarity ratio between the coding matrix of the day and each of its reference coding matrices as the weight of the actual foot traffic of each floor in the current time period of the corresponding historical days. Sum the weighted data to obtain the second prediction factor for each floor in the current time period.

[0009] In this coding matrix, each row corresponds to each time period of the day, and each column corresponds to each floor.

[0010] The specific process for determining the predicted pedestrian flow error for each floor during the current time period is as follows: For each floor in the current time period, calculate the difference between the predicted and actual pedestrian flow for each reference correction time period, and perform positive fusion with the negative correlation mapping of the predicted pedestrian flow. Average the positive fusion results obtained from all reference correction time periods to obtain the pedestrian flow prediction error for each floor in the current time period.

[0011] The specific process for obtaining the predicted and corrected pedestrian flow values ​​for each floor during the current time period is as follows: Calculate the difference between the natural number 1 and the predicted pedestrian flow error for each floor in each time period, and then perform a positive fusion with the predicted pedestrian flow value obtained for each floor in each time period to obtain the corrected pedestrian flow prediction value for each floor in each time period.

[0012] Specifically, determining the elevator's load availability involves: For the ratio of the predicted passenger flow correction value for each floor in the current time period to the actual load capacity of the elevator, when the ratio is greater than or equal to a preset threshold, a virtual response request is sent to the corresponding floor; otherwise, it is not triggered.

[0013] Specifically, adjusting the fixed inertia weights of the particle swarm optimization algorithm involves: If the elevator issues a virtual response request, for the current time period, calculate the total number of predicted correction values ​​for passenger flow on all floors; calculate the average actual passenger flow of all floors in the same time period as the current time in historical operation data; calculate the ratio between the total number and the average value, and then multiply it by a preset fixed weight to obtain the adaptive inertia weight.

[0014] Secondly, embodiments of this application also provide an elevator group control energy-saving scheduling system based on load prediction, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of any of the methods described above.

[0015] This application has at least the following beneficial effects: This application first analyzes the overall distribution characteristics of actual passenger flow in the historical operation data of elevator groups with the same time nature as the current time for each floor, determines the first predictive factor for each floor in the current time period, and uses historical data, especially elevator operation data of the same week and time period as the current time, to establish a benchmark prediction model, reflecting the periodicity and regularity of passenger flow, providing a relatively accurate preliminary prediction, helping the system to identify regular time periods with stable passenger flow, and helping to predict the load of elevators on each floor in advance during the time period, avoiding overcrowding or empty elevators on certain floors.

[0016] Secondly, the similarity between the first predictor and the distribution of actual pedestrian traffic in historical operating data is analyzed to determine the second predictor for each floor in the current time period. By comparing the first predicted value with the distribution of actual pedestrian traffic in history, the prediction is further adjusted to ensure that the prediction is closer to the actual situation, can better cope with possible changes in the actual environment, and more accurately reflect the specific pedestrian traffic situation of the floor in the current time period, thereby avoiding over-prediction or underestimation of pedestrian traffic.

[0017] Furthermore, the concept of a reference correction time period is introduced to compare the current time period with similar past time periods, thereby further refining the passenger flow forecast. By correcting the error, a more accurate passenger flow forecast correction value can be obtained, which can adapt to the special characteristics of different floors and time periods, making elevator scheduling more flexible, avoiding errors caused by a single forecast mode, reducing fluctuations in elevator load, and improving the stability of the elevator group control system.

[0018] Based on the predicted passenger flow and the actual elevator load capacity, the system determines whether to issue a virtual response request. Matching the elevator's actual load capacity with the predicted passenger flow is crucial. At this stage, if the predicted value indicates that the elevator load is close to saturation during a certain period, the system may trigger a virtual response request to schedule the elevator in advance. By determining the elevator load in advance, elevator congestion can be avoided, scheduling responses can be initiated earlier, waiting times can be reduced, and more elevator resources can be allocated during peak hours, preventing passengers on certain floors from waiting too long due to insufficient elevator capacity.

[0019] Finally, if the elevator issues a virtual response request, the fixed inertia weights of the particle swarm optimization algorithm are adjusted based on the predicted passenger flow values ​​for all floors in the current time period. The algorithm then searches for the optimal solution for response scheduling. After the virtual response request is issued, the inertia weights of the elevator scheduling are adjusted using the particle swarm optimization algorithm to optimize the elevator scheduling scheme. The particle swarm optimization algorithm finds the optimal solution by simulating group behavior, thereby achieving more efficient scheduling decisions. By dynamically adjusting the inertia weights, it can quickly find the optimal solution in complex elevator scheduling environments, ensuring optimal resource allocation and improving the scheduling efficiency of the entire elevator group. Attached Figure Description

[0020] Figure 1 A flowchart illustrating the steps of an elevator group control energy-saving scheduling method based on load prediction, provided in one embodiment of this application. Figure 2 A flowchart illustrating the process of obtaining pedestrian flow prediction correction values ​​according to one embodiment of this application. Detailed Implementation

[0021] In the description of the embodiments in this application, the words "exemplary," "or," and "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary," "or," and "for example" is intended to present the relevant concepts in a specific manner.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in this application's specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.

[0023] It should also be noted that the terms "first" and "second" in this application and its accompanying drawings are used to distinguish similar objects, rather than to describe a specific order or sequence. The methods disclosed in the embodiments of this application or the methods shown in the flowcharts include one or more steps for implementing the method. Without departing from the scope of protection of this application, the execution order of multiple steps can be interchanged, and some steps can also be deleted.

[0024] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0025] The following description, in conjunction with the accompanying drawings, details the specific scheme of the elevator group control energy-saving scheduling method and system based on load prediction provided in this application.

[0026] Please see Figure 1 The diagram illustrates a flowchart of an elevator group control energy-saving scheduling method based on load prediction according to an embodiment of this application. The method includes the following steps: The first step: Obtain the operating data of the elevator group.

[0027] Obtain the operational data of the elevator group. The operational data includes: the current number of floors in the building, the number of elevators, the current status of each elevator (floor, operating status: up / down / stationary, floor reached), actual call requests (floor, number of requests, time), and elevator passenger flow load data.

[0028] The current status of the elevator and the actual response data are obtained through the elevator control system; the passenger flow load data of the elevator is obtained through the infrared counter at the elevator entrance. Specifically, passenger flow load refers to the total number of people entering all elevators in the elevator group on a certain floor within a certain period of time.

[0029] For each day, historical operational data for the elevator group is acquired. This historical operational data includes historical actual response requests and historical passenger flow load data for the elevators. The number of days for which historical operational data is acquired is [number missing]. The implementer can set the day according to the implementation scenario without special restrictions. In this embodiment, The value is 45.

[0030] The second step is to use the historical operation data of the elevator group to analyze the periodic distribution characteristics of the actual passenger flow on each floor during the current time period, and the degree of similarity between the actual passenger flow distribution and the historical operation data, so as to determine the predicted passenger flow value for each floor during the current time period.

[0031] In elevator group scheduling and control, traditional scheduling algorithms typically select the optimal elevator for scheduling based on real-time response requests and the current operating status of each elevator, enabling rapid response to scheduling tasks. However, these methods do not fully consider passenger flow load data at different times, thus only initiating elevator scheduling when the elevator control system receives an actual response request. During peak hours, passenger flow is high, while some elevators may be located on floors with lower demand, resulting in longer waiting times for passengers. Furthermore, during off-peak hours, the elevator control system may fail to predict changes in passenger flow, potentially scheduling multiple elevators to operate simultaneously, causing some elevators to run empty, leading to energy waste. Therefore, accurate passenger flow load prediction and the precision of the prediction data are crucial for improving the efficiency and energy-saving effects of elevator group scheduling.

[0032] Traditional methods for predicting pedestrian traffic load typically rely on short-term daily historical data and employ simple time-series averaging. However, these methods fail to adequately consider the significant impact of factors such as building type, weekdays, and working hours on pedestrian traffic patterns, resulting in low accuracy of the prediction models. They struggle to effectively reflect actual pedestrian flow fluctuations and elevator demand, thus affecting the efficiency and energy-saving effects of elevator group scheduling and control. Traditional methods assume that pedestrian traffic distribution follows a fixed pattern across different dates and time periods. In reality, pedestrian traffic patterns vary considerably between different types of buildings (such as office buildings, shopping malls, and entertainment venues). Treating these heterogeneous data indiscriminately can easily lead to predictions that deviate from reality and fail to accurately reflect true pedestrian flow patterns.

[0033] For certain types of buildings, such as office buildings and school buildings, there is a clear and fixed pattern of upward and downward traffic flow during weekdays, while significant differences in distribution characteristics emerge on non-working days. Therefore, this application uses a 7-day period as a cycle and considers the distribution characteristics of passenger flow data with the same week number as the current time in the historical elevator operation data to extract periodic and regular passenger flow characteristics, thereby obtaining statistically significant baseline prediction values.

[0034] Specifically, a day is divided into a preset number of time periods and numbered in chronological order. In this embodiment, the length of each time period is 15 minutes, which can be adjusted by the implementer according to the actual situation. In the historical operation data corresponding to the current time, the average actual flow of people on each floor in the same time period with the same week number as the current time is obtained as the first prediction factor for each floor in the current time period.

[0035] Specifically, in this embodiment, the formula for the first predictor is as follows: , ; In the formula, count represents the counting parameter; D represents the number of days in the historical running data; % represents the modulo operation; d represents the order of the week number in the historical running data that is the same as the current time. This indicates the number of days in the historical operation. Heavenly The first time period The actual number of people on each floor; This indicates the floor function; Indicates the first The first time period The first predictor of each floor. Here, the k-th time period represents the current time period.

[0036] By calculating the average pedestrian flow corresponding to the number of days with the same week number as the current time in historical data, a baseline prediction value reflecting the cyclical pattern of pedestrian flow is obtained. This method effectively eliminates the interference of special factors such as rest days on the prediction results, ensuring that the prediction more closely matches the passenger flow pattern of regular weekdays.

[0037] However, the first predictor relies solely on the average of the same week number in historical operational data, ignoring potential complex variations and uncertainties within that data. When special circumstances cause disruptions, the first predictor fails to reflect the impact of these disruptions on pedestrian flow, leading to a discrepancy between its results and actual demand.

[0038] Furthermore, in order to better quantify the impact of fluctuations in pedestrian traffic under special circumstances on cyclical regularity, the similarity between the first predictor factor and the pedestrian traffic load data in historical operating data is analyzed to obtain a more realistic pedestrian traffic pattern.

[0039] The building's primary predictor and historical passenger flow data are encoded: each row of the encoding matrix corresponds to a time period, and each column corresponds to a floor, resulting in their respective encoding matrices. , ;in, The encoding matrix representing the first predictor of the day; This represents the d-th reference coding matrix in the historical operational data of the day; K represents the number of time periods in a day; Indicates the total number of floors; express The matrix is ​​used to calculate the similarity between the encoding matrix of the first predictor of the day and each of its reference encoding matrices. In this embodiment, the similarity between the two matrices is determined by normalized mutual information. Normalized mutual information is a prior art technique, and its specific process will not be described in detail.

[0040] Calculate the similarity between the coding matrix of the day and each of its reference coding matrices. Use the similarity ratio between the coding matrix of the day and each of its reference coding matrices as the weight of the actual foot traffic of each floor in the current time period of the corresponding historical days. Sum the weighted data to obtain the second prediction factor for each floor in the current time period.

[0041] Specifically, in this embodiment, the formula for the second predictor is as follows: In the formula, This represents the second predictor for the f-th floor in the k-th time period; This represents the similarity between the encoding matrix of the first predictor of the fth floor in the kth time period and the dth reference encoding matrix; This indicates the number of days in the historical data. This indicates the number of days in the historical operation. Heavenly The first time period The actual number of people on each floor.

[0042] It should be understood that, in In the diagram, the numerator represents the encoding matrix of the first predictor. and the Encoding matrix of historical operation data The similarity between them, where the denominator represents the sum of similarities between the encoding matrix of the first predictor and the encoding matrices of the passenger flow load data for all historical operating data for all days, can be represented by this linear normalization method. and The similarity between them The similarity ratio between the traffic flow data and the complex data encoding matrix of all historical operational data is calculated, and a normalized similarity weight corresponding to the number of days is assigned to the traffic flow load data of each day's historical operational data. The larger the value, the higher the value. Human traffic patterns in historical operational data and The more similar the pedestrian flow patterns, the higher the weight will be given to the historical pedestrian flow load data for that day. This will help adjust the prediction results to more accurately reflect the actual situation. By introducing diverse historical operating patterns, the predictive results are made more adaptable to interference from special circumstances.

[0043] So, the first The first time period The predicted pedestrian flow for a floor can be expressed as: in, , They are respectively , The preset weight parameters satisfy Its size can be set by the implementer according to the implementation scenario, without special restrictions. In this embodiment, , The values ​​are 0.6 and 0.4 respectively; This indicates rounding to the nearest integer.

[0044] The third step is to preset reference correction time periods for each time period, compare the overall change characteristics of the difference between the predicted and actual pedestrian flow in all reference correction time periods of the current time period, determine the pedestrian flow prediction error for each floor in the current time period, correct the pedestrian flow prediction value, and obtain the corrected pedestrian flow prediction value for each floor in the current time period.

[0045] The first and second predictive factors predict pedestrian flow from the perspectives of periodic regularity and similarity of pedestrian flow patterns, respectively. In order to further improve the accuracy of the pedestrian flow prediction correction value, the pedestrian flow prediction correction value is corrected in real time by analyzing the error difference between the actual pedestrian flow pattern before the day and the pedestrian flow prediction value.

[0046] Specifically, the first day A period of time before Using a time period as a reference correction period for the k-th time period of the day, we analyze the prediction error of pedestrian flow at each level and construct an error correction term. The size can be set by the implementer according to the implementation scenario, without special restrictions. In this embodiment... The value is 4; where the reference correction period is the same day as the current period.

[0047] Furthermore, for each floor in the current time period, the difference between the predicted and actual pedestrian flow for each reference correction time period is calculated, and positive fusion is performed with the negative correlation mapping of the predicted pedestrian flow. The average of the positive fusion results obtained from all reference correction time periods is calculated to obtain the pedestrian flow prediction error for each floor in the current time period. In this embodiment, the difference between two variables is calculated using the difference value, and the negative correlation mapping result of the variables is calculated using the reciprocal of the variables. It should be noted that to prevent the denominator from being 0, a preset value needs to be added to the denominator. In this embodiment, the preset value is 0.01. The positive fusion of multiple variables is performed using a multiplication calculation method. The formula for calculating the pedestrian flow prediction error of the f-th floor in the k-th time period of the day is as follows: In the formula, Indicates the first The first time period Error in predicting pedestrian traffic on each floor Indicates the first The t-th reference correction time period of the t-th time period Predicted pedestrian traffic for each floor Indicates the first The t-th reference correction time period of the t-th time period The actual foot traffic on each floor This represents the preset value, which is 0.01; T represents the first... The number of reference correction time periods for each time period.

[0048] Special, when When the value is less than or equal to 4, calculate the number of historical running data. The first time period before all time periods The average actual foot traffic on each floor and Perform error calculation; If so, then there were only two time periods with actual pedestrian traffic previously. Then replace To calculate .

[0049] Finally, the specific method for correcting the predicted pedestrian flow for each floor in each time period is as follows: Calculate the difference between the natural number 1 and the predicted pedestrian flow error for each floor in each time period, and then perform a forward fusion with the predicted pedestrian flow value obtained for each floor in each time period to obtain the corrected pedestrian flow prediction value for each floor in each time period. In this embodiment, the forward fusion of multiple variables uses a multiplication calculation method. The flowchart for obtaining the corrected pedestrian flow prediction value is shown below. Figure 2 As shown.

[0050] When the pedestrian flow prediction error is negative, it indicates that the predicted pedestrian flow in the preceding period was less than the actual pedestrian flow. To avoid error accumulation leading to inaccurate predictions, the difference between the natural number 1 and the pedestrian flow prediction error is used to increase the pedestrian flow prediction correction value for that period. Conversely, when the pedestrian flow prediction error is positive, it indicates that the predicted pedestrian flow in the preceding period was higher than the actual pedestrian flow. In this case, the pedestrian flow prediction correction value for that period is decreased. By continuously comparing the predicted pedestrian flow with the actual pedestrian flow, the prediction results can be adjusted in real time, thereby reducing deviation. This dynamic correction method allows the prediction to be continuously optimized according to changes in the scenario, thus improving the accuracy of pedestrian flow prediction.

[0051] The fourth step: Based on the predicted passenger flow correction value and the actual load capacity of the elevator, determine the elevator's load availability, and based on the predicted passenger flow correction values ​​of all floors in the current time period, adjust the fixed inertia weights of the particle swarm algorithm to find the optimal solution for response scheduling.

[0052] Furthermore, elevator group control and scheduling are performed based on the predicted and corrected passenger flow data and the elevator group's operational data. The specific steps are as follows: For the first The predicted passenger flow for each floor within a given time period is sorted from largest to smallest. A virtual response request is generated in the elevator control system, and elevators are dispatched to floors with higher passenger flow to wait. Elevators responding to the virtual response request do not include those that are already in operation. The purpose is to reduce users' waiting time by scheduling elevators in advance during peak hours.

[0053] Furthermore, the predicted passenger flow for each floor is compared with the actual elevator load capacity. ,in This indicates the elevator's actual load capacity. If the elevator control system does not trigger a virtual response request for that floor, it will remain in standby mode; otherwise, it will trigger a virtual response request for that floor. The purpose is to allow elevators to respond to actual response requests and be scheduled when passenger flow is predicted to be low on multiple floors, keeping some elevators in standby mode to avoid frequent empty runs and wasting resources, thus improving the energy efficiency of the elevator system.

[0054] Furthermore, for the actual response requests of the elevator system, an adaptive inertial weight is constructed based on the predicted correction value of passenger flow. Specifically: for the current time period, the total number of predicted correction values ​​of passenger flow for all floors is calculated; the average actual passenger flow of all floors in the same time period as the current time is calculated from the historical operation data; the ratio between the total number and the average value is calculated, and then multiplied by the preset fixed weight to obtain the adaptive inertial weight.

[0055] In this embodiment, the specific formula for the adaptive inertia weight is as follows: ;in, This represents the total number of predicted and corrected pedestrian traffic values ​​for all floors during the current time period. Indicates the first in the historical operation data Average total number of people on all floors over a given time period The inertial weights represent the particle swarm optimization algorithm. The method for setting these weights is a well-known technique and will not be elaborated further. Indicates in The predicted correction value of passenger flow within the specified time period is relative to the historical data. The magnitude of the average total number of people within a given time period.

[0056] Furthermore, a larger value indicates excessive current foot traffic, meaning there is more foot traffic compared to historical data for this time period. To improve the global exploration capability of the particle swarm optimization algorithm, the exploration space is increased to find the optimal solution under high traffic load conditions; when Furthermore, a smaller value indicates lower current foot traffic, and even lower foot traffic compared to historical data for this time period, thus reducing the likelihood of such a decrease. The particle swarm optimization (PSO) algorithm tends to perform localized, fine-grained searches to quickly find the optimal solution for response scheduling. It dynamically adjusts the search strategy based on passenger flow intensity, thereby achieving adaptive exploration of elevator scheduling schemes and improving the scheduling efficiency and energy-saving effect of the elevator control system. The PSO algorithm used for elevator scheduling is a well-known existing technology and will not be elaborated upon in this application.

[0057] Based on the same inventive concept as the above methods, this application also provides an elevator group control energy-saving scheduling system based on load prediction, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any of the above methods.

[0058] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the description; sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0059] It will be apparent to those skilled in the art that this application is not limited to the details of the exemplary embodiments described above, and that this application can be implemented in other specific forms without departing from its essential characteristics. Therefore, the embodiments described above should be considered exemplary and non-limiting in all respects; modifications to the technical solutions described in the foregoing embodiments, or equivalent substitutions of some technical features, without causing the essence of the corresponding technical solutions to deviate from the scope of the technical solutions in the embodiments of this application, should all be included within the protection scope of this application.

Claims

1. A load prediction based energy saving dispatching method for elevator group control, characterized in that, The method includes the following steps: By using historical operation data of the elevator group, we can analyze the periodic distribution characteristics of the actual passenger flow on each floor during the current time period and the degree of similarity between the distribution of actual passenger flow and the historical operation data, and determine the predicted value of passenger flow on each floor during the current time period. Preset reference correction time periods for each time period, compare the overall change characteristics of the difference between the predicted and actual pedestrian flow in all reference correction time periods of the current time period, determine the pedestrian flow prediction error for each floor in the current time period, correct the pedestrian flow prediction value, and obtain the pedestrian flow prediction correction value for each floor in the current time period. Based on the predicted passenger flow correction value and the actual load capacity of the elevator, the elevator's load availability is determined. Based on the predicted passenger flow correction values ​​of all floors in the current time period, the fixed inertia weight of the particle swarm algorithm is adjusted to find the optimal solution for response scheduling.

2. The load prediction based energy saving dispatching method of elevator group control according to claim 1, characterized in that, The process of determining the predicted pedestrian flow for each floor in the current time period is as follows: Based on the periodic distribution characteristics, determine the first prediction factor for each floor in the current time period; Analyze the similarity between the first predictor and the distribution of actual pedestrian traffic in historical operational data to determine the second predictor for each floor in the current time period; For each time period and each floor, the first and second predictive factors are weighted respectively, and the weighted sum is rounded down to obtain the predicted pedestrian flow value.

3. The load prediction based energy saving dispatching method of elevator group control according to claim 2, wherein, The determination of the first prediction factor for each floor in the current time period is specifically as follows: In the historical data corresponding to the current moment, the average actual pedestrian flow of each floor during the same time period with the same week number as the current moment is obtained, and used as the first predictor for each floor during the current time period.

4. The load prediction based energy saving scheduling method of elevator group control according to claim 2, wherein, The determination of the second prediction factor for each floor in the current time period is specifically as follows: Encode all the first predictor factors obtained on the same day to obtain the encoding matrix; Encode all actual passenger flow for each day in the historical operational data of the day, and obtain each reference coding matrix corresponding to the current moment; Calculate the similarity between the coding matrix of the day and each of its reference coding matrices. Use the similarity ratio between the coding matrix of the day and each of its reference coding matrices as the weight of the actual foot traffic of each floor in the current time period of the corresponding historical days. Sum the weighted data to obtain the second prediction factor for each floor in the current time period.

5. The load prediction based energy saving dispatching method of elevator group control according to claim 4, wherein, Each row of the encoding matrix corresponds to a time period of the day, and each column corresponds to a floor.

6. The load prediction based energy saving scheduling method of elevator group control according to claim 1, wherein, The specific process for determining the predicted pedestrian flow error for each floor during the current time period is as follows: For each floor in the current time period, calculate the difference between the predicted and actual pedestrian flow for each reference correction time period, and perform positive fusion with the negative correlation mapping of the predicted pedestrian flow. Average the positive fusion results obtained from all reference correction time periods to obtain the pedestrian flow prediction error for each floor in the current time period.

7. The load prediction based energy saving scheduling method of elevator group control according to claim 1, wherein, The specific process for obtaining the predicted and corrected pedestrian flow values ​​for each floor during the current time period is as follows: The difference between the natural number 1 and the predicted passenger flow error of each floor in each time period is calculated, and then the predicted passenger flow value of each floor in each time period is positively fused to obtain a predicted passenger flow correction value of each floor in each time period.

8. The load prediction based energy saving scheduling method of elevator group control according to claim 1, wherein, The load empty condition of the elevator is determined, and specifically, For the ratio of the predicted passenger flow correction value of each floor in the current time period to the actual load capacity of the elevator, when the ratio is greater than or equal to a preset threshold, a virtual call request is sent for the corresponding floor; otherwise, no triggering is performed.

9. The load prediction based energy saving scheduling method of elevator group control according to claim 1, wherein, The fixed inertia weight of the particle swarm algorithm is adjusted, and specifically, If the elevator sends a virtual call request, for the current time period, the total number of predicted passenger flow correction values of all floors is calculated; the average value of the actual passenger flow of all floors in the same time period as the current time in the historical operation data is calculated; The ratio between the total number and the average value is calculated, and then multiplied by a preset fixed weight to obtain an adaptive inertia weight.

10. An elevator group control energy-saving scheduling system based on load prediction, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, The processor executes the computer program to implement the steps of the method of any one of claims 1-9.

Images