Energy-saving elevator control system

Abstract

The invention discloses an energy-saving elevator control system, and relates to the technical field of energy-saving elevators. Comprising a historical data collection and analysis module, a regular self-inspection scheduling module, a real-time data acquisition module, a key data extraction and processing module, a deep learning evaluation and state classification module and a dynamic self-inspection period adjustment module, different self-inspection periods are set for each elevator, it is ensured that each elevator is subjected to appropriate inspection frequency according to the actual operation requirement of the elevator, and potential faults caused by untimely self-inspection are avoided. The elevator running state is intelligently evaluated, the self-checking period is dynamically adjusted, the fault early warning and maintenance efficiency of the high-frequency running elevator is improved, and potential faults and safety risks are reduced. And meanwhile, the self-checking period is optimized, unnecessary checking and energy consumption are reduced, the elevator operation efficiency is improved, the purposes of energy conservation and emission reduction are achieved, and the operation efficiency is improved while the elevator safety is guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of energy-saving elevators, and in particular to an energy-saving elevator control system. Background Art

[0002] Energy-saving elevator control refers to the technology that reduces energy consumption and improves overall energy efficiency by optimizing the operation mode and control strategy of elevators in multi-elevator operating environments such as shopping malls or office buildings. Specifically, energy-saving elevator control systems achieve efficient operation through precise scheduling algorithms, intelligent prediction and dynamic adjustment. For example, in a multi-elevator environment, the system can reasonably allocate the operating time and load of elevators according to real-time needs, reduce no-load operation, optimize the scheduling order of elevators going up and down, and ensure the balance and efficient operation of each elevator between different floors. At the same time, the system can also flexibly adjust the operating speed and dwell time of the elevator according to the floor demand forecast to reduce unnecessary pauses and waiting. Modern energy-saving elevators also widely use variable frequency drive technology, which enables the elevator motor to dynamically adjust the power according to the actual load and speed requirements, thereby avoiding energy waste. In addition, through multi-elevator coordinated scheduling, the overall efficiency can be further improved, the waste of power resources can be reduced, and the energy conservation and emission reduction goals can be achieved. The comprehensive application of these technologies not only significantly reduces the energy consumption of elevators, but also improves the operational efficiency and user experience of the building.

[0003] The prior art has the following deficiencies:

[0004] The existing technology usually performs a unified periodic self-inspection on all elevators to identify whether there are any deviations from the normal state during their operation. However, the operating state and workload of each elevator are different, so a unified self-inspection scheme may not meet the inspection needs of high-frequency elevators. Due to frequent starts and stops and greater workloads, the key components of high-frequency elevators (such as motors, brake systems, transmission devices, and door control systems, etc.) are more prone to fatigue loss or premature aging. In this case, a unified periodic self-inspection may not be able to meet the special needs of high-frequency elevators, resulting in a long self-inspection cycle or insufficient inspection, and missing the critical fault warning period. Faults that are not discovered in time may gradually evolve into serious problems, eventually leading to elevator shutdowns, service interruptions, and even safety accidents. Therefore, it is necessary to formulate more flexible and targeted self-inspection strategies based on the actual operating conditions of different elevators to ensure that high-frequency elevators can operate in the best condition and avoid potential faults and safety risks.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] The purpose of the present invention is to provide an energy-saving elevator control system. By intelligently evaluating the elevator operating status and dynamically adjusting the self-check cycle, the solution effectively improves the elevator fault warning capability and maintenance efficiency, ensures that high-frequency operating elevators are inspected in a timely manner, and reduces potential faults and safety risks. At the same time, reasonable adjustment of the self-check cycle optimizes the use of elevator resources, reduces unnecessary inspections and energy consumption, improves the elevator's operating efficiency and energy efficiency, and achieves the goal of energy conservation and emission reduction. While ensuring the safety of the elevator, the solution also improves operational efficiency and energy-saving effects to solve the problems in the above-mentioned background technology.

[0007] In order to achieve the above-mentioned purpose, the present invention provides the following technical solutions: an energy-saving elevator control system, including a historical data collection and analysis module, a periodic self-check scheduling module, a real-time data acquisition module, a key data extraction and processing module, a deep learning evaluation and status classification module, and a dynamic self-check cycle adjustment module:

[0008] The historical data collection and analysis module collects and analyzes the historical operation data of each elevator, sets different self-inspection cycles for each elevator, ensures that each elevator performs appropriate inspection frequency according to its actual operation needs, and avoids potential failures due to untimely self-inspection;

[0009] The regular self-inspection scheduling module, according to the set self-inspection cycle, all elevators conduct regular self-inspections according to their respective cycles, ensuring that each elevator is fully inspected within the preset self-inspection cycle, and that potential problems are discovered and handled in a timely manner to maintain the normal operation of the elevator;

[0010] Real-time data acquisition module, which collects the operation data of each elevator in real time during the operation of the elevator;

[0011] The key data extraction and processing module extracts key features reflecting the high-frequency operation of the elevator from the real-time collected operation data for each elevator during the monitoring window, analyzes the extracted key features, inputs the analyzed features into the pre-trained deep learning model, analyzes the elevator operation data, and evaluates the current operation status of the elevator;

[0012] The deep learning evaluation and status classification module divides the current elevator operation status into two categories: "high-frequency operation" and "smooth operation" according to the evaluation results of the deep learning model;

[0013] The dynamic self-inspection cycle adjustment module continues to perform routine self-inspections according to the preset self-inspection cycle for elevators evaluated as "smooth operation"; for elevators evaluated as "high-frequency operation", the self-inspection cycle is dynamically adjusted according to the evaluation results of the deep learning model, the self-inspection interval is shortened, potential faults are discovered and handled in advance, and the risk of shutdown and safety accidents is reduced.

[0014] Preferably, for each elevator, key features reflecting the high-frequency operation of the elevator are extracted from the operating data collected in real time, the extracted features including the time difference of the elevator operation cycle and the frequency of the elevator position adjustment. During the set monitoring window, after analyzing the extracted features, an elevator operation time difference index and an elevator position adjustment index are generated respectively. The elevator operation time difference index quantifies the time difference required for the elevator to go from one floor to another, reflecting the response speed and efficiency of the elevator under high-frequency operation; the elevator position adjustment index quantifies the frequency of fine-tuning of the elevator's position during the process of going up and down floors, reflecting the precise control and adjustment requirements of the elevator under high-frequency operation.

[0015] Preferably, in the monitoring window, the specific steps of analyzing the elevator operation cycle time difference and generating the elevator operation time difference index are as follows:

[0016] In the monitoring window, the time difference data extracted from the elevator operation cycles between multiple floors are used to generate the time difference fluctuation. The calculation expression is as follows:

[0017]

[0018] , where T cycle (j) represents the time difference of the jth operation cycle, T cycle (j-1) is the time difference of the last operation cycle, α is the weighting coefficient, which is used to adjust the influence of the relative change rate, T fluct (j) is the time difference fluctuation value, which indicates the relative change degree of the time difference between the i-th operation cycle and the previous operation cycle;

[0019] In order to further understand the operating status of the elevator, the cumulative index of time difference fluctuation is constructed, and the calculation expression is as follows:

[0020]

[0021] , where I run (t) is the time difference fluctuation cumulative index, that is, the time difference fluctuation cumulative index at time point t, t is the time point, Δt is the monitoring window length, is the exponential decay factor, exp is the natural exponential function, τ is the decay time constant, and i is the time series index;

[0022] Finally, by calculating the cumulative index I of the time difference fluctuation run (t) is further processed to generate the elevator running time difference index, and the calculation expression is as follows:

[0023]

[0024] , where I diffis the elevator running time difference index, β is the scaling factor, and λ is the sensitivity adjustment factor.

[0025] Preferably, in the monitoring window, the frequency of elevator position adjustment is analyzed, and the specific steps of generating the elevator position adjustment index are as follows:

[0026] In the monitoring window, first collect the elevator operation data in real time, and define the dynamic frequency factor of the elevator position adjustment based on the acquired data. The calculation expression is as follows:

[0027]

[0028] , where N(t) is the number of times the elevator adjusts its position within the monitoring window t, S(t) is the total number of floors the elevator travels within the monitoring window t, ΔT(t) is the monitoring window duration, and α load (t) is the load factor of the elevator within the monitoring window t, F adj (t) is the elevator position dynamic frequency factor;

[0029] When evaluating the frequency of elevator position adjustment, a weight factor of the adjustment amplitude is introduced to enhance the evaluation of the need for fine adjustment of the elevator during frequent operation. The calculation expression is as follows:

[0030]

[0031] , where W adj (t) is the adjustment amplitude weight factor, Δd p is the amplitude of the pth adjustment, max(Δd p ) is the maximum adjustment amplitude within the monitoring window t, and ω is the amplitude weighting index;

[0032] Based on the dynamic frequency factor F of the elevator position adj (t) and the adjustment amplitude weight factor W adj (t) Generate the elevator position adjustment index, the calculation expression is as follows:

[0033]

[0034] , where Δt p is the duration of the pth position adjustment, F adjustment is the elevator position adjustment index.

[0035] Preferably, the elevator operation time difference index and the elevator position adjustment index generated by analyzing the elevator operation cycle time difference and the frequency of elevator position adjustment extracted from the elevator operation data are input into a pre-learned deep learning model, and the working frequency index is generated by the deep learning model, and the current elevator operation status is intelligently evaluated by the working frequency index.

[0036] Preferably, the operating frequency index generated when the elevator operating state is intelligently evaluated by the pre-learned deep learning model during the monitoring window is compared and analyzed with a pre-set reference threshold of the operating frequency index, and the current elevator operating state is divided, and the division steps are as follows:

[0037] If the working frequency index is greater than or equal to a preset working frequency index reference threshold, the current elevator operation state is classified as high-frequency operation;

[0038] If the operating frequency index is less than a preset operating frequency index reference threshold, the current elevator operation state is classified as stable operation.

[0039] Preferably, for elevators evaluated as "high-frequency operation", the specific steps of dynamically adjusting their self-inspection cycles and shortening the self-inspection intervals are as follows according to the evaluation results of the deep learning model:

[0040] The elevator is evaluated as being in high-frequency operation state, and the self-inspection cycle is dynamically adjusted to shorten the self-inspection interval. Based on the evaluation results of the deep learning model, the elevator's operating frequency index Elevator is Work Calculate the adjusted self-test period, the calculation expression is as follows:

[0041]

[0042] , where T self_check_orig is the preset self-test cycle, is the adjustment factor, Elevator ref is the operating frequency index reference threshold, T self_check_new is the adjusted self-test cycle;

[0043] After adjusting the self-check cycle, the running status of the elevator is tracked in real time, and the fault warning is quantified. The calculation expression is as follows:

[0044]

[0045] , where F warning is the fault warning value, τ is the weight factor, γ q is the weight factor of the performance indicator q, K q is the performance index of the qth elevator, and n is the total number of performance indexes;

[0046] Based on the dynamically adjusted self-check cycle and health status assessment results, the elevator maintenance and service strategy will be adjusted accordingly to generate the maintenance priority. The calculation expression is as follows:

[0047] M priority =δ·F warning +κ·(Elevator Work-Elevator ref )

[0048] , where M priority is the maintenance priority of the elevator, δ is the fault warning weight factor, and K is the working frequency weight factor.

[0049] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0050] The present invention uses a deep learning model to intelligently evaluate the operating status of the elevator and can accurately identify the potential risks of high-frequency operating elevators. When the elevator is in a high-frequency operating state, the system will dynamically adjust the self-inspection cycle to shorten the self-inspection interval. This strategy ensures that elevators that run at a high frequency can be inspected more frequently at an early stage, so as to promptly detect possible fatigue loss or aging of key components (such as motors, braking systems, transmission devices, etc.). This not only effectively avoids the accumulation of faults and reduces the risk of elevator outages, but also ensures that the elevator is always in the best operating state throughout its life cycle. Compared with the traditional unified periodic self-inspection, the dynamically adjusted self-inspection strategy can significantly improve the maintenance efficiency and fault warning capability of the elevator, reduce the occurrence rate of faults, and avoid safety accidents and operation interruptions caused by faults that are not discovered in time.

[0051] Through intelligent evaluation and analysis of elevator operation data, the energy-saving elevator control system of the present invention can reasonably adjust the self-inspection cycle according to the actual workload and operating frequency of the elevator, so that elevators running at high frequency receive more attention and resource support, while maintaining a standard self-inspection cycle for elevators running at low frequency and smoothly. This strategy allows the elevator to avoid excessive inspections and reduce unnecessary maintenance operations while ensuring safety, thereby saving operating costs. In addition, the system can monitor the operating frequency and operating status of the elevator in real time, adjust the operating parameters of the elevator (such as operating speed, dwell time, etc.) according to actual needs, and further improve the energy efficiency of the elevator. With the support of the energy-saving elevator control solution, the operating efficiency of the elevator is optimized, which not only ensures the safety and reliability of the equipment, but also significantly reduces energy consumption, achieving the goal of energy conservation and emission reduction. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0053] Figure 1 It is a module schematic diagram of the energy-saving elevator control system of the present invention. DETAILED DESCRIPTION

[0054] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0055] The present invention provides Figure 1 The energy-saving elevator control system shown includes a historical data collection and analysis module, a regular self-inspection scheduling module, a real-time data acquisition module, a key data extraction and processing module, a deep learning evaluation and status classification module, and a dynamic self-inspection cycle adjustment module:

[0056] The historical data collection and analysis module collects and analyzes the historical operation data of each elevator, sets different self-inspection cycles for each elevator, ensures that each elevator performs appropriate inspection frequency according to its actual operation needs, and avoids potential failures due to untimely self-inspection;

[0057] By personalizing the self-inspection cycle, it is ensured that each elevator is inspected at an appropriate frequency according to its actual operating needs, avoiding potential failures of high-frequency elevators due to untimely self-inspection, and avoiding waste of resources due to excessive self-inspection of low-frequency elevators.

[0058] The regular self-inspection scheduling module, according to the set self-inspection cycle, all elevators conduct regular self-inspections according to their respective cycles, ensuring that each elevator is fully inspected within the preset self-inspection cycle, and that potential problems are discovered and handled in a timely manner to maintain the normal operation of the elevator;

[0059] All elevators are regularly inspected according to their set self-inspection cycles to ensure that each elevator is fully inspected at the specified time intervals, and potential problems are discovered and dealt with in a timely manner. The purpose of this is to ensure that the elevators remain in good working condition during daily operation and to avoid elevator failures or accidents due to failures or lack of maintenance. The self-inspection cycle of each elevator is set according to its specific usage (such as frequency of use, load, operating time, etc.), so the time and frequency of inspections will vary. This approach can ensure the long-term reliability and safety of the elevator, and at the same time, timely discover problems through regular inspections to avoid large losses or safety hazards caused by the accumulation of failures.

[0060] The preset self-inspection cycle needs to be reasonably set according to factors such as the frequency of use, load conditions, operating environment, and historical fault records of the elevator. The cycle should be short enough to ensure that the elevator can be inspected and maintained in time before potential faults occur, so as to avoid the development of serious faults due to neglect of small problems. The setting of the self-inspection cycle should not only consider the actual use intensity of the elevator, but also match the performance degradation law of the key components of the elevator (such as motors, brake systems, door control systems, etc.). Regular self-inspection can ensure that each elevator always maintains the best working condition during operation, promptly detects abnormalities that may cause safety hazards, and through timely maintenance, avoids elevator failures and shutdowns or safety threats to passengers, thereby maintaining the long-term stable operation of the elevator.

[0061] Real-time data acquisition module, which collects the operation data of each elevator in real time during the operation of the elevator;

[0062] These sensors include current sensors, temperature sensors, load sensors, speed sensors, door control sensors, etc., which can monitor the motor load, operating speed, temperature changes, door opening and closing frequency and load conditions of the elevator. All collected data is transmitted to the central monitoring system through the network to ensure the real-time and accuracy of the data. The acquisition of real-time data provides a basis for subsequent operation status evaluation, ensuring that every detail of the elevator operation can be accurately monitored and recorded, and providing support for intelligent evaluation and self-inspection cycle adjustment.

[0063] The key data extraction and processing module extracts key features reflecting the high-frequency operation of the elevator from the real-time collected operation data for each elevator during the monitoring window, analyzes the extracted key features, inputs the analyzed features into the pre-trained deep learning model, analyzes the elevator operation data, and evaluates the current operation status of the elevator;

[0064] For each elevator, key features reflecting the high-frequency operation of the elevator are extracted from the real-time collected operation data. The extracted features include the time difference of the elevator operation cycle and the frequency of the elevator position adjustment. During the set monitoring window, the extracted features are analyzed to generate the elevator operation time difference index and elevator position adjustment index respectively. The elevator operation time difference index quantifies the time difference required for the elevator to go from one floor to another, reflecting the response speed and efficiency of the elevator under high-frequency operation; the elevator position adjustment index quantifies the frequency of fine-tuning the elevator's position when going up and down floors, reflecting the need for precise control and adjustment of the elevator under high-frequency operation.

[0065] When the elevator cycle time difference increases, it usually indicates that the elevator is in a high-frequency operation state. The elevator cycle time difference reflects the fluctuation of the time required to go from one floor to another. In the case of high-frequency operation, the elevator may start and stop frequently, especially in the case of uneven load, large passenger flow or the elevator constantly going up and down floors, the operation time difference will increase significantly. This is because the elevator will be affected by various factors during operation, such as load changes, floor distance, inertia during starting and stopping, etc., resulting in slightly different operation times each time. As the frequency of elevator use increases, especially during busy hours or high-load conditions, the fluctuation of the time difference will be more obvious, reflecting the working state of the elevator under long-term high-frequency operation.

[0066] In the monitoring window, the specific steps for analyzing the elevator operation cycle time difference and generating the elevator operation time difference index are as follows:

[0067] In the monitoring window, the time difference data extracted from the elevator operation cycle between multiple floors is used to generate the time difference fluctuation, which is used to reflect the degree of change of the elevator operation status. The calculation expression is as follows:

[0068]

[0069] , where T cycle (j) represents the time difference of the jth operation cycle, T cycle (j-1) is the time difference of the last operation cycle, α is the weighting coefficient, which is used to adjust the influence of the relative change rate, T fluct (j) is the time difference fluctuation value, which indicates the relative change degree of the time difference between the i-th operation cycle and the previous operation cycle;

[0070] In this way, larger fluctuation values ​​will be more prominently highlighted, reflecting the complexity and instability of elevators at high frequency operation.

[0071] In order to further understand the operating status of the elevator, a cumulative index of time difference fluctuation is constructed. This index performs weighted accumulation of fluctuation values ​​of multiple time periods to reflect the overall operating trend of the elevator within the monitoring window. The calculation expression is as follows:

[0072]

[0073] , where I run (t) is the time difference fluctuation cumulative index, that is, the time difference fluctuation cumulative index at time point t, t is the time point, Δt is the monitoring window length, is the exponential decay factor, exp is the natural exponential function, τ is the decay time constant, and i is the time series index;

[0074] Exponential decay is used to weight fluctuations within a short distance and simulate the impact of different time periods during elevator operation, so as to accurately evaluate the elevator operation status.

[0075] Finally, by calculating the cumulative index I of the time difference fluctuation run (t) is further processed to generate the elevator running time difference index, and the calculation expression is as follows:

[0076]

[0077] , where I diff is the elevator running time difference index, β is the scaling factor used to adjust the sensitivity of the formula, and λ is the sensitivity adjustment factor used to control the attenuation rate.

[0078] In the monitoring window, the larger the performance value of the elevator operation time difference index generated after analyzing the elevator operation cycle time difference, it usually indicates that the current elevator is in a high-frequency operation state. During the high-frequency operation of the elevator, the time difference of each operation cycle increases due to frequent starts and stops and load fluctuations. A large time difference means that the elevator's operation time between different floors fluctuates significantly, which is usually the result of long-term high-frequency operation of the elevator. For example, the elevator needs to respond to passenger needs frequently, and the load and start / stop times increase, resulting in inconsistent operation cycles each time. On the contrary, when the elevator is in a stable operation state, the operation cycle time difference is small, and the elevator's operation time between different floors tends to be consistent, indicating that the elevator load is light and the frequency of use is low.

[0079] When the elevator position is adjusted frequently, it usually indicates that the elevator is in a high-frequency operation state. The precise position adjustment of the elevator is to ensure that the elevator can stop accurately at each floor and ensure smooth boarding and alighting of passengers. In the case of high-frequency operation, the elevator needs to start and stop frequently and may carry different loads. With frequent starting, stopping and passenger loading, the elevator's control system will require more fine-tuning and position correction, especially in environments with high loads or large passenger flow. The increase in the frequency of elevator fine-tuning is usually due to load fluctuations, frequent boarding and alighting of passengers, or the heavy operating load of the elevator itself, which causes the elevator to require more adjustments every time it arrives at a floor to ensure accurate positioning and smooth operation.

[0080] In the monitoring window, the frequency of elevator position adjustment is analyzed and the specific steps to generate the elevator position adjustment index are as follows:

[0081] In the monitoring window, first collect the elevator operation data in real time, especially the position adjustment of the elevator during operation. Based on the acquired data, define the dynamic frequency factor of the elevator position adjustment. The calculation expression is as follows:

[0082]

[0083] , where N(t) is the number of times the elevator adjusts its position within the monitoring window t, S(t) is the total number of floors the elevator travels within the monitoring window t, ΔT(t) is the monitoring window duration, and α load (t) is the load factor of the elevator within the monitoring window t, F adj (t) is the elevator position dynamic frequency factor;

[0084] The purpose of this step is to normalize the elevator's adjustment frequency with the load, travel distance and time to avoid errors caused by different time spans or numbers of traveled floors, and to ensure the accuracy of the calculation results.

[0085] When evaluating the frequency of elevator position adjustment, considering only the number of adjustments cannot fully reflect the high-frequency operation status of the elevator. The weight factor of the adjustment amplitude is introduced to enhance the evaluation of the fine adjustment needs of the elevator during frequent operation. The calculation expression is as follows:

[0086]

[0087] , where W adj (t) is the adjustment amplitude weight factor, Δd p is the amplitude of the pth adjustment, max(Δd p ) The maximum adjustment amplitude within the monitoring window t, ω is the amplitude weighting index, which is used to control the influence of each position adjustment amplitude on the weight factor;

[0088] The purpose of this step is to weight each adjustment amplitude so that a larger adjustment amplitude has a higher weight in the calculation of the frequency factor, thereby more accurately reflecting the actual performance of the elevator under high-frequency operation conditions.

[0089] Based on the dynamic frequency factor F of the elevator position adj (t) and the adjustment amplitude weight factor W adj (t) Generate the elevator position adjustment index, the calculation expression is as follows:

[0090]

[0091] , where Δt p is the duration of the pth position adjustment, F adjustment is the elevator position adjustment index.

[0092] Under the monitoring window, the larger the performance value of the elevator position adjustment index generated after analyzing the frequency of the elevator position adjustment, the larger the performance value of the elevator position adjustment index, which usually indicates that the elevator is in a high-frequency operation state, and vice versa, it indicates that the elevator is in a stable operation state. When the elevator frequently adjusts its position, it means that the elevator frequently starts and stops, the load changes, or the passenger flow is large during the operation between different floors, requiring the control system to make more fine-tuning to ensure that the elevator can accurately align with the floor every time it stops. This frequent adjustment usually occurs in an environment where the elevator is in high-frequency operation for a long time, such as in a building during peak hours or when there are a large number of passengers. On the contrary, when the elevator is in a relatively stable operation state, the load changes are small, the start-stop frequency is low, and the demand for elevator position adjustment is reduced, resulting in a smaller adjustment factor.

[0093] The elevator operation time difference index and elevator position adjustment index generated by analyzing the elevator operation cycle time difference and the frequency of elevator position adjustment extracted from the elevator operation data are input into the pre-learned deep learning model, and the working frequency index is generated by the deep learning model. The working frequency index is used to intelligently evaluate the current elevator operation status.

[0094] A pre-learned deep learning model refers to a deep neural network model trained and optimized using historical data during the long-term accumulation and analysis of elevator operation data. The model can intelligently analyze the operation status of the elevator. During the training phase, this model is learned through a large amount of elevator operation data, including various operation characteristics of the elevator (such as the time difference of the elevator operation cycle, the number of position adjustments, etc.), and the relationship between these characteristics and the actual operation status of the elevator, such as faults, loads, and frequencies. By learning these historical data, the deep learning model can extract potential rules and patterns from them, so that the model can accurately judge the operating frequency, operation intensity, and potential failure risks of the elevator. Specifically, the model learning process will consider data from multiple dimensions such as the use environment, load conditions, and operating frequency of different elevators, and continuously optimize the weights through the back propagation algorithm until the model's prediction accuracy on the test data reaches the expected level.

[0095] After the model training is completed, the pre-learned deep learning model can be used to analyze the elevator's operating data in real time. When new elevator data is input into the model, the model will intelligently evaluate the current elevator's operating status based on the learned knowledge and rules, and generate an operating frequency index. This operating frequency index is a quantitative indicator of the elevator's operating status, which can reflect whether the elevator is currently in high-frequency operation, whether there is a potential risk of failure, and whether more frequent self-inspection and maintenance are required. Through the evaluation of the deep learning model, managers can obtain real-time and accurate feedback on the elevator status, thereby adjusting the self-inspection cycle and optimizing the elevator scheduling strategy according to the actual needs of different elevators, further improving the elevator's operating efficiency and safety. In addition, the application of the deep learning model makes the elevator monitoring no longer rely on traditional fixed rules, but dynamically adjusts according to the actual operation of each elevator, ensuring that the elevator can achieve the best performance under different workloads and usage scenarios.

[0096] The deep learning model is not limited here, and can realize the elevator running time difference index I diff and elevator position adjustment index F adjustment Perform comprehensive analysis to generate the operating frequency index Elevator Work In order to realize the technical solution of the present invention, the present invention provides a specific implementation method;

[0097] Working frequency index Elevator Work The generation formula is as follows:

[0098]

[0099] , where Q1 and Q2 are the elevator running time difference index I diff and elevator position adjustment index F adjustment The preset proportional coefficient, and Q1 and Q2 are both greater than 0.

[0100] It can be seen from the working frequency index performance value that, under the monitoring window, the larger the elevator operation time difference index performance value generated after analyzing the elevator operation cycle time difference, and the larger the elevator position adjustment index performance value generated after analyzing the frequency of the elevator position adjustment, the larger the working frequency index performance value generated when the elevator operation status is intelligently evaluated through the pre-learned deep learning model under the monitoring window, indicating that the current elevator is in a high-frequency operation state, otherwise, it indicates that the current elevator is in a stable operation state.

[0101] The deep learning evaluation and status classification module divides the current elevator operation status into two categories: "high-frequency operation" and "smooth operation" according to the evaluation results of the deep learning model;

[0102] The operating frequency index generated by the intelligent evaluation of the elevator operation status through the pre-learned deep learning model during the monitoring window is compared and analyzed with the pre-set reference threshold of the operating frequency index to divide the current elevator operation status. The division steps are as follows:

[0103] If the working frequency index is greater than or equal to a preset working frequency index reference threshold, the current elevator operation state is classified as high-frequency operation;

[0104] If the operating frequency index is less than a preset operating frequency index reference threshold, the current elevator operation state is classified as stable operation.

[0105] The high-frequency operation state means that the elevator is subjected to high loads and frequent operations for a long time, usually experiencing more starts and stops, shorter operation cycles and frequent inter-floor dispatching; the stable operation state means that the elevator operates under conditions of lower loads and less operating frequency, usually occurring during off-peak periods or when the load is light.

[0106] Dynamic self-inspection cycle adjustment module, for elevators evaluated as "smooth operation", continue to perform regular self-inspection according to the preset self-inspection cycle;

[0107] For elevators that are evaluated as "smoothly running", the purpose of continuing to perform regular self-inspections according to the preset self-inspection cycle is to ensure that the elevator maintains its safety and reliability during stable operation. Even if the elevator does not show obvious signs of high-frequency operation or failure, regular self-inspections can still promptly discover some potential problems or gradually emerging hidden dangers, such as wear and tear of mechanical parts, minor failures in the electronic control system, or a slight decrease in elevator operating efficiency. Through continuous self-inspection, it can ensure that the various systems of the elevator are always in the best working condition, discover and eliminate hidden dangers in advance, and avoid major failures caused by the accumulation of small problems, thereby extending the service life of the elevator, reducing the occurrence of sudden failures, and improving passenger safety and user experience. In addition, regular self-inspections can also optimize the maintenance management of the elevator, ensure that the equipment meets safety standards at all times, and avoid potential safety risks that are overlooked when the elevator is running smoothly.

[0108] For elevators that are assessed as "high-frequency operation", the self-inspection cycle is dynamically adjusted based on the evaluation results of the deep learning model, the self-inspection interval is shortened, potential faults are discovered and handled in advance, and the risk of downtime and safety accidents is reduced;

[0109] For elevators that are assessed as "high-frequency operation", the specific steps for dynamically adjusting their self-inspection cycles and shortening the self-inspection intervals are as follows:

[0110] The elevator is evaluated as being in high-frequency operation state, and the self-inspection cycle is dynamically adjusted to shorten the self-inspection interval. Based on the evaluation results of the deep learning model, the elevator's operating frequency index Elevator is Work Calculate the adjusted self-test period, the calculation expression is as follows:

[0111]

[0112] , where T self_check_orig is the preset self-test cycle, It is a regulating factor used to control the influence of the operating frequency index on the self-test cycle. ref is the operating frequency index reference threshold, T self_check_new is the adjusted self-test cycle;

[0113] If the elevator is assessed to be in high-frequency operation state, the self-inspection cycle will be shortened, thereby ensuring that the high-frequency operation state of the elevator can be monitored and maintained more frequently and potential faults can be discovered in time.

[0114] After adjusting the self-check cycle, the elevator's operating status is tracked in real time, and the health status is continuously evaluated in the new self-check cycle. To this end, the deep learning model will continue to track the high-frequency operating characteristics of the elevator, and perform regular inspections according to the new self-check cycle to quantify the fault warning. The calculation expression is as follows:

[0115]

[0116] , where F warning is the fault warning value, τ is the weight factor, and controls the working frequency index Elevator Work and fault warning value F warning The relationship between γ q is the weight factor of the performance index q, indicating the influence of various health parameters of the elevator on fault warning. q is the performance index of the qth elevator, including but not limited to the operating temperature of the motor, the working pressure of the brake system, and the load condition of the elevator, and n is the total number of performance indexes;

[0117] Through comprehensive calculation, a fault warning value F is obtained. warning , used to determine whether the elevator is about to fail or have a safety problem. warning If the preset threshold is exceeded, a fault warning is triggered and maintenance response is accelerated.

[0118] Based on the dynamically adjusted self-inspection cycle and health status assessment results, the elevator maintenance and service strategy will be adjusted accordingly. In the case of high-frequency operation of elevators, not only will the self-inspection cycle be shortened, but the daily scheduling and operation load of the elevator will also need to be adjusted to generate maintenance priorities. The calculation expression is as follows:

[0119] M priority =δ·F warning +κ·(Elevator Work -Elevator ref )

[0120] , where M priority is the maintenance priority of the elevator, δ is the fault warning weight factor, which is used to adjust the impact of fault warning on maintenance priority, and κ is the working frequency weight factor, which controls the impact of working frequency index on maintenance priority.

[0121] Higher M priority A low value means that the elevator needs more urgent maintenance and inspection to ensure its stable operation and minimize service interruptions and safety hazards caused by elevator failures. This dynamic adjustment mechanism can optimize the elevator's service strategy and improve the elevator's operating safety and efficiency by comprehensively considering the elevator's health status and work intensity.

[0122] By introducing a maintenance priority adjustment formula, we are able to optimize the service schedule of the elevators based on their current status

[0123] For elevators that are evaluated as "high-frequency operation", the self-inspection cycle is dynamically adjusted according to the evaluation results of the deep learning model. The purpose of shortening the self-inspection interval is to deal with the risk of failure that may be aggravated during the frequent use of the elevator. Elevators that operate at a high frequency are subject to greater loads and frequent start and stop operations, which makes the elevator components (such as motors, brake systems, transmission devices, etc.) prone to fatigue, wear or aging, and potential failures may also occur earlier. By shortening the self-inspection interval, the operating status of the elevator can be monitored more frequently, and hidden problems caused by high-frequency operation can be discovered in time to prevent small failures from turning into big problems. In this way, potential failures can be identified and handled in advance, reducing the risk of equipment downtime, ensuring the continuous safe and efficient operation of the elevator, reducing downtime and safety accidents, and improving users' sense of security and user experience.

[0124] The present invention uses a deep learning model to intelligently evaluate the operating status of the elevator and can accurately identify the potential risks of high-frequency operating elevators. When the elevator is in a high-frequency operating state, the system will dynamically adjust the self-inspection cycle to shorten the self-inspection interval. This strategy ensures that elevators that run at a high frequency can be inspected more frequently at an early stage, so as to promptly detect possible fatigue loss or aging of key components (such as motors, braking systems, transmission devices, etc.). This not only effectively avoids the accumulation of faults and reduces the risk of elevator outages, but also ensures that the elevator is always in the best operating state throughout its life cycle. Compared with the traditional unified periodic self-inspection, the dynamically adjusted self-inspection strategy can significantly improve the maintenance efficiency and fault warning capability of the elevator, reduce the occurrence rate of faults, and avoid safety accidents and operation interruptions caused by faults that are not discovered in time.

[0125] Through intelligent evaluation and analysis of elevator operation data, the energy-saving elevator control system of the present invention can reasonably adjust the self-inspection cycle according to the actual workload and operating frequency of the elevator, so that elevators running at high frequency receive more attention and resource support, while maintaining a standard self-inspection cycle for elevators running at low frequency and smoothly. This strategy allows the elevator to avoid excessive inspections and reduce unnecessary maintenance operations while ensuring safety, thereby saving operating costs. In addition, the system can monitor the operating frequency and operating status of the elevator in real time, adjust the operating parameters of the elevator (such as operating speed, dwell time, etc.) according to actual needs, and further improve the energy efficiency of the elevator. With the support of the energy-saving elevator control solution, the operating efficiency of the elevator is optimized, which not only ensures the safety and reliability of the equipment, but also significantly reduces energy consumption, achieving the goal of energy conservation and emission reduction.

[0126] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. Energy-saving elevator control system, characterized in that: It includes historical data collection and analysis module, regular self-inspection scheduling module, real-time data acquisition module, key data extraction and processing module, deep learning evaluation and status classification module and dynamic self-inspection cycle adjustment module: The historical data collection and analysis module collects and analyzes the historical operation data of each elevator, sets different self-inspection cycles for each elevator, ensures that each elevator performs appropriate inspection frequency according to its actual operation needs, and avoids potential failures due to untimely self-inspection; The regular self-inspection scheduling module, according to the set self-inspection cycle, all elevators conduct regular self-inspections according to their respective cycles, ensuring that each elevator is fully inspected within the preset self-inspection cycle, and that potential problems are discovered and handled in a timely manner to maintain the normal operation of the elevator; Real-time data acquisition module, which collects the operation data of each elevator in real time during the operation of the elevator; The key data extraction and processing module extracts key features reflecting the high-frequency operation of the elevator from the real-time collected operation data for each elevator during the monitoring window, analyzes the extracted key features, inputs the analyzed features into the pre-trained deep learning model, analyzes the elevator operation data, and evaluates the current operation status of the elevator; The deep learning evaluation and status classification module divides the current elevator operation status into two categories: "high-frequency operation" and "smooth operation" according to the evaluation results of the deep learning model; The dynamic self-inspection cycle adjustment module continues to perform routine self-inspections according to the preset self-inspection cycle for elevators evaluated as "smooth operation"; for elevators evaluated as "high-frequency operation", the self-inspection cycle is dynamically adjusted according to the evaluation results of the deep learning model, the self-inspection interval is shortened, potential faults are discovered and handled in advance, and the risk of downtime and safety accidents is reduced.

2. The energy-saving elevator control system according to claim 1, characterized in that: For each elevator, key features reflecting the high-frequency operation of the elevator are extracted from the real-time collected operation data. The extracted features include the time difference of the elevator operation cycle and the frequency of the elevator position adjustment. During the set monitoring window, the extracted features are analyzed to generate the elevator operation time difference index and elevator position adjustment index respectively. The elevator operation time difference index quantifies the time difference required for the elevator to go from one floor to another, reflecting the response speed and efficiency of the elevator under high-frequency operation; the elevator position adjustment index quantifies the frequency of fine-tuning the elevator's position when going up and down floors, reflecting the need for precise control and adjustment of the elevator under high-frequency operation.

3. The energy-saving elevator control system according to claim 2, characterized in that: In the monitoring window, the specific steps for analyzing the elevator operation cycle time difference and generating the elevator operation time difference index are as follows: In the monitoring window, the time difference data extracted from the elevator operation cycles between multiple floors are used to generate the time difference fluctuation. The calculation expression is as follows: , where T cycle (j) represents the time difference of the jth operation cycle, T cycle (j-1) is the time difference of the last operation cycle, α is the weighting coefficient, which is used to adjust the influence of the relative change rate, T fluct (j) is the time difference fluctuation value, which indicates the relative change degree of the time difference between the i-th operation cycle and the previous operation cycle; In order to further understand the operating status of the elevator, the cumulative index of time difference fluctuation is constructed, and the calculation expression is as follows: , where I run (t) is the time difference fluctuation cumulative index, that is, the time difference fluctuation cumulative index at time point t, t is the time point, Δt is the monitoring window length, is the exponential decay factor, exp is the natural exponential function, τ is the decay time constant, and i is the time series index; Finally, by calculating the cumulative index I of the time difference fluctuation run (t) is further processed to generate the elevator running time difference index, and the calculation expression is as follows: , where I diff is the elevator running time difference index, β is the scaling factor, and λ is the sensitivity adjustment factor.

4. The energy-saving elevator control system according to claim 2, characterized in that: In the monitoring window, the frequency of elevator position adjustment is analyzed and the specific steps to generate the elevator position adjustment index are as follows: In the monitoring window, first collect the elevator operation data in real time, and define the dynamic frequency factor of the elevator position adjustment based on the acquired data. The calculation expression is as follows: ， Where N(t) is the number of times the elevator adjusts its position within the monitoring window t, S(t) is the total number of floors the elevator travels within the monitoring window t, ΔT(t) is the monitoring window duration, and α load (t) is the load factor of the elevator within the monitoring window t, F adj (t) is the elevator position dynamic frequency factor; When evaluating the frequency of elevator position adjustment, a weight factor of the adjustment amplitude is introduced to enhance the evaluation of the need for fine adjustment of the elevator during frequent operation. The calculation expression is as follows: , where W adj (t) is the adjustment amplitude weight factor, Δd p is the amplitude of the pth adjustment, max(Δd p ) is the maximum adjustment amplitude within the monitoring window t, and ω is the amplitude weighting index; Based on the dynamic frequency factor F of the elevator position adj (t) and the adjustment amplitude weight factor W adj (t) Generate the elevator position adjustment index, the calculation expression is as follows: , where Δt p is the duration of the pth position adjustment, F adjustment is the elevator position adjustment index.

5. The energy-saving elevator control system according to claim 2, characterized in that: The elevator operation time difference index and elevator position adjustment index generated by analyzing the elevator operation cycle time difference and the frequency of elevator position adjustment extracted from the elevator operation data are input into the pre-learned deep learning model, and the working frequency index is generated by the deep learning model. The working frequency index is used to intelligently evaluate the current elevator operation status.

6. The energy-saving elevator control system according to claim 5, characterized in that: The operating frequency index generated by the intelligent evaluation of the elevator operation status through the pre-learned deep learning model during the monitoring window is compared and analyzed with the pre-set reference threshold of the operating frequency index to divide the current elevator operation status. The division steps are as follows: If the working frequency index is greater than or equal to a preset working frequency index reference threshold, the current elevator operation state is classified as high-frequency operation; If the operating frequency index is less than a preset operating frequency index reference threshold, the current elevator operation state is classified as stable operation.

7. The energy-saving elevator control system according to claim 6, characterized in that: For elevators that are assessed as "high-frequency operation", the specific steps for dynamically adjusting their self-inspection cycles and shortening the self-inspection intervals are as follows: The elevator is evaluated as being in high-frequency operation state, and the self-inspection cycle is dynamically adjusted to shorten the self-inspection interval. Based on the evaluation results of the deep learning model, the elevator's operating frequency index Elevator is Work Calculate the adjusted self-test period, the calculation expression is as follows: , where T self_check_orig is the preset self-test cycle, is the adjustment factor, Elevator ref is the operating frequency index reference threshold, T self_check_new is the adjusted self-test cycle; After adjusting the self-check cycle, the running status of the elevator is tracked in real time, and the fault warning is quantified. The calculation expression is as follows: , where F warning is the fault warning value, τ is the weight factor, γ q is the weight factor of the performance indicator q, K q is the performance index of the qth elevator, and n is the total number of performance indexes; Based on the dynamically adjusted self-check cycle and health status assessment results, the elevator maintenance and service strategy will be adjusted accordingly to generate the maintenance priority. The calculation expression is as follows: M priority =δ·F warning +κ·(Elevator Work -Elevator ref ), where M priority is the maintenance priority of the elevator, δ is the fault warning weight factor, and κ is the operating frequency weight factor.

Images