Intelligent building management monitoring system based on Internet of Things

Abstract

The invention relates to the field of building management monitoring, and particularly discloses an intelligent building management monitoring system based on the Internet of Things, which comprises the following steps: firstly, acquiring a floor image of an elevator to be taken acquired by a camera, historical elevator use frequency text data acquired by a database and elevator operation data at a plurality of preset time points; and then feature extraction and correlation analysis are carried out on the three by utilizing a deep learning technology, and finally an elevator scheduling scheme is generated through a generator, so that the elevator operation efficiency and the energy utilization rate are improved, the user experience can be optimized, and the intelligent level of management decision making can be improved.

Description

Technical Field

[0001] The present application relates to the field of building management and monitoring, and more specifically, to an intelligent building management and monitoring system based on the Internet of Things. Background Art

[0002] Intelligent buildings are organically integrated through communication network systems, integrating structures, systems, services, management and their optimal combination, making buildings safe, convenient, efficient and energy-saving. Intelligent buildings are a marginal and interdisciplinary subject involving computer technology, automatic control, communication technology, construction technology, etc., and more and more new technologies are being applied in intelligent buildings.

[0003] As the number of floors in buildings increases, elevators are an indispensable device for high-rise buildings. They can transport passengers or goods quickly between floors. When using, passengers need to press buttons to select the floor they want to reach, and the elevator control terminal controls the car to reach the floor specified by the passenger. In addition, in order to meet the passenger's riding needs during rush hours, multiple elevators are usually installed in buildings. However, except for rush hours, multiple elevators are always in operation at other times, and only sporadic passengers will occasionally take the elevator during idle time outside the peak hours. Only one or two elevators are needed to fully meet the passengers' demand for elevators during idle time. Therefore, the current elevator management method in which each elevator is always in operation is extremely power-consuming.

[0004] Therefore, an intelligent building management and monitoring system based on the Internet of Things is desired. Summary of the invention

[0005] In order to solve the above technical problems, this application is proposed. The embodiment of this application provides an intelligent building management and monitoring system based on the Internet of Things, which first obtains the image of the elevator floor to be taken collected by the camera, the historical elevator usage frequency text data collected by the database, and the elevator operation data at multiple predetermined time points, and then uses deep learning technology to perform feature extraction and correlation analysis on the three, and finally generates an elevator scheduling plan through a generator, thereby improving the efficiency and energy utilization of elevator operation, and can also optimize user experience and improve the intelligence level of management decision-making.

[0006] According to one aspect of the present application, there is provided an intelligent building management and monitoring system based on the Internet of Things, which includes:

[0007] The intelligent building elevator data acquisition module is used to obtain the image of the elevator floor to be taken collected by the camera, the historical elevator usage frequency text data collected by the database, and the elevator operation data at multiple predetermined time points;

[0008] The intelligent building elevator data extraction module is used to extract the image block association feature vector of the elevator floor area to be taken and the elevator operation efficiency feature vector from the elevator floor image to be taken collected by the camera, the historical elevator usage frequency text data collected by the database, and the elevator operation data at the plurality of predetermined time points;

[0009] The elevator dispatching scheme generating module is used to generate an elevator dispatching scheme based on the associated feature vector of the image block of the floor area of ​​the elevator to be taken and the feature vector of the elevator operation efficiency.

[0010] Compared with the prior art, the present application provides an intelligent building management and monitoring system based on the Internet of Things, which first obtains the image of the elevator floor to be taken collected by the camera, the historical elevator usage frequency text data collected by the database, and the elevator operation data at multiple predetermined time points, and then uses deep learning technology to perform feature extraction and correlation analysis on the three, and finally generates an elevator scheduling plan through a generator, thereby improving the efficiency and energy utilization of the elevator operation, and can also optimize the user experience and enhance the intelligence level of management decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0012] Figure 1 The block diagram is a smart building management and monitoring system based on the Internet of Things according to an embodiment of the present application.

[0013] Figure 2 The present invention is a block diagram of an intelligent building elevator data extraction module in an intelligent building management and monitoring system based on the Internet of Things according to an embodiment of the present application.

[0014] Figure 3 The present invention is a block diagram of an elevator floor image feature extraction unit in an IoT-based intelligent building management and monitoring system according to an embodiment of the present application.

[0015] Figure 4 The present invention is a block diagram of an elevator dispatching scheme generating module in an intelligent building management and monitoring system based on the Internet of Things according to an embodiment of the present application.

[0016] Figure 5 is a block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0017] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0018] Figure 1 FIG. 1 is a block diagram of an intelligent building management and monitoring system based on the Internet of Things according to an embodiment of the present application. Figure 1 As shown, according to the embodiment of the present application, the smart building management and monitoring system 100 based on the Internet of Things includes: a smart building elevator data acquisition module 110, which is used to obtain the image of the elevator floor to be taken collected by the camera, the historical elevator usage frequency text data collected by the database, and the elevator operation data at multiple predetermined time points; a smart building elevator data extraction module 120, which is used to extract the image block association feature vector of the elevator floor area to be taken and the elevator operation efficiency feature vector from the elevator floor image to be taken collected by the camera, the historical elevator usage frequency text data collected by the database, and the elevator operation data at multiple predetermined time points; an elevator scheduling plan generation module 130, which is used to generate an elevator scheduling plan based on the image block association feature vector of the elevator floor area to be taken and the elevator operation efficiency feature vector.

[0019] In the above-mentioned intelligent building management and monitoring system 100 based on the Internet of Things, the intelligent building elevator data acquisition module 110 is used to obtain the image of the elevator floor to be taken collected by the camera, the historical elevator usage frequency text data collected by the database, and the elevator operation data at multiple predetermined time points. It should be understood that the intelligent building optimizes and combines various aspects such as building structure, facility system, and service management through an integrated communication network system, so as to achieve safe, convenient, efficient and energy-saving operation of the building. This is a multidisciplinary field involving computer technology, automatic control, communication technology, and construction engineering, and new technologies are constantly being integrated to enhance the functions and benefits of intelligent buildings. With the increase in building height, elevators, as indispensable equipment, play an important role in quickly transporting passengers or goods to and from various floors. In daily use, passengers select the target floor by pressing buttons, and the elevator control system is responsible for dispatching the car to the corresponding floor according to passenger needs. In order to cope with the needs of rush hour commuting, multiple elevators are usually installed in the building. However, except for peak hours, these elevators are often in operation for a long time. Even if only a small number of passengers occasionally need to use the elevator during low peak hours, multiple elevators are required to continue to operate. This elevator management method leads to a large amount of waste of electric energy. Therefore, in the technical solution of this application, by acquiring the image of the elevator floor to be taken by the camera, the historical elevator usage frequency text data collected by the database and the elevator operation data at multiple predetermined time points, and combining deep learning technology, an elevator scheduling plan is generated to enhance the operation efficiency and energy efficiency of the elevator, improve the user experience, and enhance the intelligence level of management decision-making.

[0020] Specifically, the images of the elevator floors to be taken collected by the camera can provide real-time information on elevator demand. By analyzing these images, the system can identify the number of people waiting to take the elevator in each elevator lobby, their locations, and the destination floors. This information is crucial for formulating real-time elevator scheduling plans, which can reduce passenger waiting time and improve the operating efficiency of elevators. The historical elevator usage frequency text data records the usage of each elevator over a period of time. These data can include the peak and trough periods of the elevator, as well as the frequency of use in each time period. The elevator operation data at multiple predetermined time points provides detailed operating status of the elevator in different time periods. These data can include the elevator's operating speed, number of operations, fault conditions, etc. By analyzing this data, the system can identify the elevator's operating bottlenecks and potential problems, and take preventive maintenance measures to ensure the safety and reliability of the elevator. In this way, not only can the intelligent scheduling and optimization of elevator operation be achieved, but also the efficiency and service quality of building management can be improved, providing residents and visitors with a more convenient and safe elevator experience.

[0021] In the above-mentioned intelligent building management and monitoring system 100 based on the Internet of Things, the intelligent building elevator data extraction module 120 is used to extract the associated feature vector of the image block of the floor area of ​​the elevator to be taken and the elevator operation efficiency feature vector from the image of the elevator floor to be taken collected by the camera, the historical elevator usage frequency text data collected by the database, and the elevator operation data at multiple predetermined time points. It should be understood that extracting the associated feature vector of the image block of the floor area of ​​the elevator to be taken and the elevator operation efficiency feature vector from the image collected by the camera, the historical usage frequency text data, and the operation data at multiple time points can not only help the system realize intelligent elevator scheduling and optimization management, but also improve the quality and efficiency of elevator services, and provide users with a more convenient and comfortable travel experience. The extraction and analysis process of these feature vectors is an important technical support in the intelligent building management system, which directly affects the overall operation efficiency and management effect.

[0022] Figure 2 FIG. 1 is a block diagram of an intelligent building elevator data extraction module in an intelligent building management and monitoring system based on the Internet of Things according to an embodiment of the present application. Figure 2 As shown, in a specific embodiment of the present application, the intelligent building elevator data extraction module 120 includes: an elevator floor image feature extraction unit 121, which is used to perform feature extraction on the image of the elevator floor to be taken collected by the camera to obtain a feature vector associated with the image block of the elevator floor area to be taken; a historical elevator usage frequency text feature extraction unit 122, which is used to perform feature extraction on the historical elevator usage frequency text data collected by the database to obtain a historical elevator usage frequency semantic association feature vector; an elevator operation data feature extraction unit 123, which is used to perform feature extraction on the elevator operation data of the multiple predetermined time points to obtain an elevator operation data time series feature vector; an elevator operation efficiency feature fusion unit 124, which is used to fuse the historical elevator usage frequency semantic association feature vector and the elevator operation data time series feature vector to obtain the elevator operation efficiency feature vector.

[0023] It should be understood that through feature extraction, the system can extract key information and features from the image. These features can include the location distribution of passengers, the number of people waiting to ride, and the behavior patterns of passengers. For example, the system can identify the density of people in each area of ​​the elevator lobby, the queue situation of passengers, and the distribution of their destinations. This information is crucial for elevator scheduling, which can help the system predict and respond to passenger needs, thereby reducing waiting time and improving service efficiency. By analyzing the elevator lobby images in different time periods, the system can understand the changing patterns and trends of passenger needs, and then optimize the elevator scheduling strategy.

[0024] Furthermore, the historical elevator usage frequency text data records the usage of the elevator in the past period of time, including usage frequency within a daily, weekly or longer period, peak hours, changes under specific events, etc. Through feature extraction, the system can extract key semantic information from these text data, such as usage changes in different time periods, common usage patterns (such as the difference between office hours and non-office hours), preferences for specific floors, etc. The extraction of semantically associated feature vectors of historical elevator usage frequency helps to establish a model of elevator usage behavior. These feature vectors may include information such as average usage frequency, changes in peak and trough periods, and usage preferences for specific floors. Through these models, the system can predict future usage needs and adjust the elevator scheduling strategy to adapt to different time periods and demand changes.

[0025] Furthermore, feature extraction is performed on the elevator operation data at multiple predetermined time points, aiming to reveal the changes in the operation status, performance trends and potential problems of the elevator in different time periods through in-depth analysis and pattern recognition, so as to support refined operation management and optimization decisions. The extraction of time series feature vectors of elevator operation data involves extracting key time series information from the operation data collected at multiple time points. These data include the daily operation time, operation speed, number of operations, fault records, etc. of the elevator. Through feature extraction, the system can establish a time series model of elevator operation and analyze the operation mode and regularity of the elevator in different time periods, such as the operation frequency during the day and night, the operation load during peak hours, etc. The extraction of time series feature vectors helps to predict the future operation demand and failure risk of the elevator. By analyzing historical data, the system can identify periodic changes and trends in elevator operation, such as the reasons for the significant increase in elevator operation load in certain time periods, or the upward trend of elevator failure rate under specific conditions. This information is crucial for formulating preventive maintenance plans and adjusting operation strategies, and can provide early warning and respond to potential problems to ensure the safe and reliable operation of elevators.

[0026] In particular, the fusion of the semantic association feature vector of the historical elevator usage frequency and the time series feature vector of the elevator operation data is to comprehensively analyze the usage pattern and operation status of the elevator, so as to comprehensively evaluate and optimize the operation efficiency of the elevator and improve the service quality and user experience. The semantic association feature vector of the historical elevator usage frequency contains semantic information and usage patterns about the elevator. These feature vectors reflect the usage frequency of the elevator in different time periods, the changes in peak and trough periods, the preference for specific floors, etc. The time series feature vector of the elevator operation data provides detailed data during the actual operation of the elevator, such as daily operation time, operation speed, fault records, etc. These time series feature vectors can reveal specific problems and trends in the operation of the elevator, such as the peak period of the operating load, the types of faults that occur frequently, etc. The fusion of the semantic association feature vector of the historical elevator usage frequency and the time series feature vector of the elevator operation data to obtain the elevator operation efficiency feature vector can not only comprehensively evaluate and optimize the operation management of the elevator, but also improve the quality of elevator service and user experience. It is an important step and technical support for realizing intelligent building management.

[0027] Figure 3 FIG. 1 is a block diagram of an elevator floor image feature extraction unit in an IoT-based intelligent building management and monitoring system according to an embodiment of the present application. Figure 3 As shown, in a specific embodiment of the present application, the elevator floor image feature extraction unit 121 includes: an elevator floor image feature encoding subunit 1211, used to perform feature encoding on the image of the elevator floor to be taken collected by the camera to obtain multiple feature vectors of the image blocks of the elevator floor area to be taken; an elevator floor image feature construction subunit 1212, used to construct the multiple feature vectors of the image blocks of the elevator floor area to be taken into the associated feature vector of the image blocks of the elevator floor area to be taken.

[0028] It should be understood that feature encoding can transform complex elevator floor images into compact and representative feature vectors. These feature vectors can include abstract numerical descriptions of the structure, color, texture, etc. in the image, rather than simple pixel information. Through this encoding, the system can extract key visual features from the image, such as the density of people in different areas of the elevator hall, queuing conditions, and dynamic changes of pedestrians. The generation of feature vectors of multiple image blocks of the floor area to be taken by the elevator helps to achieve refined allocation and scheduling of elevator resources.

[0029] Furthermore, the feature vectors of multiple floor area image blocks of the elevator to be taken are constructed into associated feature vectors of the floor area image blocks of the elevator to be taken, aiming to achieve more accurate and comprehensive elevator lobby management and operation optimization by integrating and synthesizing the feature information of different area image blocks. The feature vectors of the floor area image blocks of the elevator to be taken usually reflect the visual characteristics and usage status of each area in the elevator lobby. These feature vectors may include information such as crowd density, movement trajectory, and spatial distribution. By constructing and integrating the feature vectors of multiple area image blocks, the system can establish a more comprehensive elevator lobby state model and accurately capture the association and influence between different areas.

[0030] In a specific embodiment of the present application, the elevator floor image feature encoding subunit 1211 includes: performing image block processing on the image of the elevator floor to be taken collected by the camera to obtain a sequence of image blocks of the elevator floor area to be taken; passing the sequence of image blocks of the elevator floor area to be taken through a convolutional neural network model of the elevator floor area to be taken as a filter to obtain the multiple feature vectors of the elevator floor area to be taken.

[0031] It should be understood that image block processing can decompose the image of the entire elevator floor into multiple small blocks, each of which represents a local area in the elevator lobby. These regional image blocks can be divided according to the structure and layout of the elevator lobby, such as the elevator door, the waiting area, the floor indicator, etc. Through block processing, the system can more finely analyze the information such as the flow density, waiting situation, and movement trajectory of each area, so as to more accurately understand the real-time status of the elevator lobby. The generation of the image block sequence of the floor area to be taken by the elevator helps to realize the real-time monitoring and analysis of the dynamic changes of the elevator lobby. Over time, the flow of people and usage patterns in the elevator lobby will change, such as the peak flow of people during the day and at night. By continuously collecting and analyzing the regional image block sequence, the system can identify these changes and adjust the elevator operation strategy accordingly to optimize the dispatch and service efficiency of the elevator. Image block processing can also simplify the processing and storage of complex image data. Compared with processing the entire large image, block processing can reduce computational complexity and resource consumption, and improve the operation efficiency of the system. The feature analysis of each small block image can be performed locally and combined with the analysis results of other areas to form a comprehensive understanding of the global situation of the elevator lobby.

[0032] Furthermore, convolutional neural networks are a type of deep learning model that is particularly suitable for processing image data. Through multi-layer convolution and pooling operations, convolutional neural networks can effectively capture local features and global structures in images and extract abstract feature information from images. Applying convolutional neural networks as filters to the image block sequences of the floor areas to be ridden in the elevator can extract and represent features for each area image block while retaining the image spatial information, thereby forming a feature vector. The convolutional neural network processes the image block sequences of the floor areas to be ridden in the elevator as a filter, and can automatically learn and extract important features suitable for elevator lobby management. These features can include crowd density, motion trajectory, spatial distribution, etc., which are crucial for elevator scheduling and resource optimization. Specifically, each layer of the convolutional neural network model of the image block of the floor area of ​​the elevator to be taken as a filter performs convolution processing, mean pooling processing based on the local feature matrix and nonlinear activation processing on the input data in the forward pass of the layer, so that the last layer of the convolutional neural network model of the image block of the floor area of ​​the elevator to be taken as a filter outputs the multiple feature vectors of the image block of the floor area of ​​the elevator to be taken, wherein the input of the convolutional neural network model of the image block of the floor area of ​​the elevator to be taken as a filter is the sequence of image blocks of the floor area of ​​the elevator to be taken.

[0033] In a specific embodiment of the present application, the historical elevator usage frequency text feature extraction unit 122 includes: passing the historical elevator usage frequency text data collected from the database through a converter-based historical elevator usage frequency text semantic encoder to obtain multiple historical elevator usage frequency text semantic feature vectors; cascading the multiple historical elevator usage frequency text semantic feature vectors into the historical elevator usage frequency semantic association feature vector.

[0034] It should be understood that the historical elevator usage frequency text data usually contains information such as the time, floor, and waiting time of passengers using the elevator. These data are huge and complex, and it may be difficult to extract the potential patterns and associations in them by direct use. Through a semantic encoder based on a transformer (such as BERT, GPT, etc.), the text data can be converted into a dense vector representation in the semantic space, that is, a semantic feature vector. The transformer-based semantic encoder can handle different types of text data, including time series and location descriptions in elevator usage records. Through pre-training and fine-tuning, this model can understand and represent complex semantics in text data, such as understanding the peak and low peak elevator usage patterns in different time periods, usage trends between different floors, etc. By generating semantic feature vectors, the system can analyze the patterns and trends in historical data more deeply, providing data support and decision-making basis for elevator management. Semantic feature vectors also facilitate semantic search and similarity comparison of historical data. The transformer-based encoder can map text data to vector representations in the semantic space, allowing the system to quickly find and identify historical cases or patterns similar to the current situation. Specifically, the historical elevator usage frequency text data collected by the database is segmented to obtain a word sequence; the embedding layer of the converter-based historical elevator usage frequency text semantic encoder is used to map each word in the word sequence into a word embedding vector to obtain a sequence of word embedding vectors; the converter-based Bert model of the converter-based historical elevator usage frequency text semantic encoder is used to perform global context semantic encoding on the sequence of embedding vectors to obtain multiple historical elevator usage frequency text semantic feature vectors.

[0035] Furthermore, multiple historical elevator usage frequency text semantic feature vectors are cascaded into historical elevator usage frequency semantic association feature vectors, aiming to more comprehensively analyze and understand the long-term trends and patterns of elevator usage by integrating and synthesizing semantic information of different time periods or different dimensions, so as to support more accurate elevator management decisions and service optimization. The historical elevator usage frequency text semantic feature vectors reflect the semantic characteristics of elevator usage in different time periods or different situations. These feature vectors can include usage patterns during peak and off-peak periods, usage preferences of different floors, usage in specific time periods, etc. By cascading feature vectors of multiple historical time periods or different dimensions, the system can establish a more comprehensive and comprehensive historical elevator usage model that can capture the long-term changes and trends in elevator usage. Cascading historical elevator usage frequency text semantic feature vectors helps to analyze and predict the global dynamics of elevator usage. The usage pattern of elevators is often affected by seasonality, regionality, and crowd behavior. By integrating feature vectors of different time periods or dimensions, the system can identify periodic changes and trends in elevator usage, such as usage patterns in different time periods such as holidays and working days. This comprehensive analysis helps the system to more accurately predict future elevator demand, thereby optimizing elevator scheduling strategies and resource allocation.

[0036] In a specific embodiment of the present application, the elevator operation data feature extraction unit 123 includes: arranging the elevator operation data of the multiple predetermined time points into an elevator operation data timing input vector; passing the elevator operation data timing input vector through an elevator operation data timing feature extractor to obtain the elevator operation data timing feature vector.

[0037] It should be understood that the elevator operation data time series input vector can capture the continuity and trend of the elevator operation status over time. Elevator operation data usually includes information on multiple dimensions such as elevator operation time, operation speed, and passenger capacity. By arranging these data into time series vectors according to time points, the system can analyze the changing patterns of elevator operation status in different time periods. This time series analysis capability is of great significance for identifying abnormal behavior in elevator operation and predicting the arrival of peak and off-peak periods.

[0038] Furthermore, the elevator operation data time series input vector is extracted by the elevator operation data time series feature extractor, aiming to extract key feature information from complex time series data through efficient feature extraction and representation, so as to support more accurate and reliable analysis, prediction and optimization of elevator operation status. The elevator operation data time series input vector contains multi-dimensional data of the elevator at different time points, such as operation time, passenger capacity, operation speed, etc. These data are continuous and dynamically changing in time, and direct use may face the challenges of high dimension and high noise. Through the elevator operation data time series feature extractor, the system can automatically extract representative and distinguishing feature vectors from the time series data, such as statistical features (mean, variance), frequency domain features (Fourier transform correlation), time domain features (autocorrelation function), etc. The elevator operation data time series feature extractor can reduce data dimension and noise, and improve the representation ability and stability of data. By reasonably selecting and designing feature extraction methods, irrelevant information can be effectively filtered out and the representation ability of elevator operation data can be enhanced. This feature extraction capability is particularly important for processing large-scale, high-dimensional time series data, which can reduce the complexity and cost of data processing and improve the efficiency and accuracy of data analysis. Specifically, the elevator operation data time series input vector is fully connected using the fully connected layer of the elevator operation data time series feature extractor to extract the high-dimensional implicit features of the feature values ​​at each position in the elevator operation data time series input vector; and the elevator operation data time series input vector is one-dimensionally encoded using the one-dimensional convolutional layer of the elevator operation data time series feature extractor to extract the high-dimensional implicit correlation features of the correlation between the feature values ​​at each position in the elevator operation data time series input vector.

[0039] In the above-mentioned intelligent building management and monitoring system 100 based on the Internet of Things, the elevator scheduling scheme generation module 130 is used to generate an elevator scheduling scheme based on the associated feature vector of the image block of the floor area of ​​the elevator to be taken and the elevator operation efficiency feature vector. It should be understood that the associated feature vector of the image block of the floor area of ​​the elevator to be taken provides information about passengers waiting for the elevator on different floors. These image blocks usually contain data such as the location, number and waiting time of the passengers waiting. Through the associated feature vector, the system can quantify and analyze the passenger demand on each floor, understand the waiting situation during peak and trough periods, and provide accurate demand forecasts for elevator scheduling. The elevator operation efficiency feature vector reflects the performance and efficiency of the elevator during operation. These features include the elevator's operating speed, average passenger capacity, operating stability, etc. By analyzing the elevator's operating efficiency, the system can evaluate the actual operation of the elevator and the potential optimization space, and provide real-time operating status feedback and support for scheduling decisions. Generating an elevator scheduling scheme based on the associated feature vector of the image block of the floor area of ​​the elevator to be taken and the elevator operation efficiency feature vector can not only improve the intelligence level and operating efficiency of elevator scheduling, but also optimize the waiting experience of passengers and the overall service quality of the system.

[0040] Figure 4 FIG. 1 is a block diagram of an elevator dispatching scheme generation module in an intelligent building management and monitoring system based on the Internet of Things according to an embodiment of the present application. Figure 4 As shown, in a specific embodiment of the present application, the elevator scheduling scheme generation module 130 includes: an elevator scheduling data feature association unit 131, which is used to associate the associated feature vector of the image block of the floor area of ​​the elevator to be taken with the elevator operation efficiency feature vector to obtain an elevator scheduling generation feature vector; an elevator scheduling data feature association unit 132, which is used to perform probability guidance calibration based on back propagation compensation on the elevator scheduling generation feature vector to obtain an optimized elevator scheduling generation feature vector; a scheduling scheme generation unit 133, which is used to pass the optimized elevator scheduling generation feature vector through a generator to generate an elevator scheduling scheme.

[0041] It should be understood that by combining the spatial distribution characteristics of passengers with the operating efficiency characteristics of elevators, the system can comprehensively analyze the waiting conditions of passengers on different floors and the operating capacity of elevators. This comprehensive analysis capability helps the system to more accurately predict and identify the elevator demand during peak and trough periods, as well as the operating bottlenecks or optimization space that the elevator system may face.

[0042] In particular, in the technical solution of the present application, the image of the floor of the elevator to be taken, the text data of the historical elevator usage frequency, and the elevator operation data are derived from different data sources. Among them, since the feature extraction process of the image block of the floor area of ​​the elevator to be taken depends on the convolutional neural network, the quality of its feature vector is affected by the image segmentation and feature extraction accuracy. If the image segmentation is uneven or the convolutional neural network model is not trained enough, the feature vector may not be consistent in some areas. The semantic encoding of the historical elevator usage frequency text data may also be affected by the data quality or encoder performance, resulting in deviations in the semantic feature extraction of some text data. In the process of extracting the temporal features of the elevator operation data, the distribution of the temporal features may be uneven due to the selection of time points or insufficient data preprocessing. When the feature vectors of local abnormal distribution are fused, these abnormal information will be superimposed, which may cause the elevator scheduling generation feature vector to have decoding deviations in the overall distribution, thereby affecting the prediction accuracy and stability of the generator and generating a scheduling plan that does not meet expectations. This offset will make the generated elevator scheduling plan unstable or inefficient in actual application, affecting the overall scheduling effect and user experience. In order to solve this problem, in the technical solution of the present application, the elevator scheduling generation feature vector is calibrated based on probability guidance of back propagation compensation.

[0043] The method comprises: determining a decoding weight matrix of the generator; performing matrix decoupling on the decoding weight matrix in units of row vectors to obtain a set of decoding weight vectors; calculating an elevator scheduling metric learning loss information item between each decoding weight vector in the set of decoding weight vectors and the elevator scheduling generated feature vector to obtain an elevator scheduling metric learning compensation representation vector; inputting the elevator scheduling metric learning compensation representation vector into a softmax activation function to obtain a probabilistic elevator scheduling metric learning compensation weight vector; calculating a position-based point multiplication of the probabilistic elevator scheduling metric learning compensation weight vector and the elevator scheduling generated feature vector to obtain a point product feature vector, and calculating a weighted sum of the point product feature vector and the elevator scheduling generated feature vector to obtain the optimized elevator scheduling generated feature vector.

[0044] The elevator scheduling metric learning loss information item between each decoding weight vector in the set of decoding weight vectors and the elevator scheduling generated feature vector is calculated by the following optimization formula to obtain an elevator scheduling metric learning compensation representation vector;

[0045] Wherein, the optimization formula is:

[0046]

[0047] in, represents the eigenvalue of the i-th position of the j-th decoding weight vector in the set of decoding weight vectors, x i represents the eigenvalue of the ith position of the feature vector generated by the elevator dispatch, L represents the length of the feature vector, and D j Represents the eigenvalue of the j-th position of the elevator scheduling metric learning compensation representation vector.

[0048] That is, the elevator scheduling generated feature vector is calibrated based on the probability guidance of back propagation compensation, which first determines the decoding weight matrix of the generator to understand the generator to distinguish different decoding features. Subsequently, the decoding weight matrix is ​​decoupled by row vectors to obtain a set of decoding weight vectors to allow the model to independently analyze the weight of each decoding and provide more refined control for feature compensation. Next, the elevator scheduling metric learning loss information item is calculated. This process involves quantitative analysis of the interaction between each decoding weight vector and the elevator scheduling generated feature vector. Through the calculation of a specific loss function, the inconsistency between the feature vector and the decoding weight is captured, providing a quantitative indicator for compensation. Next, the elevator scheduling metric learning compensation representation vector is input into the activation function, nonlinearly transformed, and converted into a probabilistic form. The application of the activation function introduces nonlinearity, enhances the model's ability to interpret the compensation weight, and makes the compensation weight more consistent with the actual distribution characteristics. Then, the elevator scheduling metric learning compensation representation vector processed by the activation function is used to calculate the probabilistic elevator scheduling metric learning compensation weight vector. This step provides probabilistic guidance for the adjustment of the feature vector in a probabilistic form, making the optimization process more consistent with statistical characteristics, thereby enhancing the accuracy of the model for feature adjustment. Finally, by calculating the dot product of the probabilistic elevator scheduling metric learning compensation weight vector and the elevator scheduling generated feature vector, and then calculating the weighted sum with the original elevator scheduling generated feature vector, the optimized elevator scheduling generated feature vector is obtained. This step combines the original feature information and compensation information to generate an adjusted feature vector, reducing the impact of abnormal distribution and improving the accuracy of generation.

[0049] Furthermore, the optimized elevator scheduling feature vector contains rich information after preprocessing and feature extraction, such as passenger waiting conditions, elevator operation status, floor demand distribution, etc. The optimized elevator scheduling feature vector is input through the generator and can be processed and analyzed by machine learning models, optimization algorithms or rule engines. The generator uses the optimized elevator scheduling feature vector to perform complex calculations and decision-making processes to generate the best elevator scheduling plan. This may involve advanced algorithmic techniques such as dynamic programming, genetic algorithms, and reinforcement learning to optimize the elevator's running route, stop floors, and running speed based on real-time or predicted data input to maximize overall service efficiency and passenger satisfaction.

[0050] In summary, the embodiment of the present application first obtains the image of the elevator floor to be taken collected by the camera, the historical elevator usage frequency text data collected by the database, and the elevator operation data at multiple predetermined time points, and then uses deep learning technology to perform feature extraction and correlation analysis on the three. Finally, a generator is used to generate an elevator scheduling plan, thereby improving the efficiency and energy utilization of the elevator operation, optimizing the user experience and improving the intelligence level of management decisions.

[0051] As described above, the intelligent building management and monitoring system 100 based on the Internet of Things according to the embodiment of the present application can be implemented in various terminal devices. In one example, the intelligent building management and monitoring system 100 based on the Internet of Things can be integrated into the terminal device as a software module and / or a hardware module. For example, the intelligent building management and monitoring system 100 based on the Internet of Things can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the intelligent building management and monitoring system 100 based on the Internet of Things can also be one of the many hardware modules of the terminal device.

[0052] Alternatively, in another example, the IoT-based smart building management and monitoring system 100 and the terminal device may also be separate devices, and the IoT-based smart building management and monitoring system 100 may be connected to the terminal device via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0053] Below, reference Figure 5 To describe an electronic device according to an embodiment of the present application.

[0054] Figure 5 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0055] The processor 11 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solution provided in the embodiment of the present application; the memory 12 can be implemented in the form of a read-only memory (Read Only Memory, ROM), a static storage device, a dynamic storage device or a random access memory (Random Access Memory, RAM), etc.

[0056] The memory 12 can store an operating system and other application programs. When the technical solution provided in the embodiments of this specification is implemented by software or firmware, the relevant program code is stored in the memory 12, and the processor 11 calls and executes the method of the intelligent building management and monitoring system based on the Internet of Things in the embodiments of this application;

[0057] Input / output interface 13, used to implement information input and output;

[0058] The communication interface 14 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.);

[0059] A bus 15 that transmits information between the various components of the device (e.g., the processor 11, the memory 12, the input / output interface 13, and the communication interface 14);

[0060] The processor 11 , the memory 12 , the input / output interface 13 and the communication interface 14 are connected to each other in communication within the device via a bus 15 .

[0061] An embodiment of the present application also provides a storage medium, which is a computer-readable storage medium. The storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned smart building management and monitoring system method based on the Internet of Things is implemented.

[0062] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0063] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0064] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0065] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0066] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0067] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0068] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0069] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0070] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0071] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method of each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs. The preferred embodiment of the embodiment of the present application is described above with reference to the accompanying drawings, and the scope of rights of the embodiment of the present application is not limited thereto. Any modifications, equivalent substitutions and improvements made by those skilled in the art without departing from the scope and essence of the embodiments of the present application shall be within the scope of rights of the embodiments of the present application.

Claims

1. An intelligent building management and monitoring system based on the Internet of Things, characterized in that: include: The intelligent building elevator data acquisition module is used to obtain the image of the elevator floor to be taken collected by the camera, the historical elevator usage frequency text data collected by the database, and the elevator operation data at multiple predetermined time points; The intelligent building elevator data extraction module is used to extract the image block association feature vector of the elevator floor area to be taken and the elevator operation efficiency feature vector from the elevator floor image to be taken collected by the camera, the historical elevator usage frequency text data collected by the database, and the elevator operation data at the plurality of predetermined time points; The elevator dispatching scheme generating module is used to generate an elevator dispatching scheme based on the associated feature vector of the image block of the floor area of ​​the elevator to be taken and the feature vector of the elevator operation efficiency.

2. The intelligent building management and monitoring system based on the Internet of Things according to claim 1 is characterized in that: The intelligent building elevator data extraction module comprises: An elevator floor image feature extraction unit, used for extracting features from the image of the elevator floor to be taken collected by the camera to obtain a feature vector associated with an image block of the elevator floor area to be taken; A historical elevator usage frequency text feature extraction unit, used to extract features from the historical elevator usage frequency text data collected from the database to obtain a historical elevator usage frequency semantic association feature vector; an elevator operation data feature extraction unit, configured to extract features from the elevator operation data at the plurality of predetermined time points to obtain a time series feature vector of the elevator operation data; The elevator operation efficiency feature fusion unit is used to fuse the historical elevator usage frequency semantic association feature vector and the elevator operation data time series feature vector to obtain the elevator operation efficiency feature vector.

3. The intelligent building management and monitoring system based on the Internet of Things according to claim 2 is characterized in that: The elevator floor image feature extraction unit comprises: The elevator floor image feature encoding subunit is used to perform feature encoding on the image of the elevator floor to be taken acquired by the camera to obtain a plurality of image block feature vectors of the elevator floor area to be taken; The elevator floor image feature construction subunit is used to construct the feature vectors of the plurality of elevator floor area image blocks to be taken into the associated feature vector of the elevator floor area to be taken.

4. The intelligent building management and monitoring system based on the Internet of Things according to claim 3 is characterized in that: The elevator floor image feature encoding subunit comprises: Performing image block processing on the image of the elevator floor to be taken acquired by the camera to obtain an image block sequence of the area of ​​the elevator floor to be taken; The sequence of image blocks of the floor area of ​​the elevator to be taken is passed through a convolutional neural network model of the image blocks of the floor area of ​​the elevator to be taken as a filter to obtain feature vectors of the plurality of image blocks of the floor area of ​​the elevator to be taken.

5. The intelligent building management and monitoring system based on the Internet of Things according to claim 4 is characterized in that: The historical elevator usage frequency text feature extraction unit includes: The historical elevator usage frequency text data collected from the database is passed through a converter-based historical elevator usage frequency text semantic encoder to obtain a plurality of historical elevator usage frequency text semantic feature vectors; The multiple historical elevator usage frequency text semantic feature vectors are cascaded into the historical elevator usage frequency semantic association feature vector.

6. The intelligent building management and monitoring system based on the Internet of Things according to claim 5 is characterized in that: The elevator operation data feature extraction unit comprises: Arranging the elevator operation data at the plurality of predetermined time points into an elevator operation data time series input vector; The elevator operation data time series input vector is passed through an elevator operation data time series feature extractor to obtain the elevator operation data time series feature vector.

7. The intelligent building management and monitoring system based on the Internet of Things according to claim 6 is characterized in that: The elevator dispatching scheme generating module comprises: An elevator dispatch data feature association unit, used to associate the associated feature vector of the image block of the floor area of ​​the elevator to be taken with the elevator operation efficiency feature vector to obtain an elevator dispatch generation feature vector; An elevator dispatching data feature optimization unit, used for performing probability guidance calibration based on back propagation compensation on the elevator dispatching generated feature vector to obtain an optimized elevator dispatching generated feature vector; The scheduling plan generating unit is used to generate an elevator scheduling plan by passing the optimized elevator scheduling generating feature vector through a generator.

8. The intelligent building management and monitoring system based on the Internet of Things according to claim 7 is characterized in that: The elevator dispatching data feature optimization unit comprises: Determining a decoding weight matrix of the generator; Decoupling the decoding weight matrix in units of row vectors to obtain a set of decoding weight vectors; Calculating an elevator scheduling metric learning loss information item between each decoding weight vector in the set of decoding weight vectors and the elevator scheduling generated feature vector to obtain an elevator scheduling metric learning compensation representation vector; Inputting the elevator scheduling metric learning compensation representation vector into a softmax activation function to obtain a probabilistic elevator scheduling metric learning compensation weight vector; The position-based point multiplication of the probabilistic elevator scheduling metric learning compensation weight vector and the elevator scheduling generated feature vector is calculated to obtain a point product feature vector, and a weighted sum of the point product feature vector and the elevator scheduling generated feature vector is calculated to obtain the optimized elevator scheduling generated feature vector.

Images