Elevator group management method based on thermal safety boundary prediction and cooperative scheduling
By monitoring the operating parameters of the elevator power module in real time, dynamically predicting the thermal safety boundary and generating a load reduction protection strategy, the blind spot problem of thermal safety management in the elevator group control system is solved, improving the reliability of elevator operation and equipment life, while optimizing the system's transportation efficiency.
Patent Information
- Application Number
- CN202511673235.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-14
- Publication Date
- 2026-03-17
AI Technical Summary
Existing elevator group control systems have blind spots in thermal safety management. The scheduling strategy is disconnected from the physical health status of the drive system, leading to the accumulation of thermal failure risks and affecting equipment reliability and user experience.
By collecting the operating parameters of the elevator power module in real time, calculating the junction temperature and power loss of the power chip, dynamically predicting the thermal safety boundary, generating a load reduction protection strategy, and coordinating the adjustment of task allocation to avoid overheating, the continuous operation of the elevator and the lifespan of the hardware are ensured.
This has enabled continuous elevator operation and improved hardware lifespan, avoiding sudden shutdowns caused by overheating, and significantly improving the operational reliability and overall transportation efficiency of elevator groups.
Smart Images
Figure CN121672293A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the field of elevator safety and dispatching control technology, in particular to an elevator group management method based on thermal safety boundary prediction and collaborative scheduling. BACKGROUND
[0002] The elevator group control system is the core of modern high-rise buildings, which optimizes the scheduling algorithm to improve transportation efficiency and service quality. However, while pursuing operational efficiency, existing systems generally ignore a key underlying physical constraint: the thermal safety and reliability of power modules in the elevator drive system.
[0003] Specifically, the existing technology mainly has the following defects: first, the scheduling strategy is disconnected from the physical state. The decision of the existing group control algorithm is only based on the position, call signal and other logical operating states of the elevator. The system may continuously assign heavy tasks to an elevator that is already under high thermal load, unintentionally causing the power device to approach the thermal failure critical point and creating a potential failure. Second, thermal management measures are passive and isolated. Current systems usually use over-temperature protection based on fixed temperature thresholds, which passively triggers load reduction or shutdown only after a crisis occurs. This after-the-fact remediation model can cause sudden service interruptions, severely affecting user experience. More importantly, this protection only applies to a single elevator, and the withdrawal of one elevator can increase the load of the remaining elevators, possibly causing a systemic decline in service capacity. Finally, there is a lack of predictive safety boundary management. The system cannot dynamically predict the thermal safety boundary of the power module based on operating parameters and cooling conditions, which prevents proactive scheduling adjustments before thermal risk accumulates, not only accelerating equipment aging, but also making maintenance dependent on regular maintenance rather than more efficient predictive maintenance.
[0004] In summary, the existing elevator group control system has a blind spot in thermal safety management, and its scheduling strategy is disconnected from the physical health status of the drive system, which severely restricts the long-term reliability of the system under high-intensity operation. SUMMARY
[0005] The present application provides an elevator group management method based on thermal safety boundary prediction and collaborative scheduling, which can improve the safety of elevator hardware operation while effectively ensuring the continuity of elevator service and prolonging the service life of core equipment.
[0006] In order to achieve the above purpose, the present application provides the following technical solutions: An elevator group management method based on thermal safety boundary prediction and collaborative scheduling, comprising the following steps: Collecting real-time operating parameters of power modules of each elevator and real-time environmental temperature in the control cabinet of each power module; Based on the real-time operating parameters of each power module, calculate the real-time junction temperature of the power chip and the real-time power loss of the power module corresponding to each power module. Based on the real-time ambient temperature, the real-time junction temperature of the power chip, and the real-time power loss of the power module, the real-time junction-to-ambient equivalent thermal resistance of the power module is analyzed. Calculate the deviation ratio between the real-time junction-to-environment equivalent thermal resistance and the pre-stored standard junction-to-environment equivalent thermal resistance, and generate a fault warning based on the deviation ratio calculation result; Based on the real-time junction-to-environment equivalent thermal resistance, real-time ambient temperature, and the pre-stored maximum allowable junction temperature of the power chip, calculate the maximum allowable power loss of the power module under the current real-time ambient temperature. A load reduction protection strategy is generated based on the real-time power loss and maximum allowable power loss of the power module.
[0007] Furthermore, the real-time operating parameters include the casing temperature of the power module and the collector current, collector-emitter saturation voltage drop, duty cycle, single turn-on energy, single turn-off energy, and switching frequency of the IGBT chip in the power module.
[0008] Furthermore, the real-time junction temperature of the power chip corresponding to the power module and the real-time power loss of the power module are calculated, including: Calculate the conduction loss of the power module based on the collector current, collector-emitter saturation voltage drop, and duty cycle. Calculate the switching losses of the power module based on the single-turn-on energy, single-turn-off energy, and switching frequency; Based on the conduction loss and switching loss, the real-time power loss of the power module is generated. The real-time junction temperature of the power chip is generated based on the real-time power loss of the power module and the case temperature.
[0009] Furthermore, the formula for calculating the real-time power loss of the power module is as follows:
[0010]
[0011]
[0012] In the formula, For the real-time power loss of the power module, For conduction loss, For switching losses, For collector current, This is the collector-emitter saturation voltage drop. Duty cycle, Energy is activated only once. For single-time power shutdown, is the switching frequency; The calculation formula of the real-time junction temperature of the power chip is as follows:
[0013] In the formula, is the real-time junction temperature of the power chip, is the shell temperature, is the pre-stored junction-to-ambient thermal resistance.
[0014] Further, the calculation formula of the real-time junction-to-ambient equivalent thermal resistance is as follows:
[0015] In the formula, is the real-time junction-to-ambient equivalent thermal resistance, is the real-time junction temperature of the power chip, is the real-time ambient temperature, is the real-time loss power of the power module.
[0016] Further, the calculation formula of the maximum allowable loss power is as follows:
[0017] In the formula, is the maximum allowable loss power, is the maximum allowable junction temperature of the power chip, is the real-time ambient temperature, is the real-time junction-to-ambient equivalent thermal resistance.
[0018] Further, according to the real-time loss power of the power module and the maximum allowable loss power, a load shedding protection strategy is generated, including: According to the real-time loss power of the power module and the maximum allowable loss power, a safety margin is calculated:
[0019] In the formula, is the safety margin, is the maximum allowable loss power, is the real-time loss power of the power module; According to the safety margin, a risk level is generated; if , the risk level is low risk, and the load shedding protection strategy includes reducing the maximum speed and acceleration of the elevator, and prolonging the opening and closing door time of the elevator; if , the risk level is high risk, and the load shedding protection strategy includes adjusting the task allocation priority of each elevator.
[0020] Further, adjusting the task allocation priority of each elevator includes: obtaining cooperative scheduling data; the cooperative scheduling data includes current operating states of each elevator, current floors of each elevator, floors of call requests, and the number of people waiting for elevators at each floor; According to the cooperative scheduling data and the deviation ratio calculation result, the comprehensive cost generated by assigning the call request to any elevator is analyzed; According to the comprehensive cost corresponding to each elevator, a task allocation priority is generated.
[0021] The principles and advantages of the present application are: In the present scheme, by calculating the equivalent thermal resistance deviation ratio in real time, early warning can be issued in the early stage of heat dissipation performance degradation, effectively avoiding sudden shutdown caused by overheating of the power module, and ensuring the continuity of elevator operation. In addition, by dynamically calculating the maximum allowable loss power under the current environment, and generating a load reduction protection strategy accordingly, it can ensure that the power chip always works within the safety boundary of its maximum allowable junction temperature, avoiding the accelerated aging of the device due to long-term overheating or instantaneous thermal shock, and significantly improving the service life of the core drive hardware and the operation reliability of the entire elevator group. Specifically, the present scheme takes the maximum allowable loss power and the real-time loss power of the power module as the key parameters of group control scheduling. When the thermal safety margin of a single elevator is insufficient, the system can cooperatively adjust the task allocation from the group level, intelligently scheduling part of the load to the elevator with better thermal state, ensuring the safety of the hardware while optimizing the overall transportation efficiency of the system. BRIEF DESCRIPTION OF DRAWINGS
[0022] Figure 1 The flowchart of an embodiment of the elevator group management method based on thermal safety boundary prediction and cooperative scheduling. DETAILED DESCRIPTION
[0023] The following will be further described in detail through specific embodiments: Embodiment 1 An elevator group management method based on thermal safety boundary prediction and cooperative scheduling is executed by a thermal safety scheduling module deployed on a group control system main controller. The thermal safety scheduling module communicates data with the frequency converter controller of each elevator through CAN bus or Ethernet. As shown in the figure, Figure 1 The steps include: S100 collects real-time operating parameters of the power modules of each elevator and the real-time ambient temperature inside the control cabinet where each power module is located. The real-time operating parameters include the casing temperature of the power modules and the collector current, collector-emitter saturation voltage drop, duty cycle, single-turn-on energy, single-turn-off energy, and switching frequency of the IGBT chips within the power modules. Specifically, a digital temperature sensor is installed inside each elevator control cabinet, away from the heat source of the power modules, in a ventilated area to collect the real-time ambient temperature inside the control cabinet at a preset sampling frequency. The casing temperature is measured by an infrared temperature sensor tightly attached to the metal heat sink substrate of the power module. The collector current, collector-emitter saturation voltage drop, duty cycle, and switching frequency of the IGBT chips in the power modules are calculated and read in real-time by the current sensor, voltage sampling circuit, and DSP digital signal processor inside the inverter controller. The single-turn-on energy and single-turn-off energy are obtained through an IGBT parameter table pre-stored in the non-volatile memory of the inverter controller. This parameter table indexes the corresponding single-turn-on energy and single-turn-off energy by querying the collector current and junction temperature. In this embodiment, the default junction temperature is used for the initial query, and the real-time junction temperature is calculated in subsequent steps and fed back for more accurate iterative queries.
[0024] S200 calculates the real-time junction temperature of the power chip and the real-time power loss of the power module for each power module based on the real-time operating parameters of each power module; including: S201, calculate the conduction loss of the power module based on the collector current, collector-emitter saturation voltage drop, and duty cycle; the formula for calculating the conduction loss is as follows:
[0025] In the formula, For conduction loss, For collector current, This is the collector-emitter saturation voltage drop. This refers to the duty cycle; in this embodiment, , and All values are instantaneous values collected. In other embodiments of this application, the average value can also be taken over one PWM cycle for calculation.
[0026] S202, based on the single-turn-on energy, single-turn-off energy, and switching frequency, calculate the switching loss of the power module; the formula for calculating the switching loss is as follows:
[0027] In the formula, For switching losses, Energy is activated only once. For single-time power shutdown, The switching frequency.
[0028] S203, generating the real-time loss power of the power module according to the turn-on loss and the switching loss; the calculation formula of the real-time loss power of the power module is as follows:
[0029] In the formula, the real-time loss power of the power module, the turn-on loss, the switching loss.
[0030] S204, generating the real-time junction temperature of the power chip according to the real-time loss power of the power module and the shell temperature; the calculation formula of the real-time junction temperature of the power chip is as follows:
[0031] In the formula, the real-time junction temperature of the power chip, the shell temperature, the pre-stored junction-to-ambient thermal resistance. Wherein, the inherent property of the power module.
[0032] S300, analyzing the real-time junction-to-ambient equivalent thermal resistance of the power module according to the real-time environmental temperature, the real-time junction temperature of the power chip and the real-time loss power of the power module; the calculation formula of the real-time junction-to-ambient equivalent thermal resistance is as follows:
[0033] In the formula, the real-time junction-to-ambient equivalent thermal resistance, the real-time junction temperature of the power chip, the real-time environmental temperature, the real-time loss power of the power module. Wherein, the dynamic parameter, reflecting all factors affecting the heat dissipation efficiency such as the working state of the heat dissipation fan, the dust accumulation of the heat sink, the environmental ventilation condition, etc. The system records and updates in a sliding time window to smooth the instantaneous fluctuations.
[0034] S400, calculating the deviation ratio of the real-time junction-to-ambient equivalent thermal resistance and the pre-stored standard junction-to-ambient equivalent thermal resistance, and generating a fault warning according to the calculation result of the deviation ratio; the calculation formula of the deviation ratio is as follows:
[0035] In the formula, the deviation ratio of the real-time junction-to-ambient equivalent thermal resistance and the pre-stored standard junction-to-ambient equivalent thermal resistance, The pre-stored standard junction-to-ambient thermal resistance is a benchmark value measured under standard laboratory conditions.
[0036] If , it is determined that the heat dissipation system is abnormal, and the system sends a first-level fault warning to the remote control center; if , it is determined that the heat dissipation system is seriously deteriorated and has the risk of immediate overheating, the system sends a second-level fault warning to the monitoring center, and executes S500 to implement subsequent load reduction protection strategies. In other embodiments of the present application, if the generated fault warning is a first-level fault warning, the fault cause can be further analyzed and uploaded to the monitoring center. Specifically, if the fan speed signal is normal, but the real-time junction-to-ambient thermal resistance increases, it is determined that the fault cause is air duct blockage or dust accumulation on the heat sink; if the real-time junction-to-ambient thermal resistance rises sharply (the rising amplitude in unit time is greater than a preset first thermal resistance change rate threshold), and the fan speed feedback is abnormal or the vibration is abnormal, it is determined that the fault cause is fan failure; if the real-time junction-to-ambient thermal resistance rises slowly and monotonously (the rising amplitude in unit time is less than a preset second thermal resistance change rate threshold), and there is no other abnormality, it is determined that the fault cause is the aging of the thermal interface material, and the preset first thermal resistance change rate threshold is greater than the preset second thermal resistance change rate threshold.
[0037] S500, according to the real-time junction-to-ambient thermal resistance, the real-time ambient temperature and the pre-stored maximum allowed junction temperature of the power chip, the maximum allowed loss power of the power module at the current real-time ambient temperature is calculated; the calculation formula of the maximum allowed loss power is:
[0038] In the formula, the maximum allowed loss power is the maximum allowed junction temperature of the power chip is 150°C in the present embodiment, the real-time ambient temperature is the real-time junction-to-ambient thermal resistance.
[0039] S600, according to the real-time loss power of the power module and the maximum allowed loss power, a load reduction protection strategy is generated. Including: S601, according to the real-time loss power of the power module and the maximum allowed loss power, a safety margin is calculated:
[0040] In the formula, the safety margin is the maximum allowed loss power is the real-time loss power of the power module is S602, according to the safety margin, a risk level is generated.
[0041] If , the risk level is low risk, and the load reduction protection strategy includes reducing the maximum speed and acceleration of the elevator, prolonging the opening and closing time of the elevator door, and reducing the power loss of the power module by limiting the performance of the faulty elevator. Specifically, by adjusting the frequency converter parameters, the maximum operating speed of the corresponding elevator is reduced by 15% from the rated value; the maximum acceleration and deceleration rate of the corresponding elevator is reduced by 20% from the standard value; the total opening and closing time of the corresponding elevator is prolonged by 1-2 seconds, reducing the associated heat effect caused by frequent starting and stopping of the door motor.
[0042] If , the risk level is high risk, and the load reduction protection strategy includes adjusting the task allocation priority of each elevator, specifically: obtain collaborative scheduling data; the collaborative scheduling data includes the current operating state of each elevator, the current floor of each elevator, the floor of the call request, and the number of waiting passengers on each floor.
[0043] According to the collaborative scheduling data and the deviation ratio calculation result, the comprehensive cost generated by assigning the call request to any elevator is analyzed; in this embodiment, a neural network model based on the error back propagation algorithm is used to analyze and predict the comprehensive cost. This method can automatically learn the deep, nonlinear mapping relationship between various complex factors in the elevator group system and the scheduling cost without relying on an accurate mathematical model defined by humans. Specifically, first, a three-layer BP neural network model is constructed, including an input layer, a hidden layer, and an output layer. In this embodiment, the deviation ratio of the real-time junction-to-environment equivalent thermal resistance of each elevator to the pre-stored standard junction-to-environment equivalent thermal resistance, the current operating state of each elevator, the current floor of each elevator, the floor of the call request, and the number of waiting passengers on each floor are used as inputs to the input layer, so the input layer has 5 nodes, and the output is the predicted comprehensive cost generated when the call request is assigned to elevator i, so there are 1 nodes in total. The comprehensive cost is a dimensionless scalar for comparing the suitability of each elevator horizontally; for the hidden layer, the following formula is used to determine the number of hidden layer nodes in this embodiment: ; where L is the number of hidden layer nodes, n is the number of input layer nodes, m is the number of output layer nodes, and a is a number between 1 and 10, which is 6 in this embodiment, so the hidden layer has 9 nodes. BP neural network usually uses Sigmoid differentiable function and linear function as the excitation function of the network. In this embodiment, the hyperbolic tangent function tansig is selected as the excitation function of the hidden layer neurons, which has an output range of (-1, 1) and a zero-centered property, which can make the convergence speed faster; since the comprehensive cost is a regression prediction value, its theoretical range is unlimited, therefore, a pure linear function purelin is selected as the excitation function of the output layer neurons.
[0044] A task allocation priority is generated according to the comprehensive cost corresponding to each elevator. Specifically, the smaller the comprehensive cost, the higher the task allocation priority; all elevators in an idle state or having a compatible running direction with the call request are screened, and the call request is allocated to the elevator with the smallest comprehensive cost.
[0045] The scheme enables the scheduling system to not only take corresponding measures when the elevator is about to overheat, but also to take early prevention, to intervene and protect the elevator whose heat dissipation performance has been significantly deteriorated although the high-temperature alarm has not been triggered, and to improve the reliability and predictive maintenance capability of the entire elevator group system.
[0046] Embodiment 2: Embodiment 2 has the same basic principle as Embodiment 1, and the difference lies in that in Embodiment 2, the comprehensive cost generated by allocating the call request to any elevator is analyzed according to the cooperative scheduling data and the bias ratio calculation result, including: A cost function is constructed; for each elevator that can respond to the call request R, the comprehensive cost is calculated to quantify the total cost of allocating the task to the elevator i. In this embodiment, the elevator whose current running state is idle or whose running direction is compatible with the call request is defined as the elevator that can respond to the call request. The cost function is defined as:
[0047]
[0048]
[0049]
[0050] In the formula, is the comprehensive cost; , , , are pre-stored weight coefficients for balancing the importance of service efficiency, energy consumption, instantaneous heat risk, and long-term health degree in decision-making; To predict the waiting time, in this embodiment, for the currently calculated elevator, the predicted path is obtained through path simulation, specifically, the system inserts the floor of the request R into the current task sequence of the elevator according to the preset dispatching rule, generates a predicted path, and simulates and calculates the total time required from the current time to serve the request; is a time normalization factor, which is the average value of the predicted waiting times of all current elevators, used to eliminate the dimension; is the predicted power consumption; and are pre-stored coefficients obtained by linear regression fitting of historical data, representing the incremental power consumption per meter of running distance and the incremental power consumption per stop, respectively; To estimate the increment of running distance, the difference between the total length of the current path and the predicted total length of the path after the insertion of the call request R is calculated, and the total length of the path is the cumulative sum of the absolute values of the differences between adjacent floors in the sequence; To estimate the increment of the number of stops, the difference between the number of stops in the predicted path and the number of stops in the current path is calculated for the elevator; is a power consumption normalization factor, which is the average of the predicted power consumption of all elevators; is a thermal risk cost component, which is a penalty term related to the safety margin of the elevator, and is defined as a piecewise function; is a health degree cost component, which is a penalty term related to the corresponding deviation ratio of the elevator.
[0051] The above is only an embodiment of the present application, and common knowledge of specific structures and characteristics in the scheme is not described in detail. Those skilled in the art know all the common technical knowledge in the field of the present application before the filing date or the priority date, can know all the prior art in the field, and have the ability to apply conventional experimental means before that date. Those skilled in the art can improve and implement the present scheme based on their own abilities under the guidance of the present application. Some typical known structures or known methods should not be an obstacle for those skilled in the art to implement the present application. It should be noted that those skilled in the art can make several modifications and improvements without departing from the structure of the present application, and these should be considered as the protection scope of the present application. The protection scope of the present application should be subject to the content of its claims, and the specific implementation in the specification can be used to explain the content of the claims.
Claims
1. An elevator group management method based on thermal safety boundary prediction and cooperative scheduling, characterized by: The method comprises the following steps: collecting real-time running parameters of power modules of each elevator and real-time ambient temperature in control cabinets where the power modules are located; calculating real-time junction temperature of power chips corresponding to each power module and real-time loss power of the power module according to the real-time running parameters of the power module; analyzing real-time junction-to-ambient equivalent thermal resistance of the power module according to the real-time ambient temperature, the real-time junction temperature of the power chip and the real-time loss power of the power module; calculating deviation ratio of the real-time junction-to-ambient equivalent thermal resistance and pre-stored standard junction-to-ambient equivalent thermal resistance, and generating a fault early warning according to the deviation ratio calculation result; calculating maximum allowable loss power of the power module under the current real-time ambient temperature according to the real-time junction-to-ambient equivalent thermal resistance, the real-time ambient temperature and the pre-stored maximum allowable junction temperature of the power chip; generating a load reduction protection strategy according to the real-time loss power of the power module and the maximum allowable loss power. 2.The elevator group management method based on thermal safety boundary prediction and cooperative scheduling according to claim 1, characterized in that: The real-time running parameters include shell temperature of the power module and collector current, collector-emitter saturation voltage drop, duty cycle, single opening energy, single closing energy and switching frequency of IGBT chips in the power module. 3.The elevator group management method based on thermal safety boundary prediction and cooperative scheduling according to claim 2, characterized in that: The calculation of the real-time junction temperature of the power chip corresponding to the power module and the real-time loss power of the power module comprises: calculating conduction loss of the power module according to the collector current, the collector-emitter saturation voltage drop and the duty cycle; calculating switching loss of the power module according to the single opening energy, the single closing energy and the switching frequency; generating real-time loss power of the power module according to the conduction loss and the switching loss; generating real-time junction temperature of the power chip according to the real-time loss power of the power module and the shell temperature.
4. The elevator group management method based on thermal safety boundary prediction and cooperative scheduling according to claim 3, characterized in that: The calculation formula of the real-time loss power of the power module is as follows: wherein Ploss is the real-time loss power of the power module, Pcon is the conduction loss, Psw is the switching loss, Ic is the collector current, Vce is the collector-emitter saturation voltage drop, D is the duty cycle, Eon is the single turn-on energy, Eoff is the single turn-off energy, fsw is the switching frequency; The calculation formula of the real-time junction temperature of the power chip is as follows: wherein Tj is the real-time junction temperature of the power chip, Tcase is the case temperature, Rjc is the pre-stored junction-to-case thermal resistance.
5. The elevator group management method based on thermal safety boundary prediction and cooperative scheduling according to claim 1, characterized in that: The calculation formula of the real-time junction-to-ambient equivalent thermal resistance is as follows: wherein, is the real-time junction-to-ambient thermal resistance, is the real-time junction temperature of the power chip, is the real-time ambient temperature, is the real-time power module loss power. 6.The elevator group management method based on thermal safety boundary prediction and cooperative scheduling according to claim 1, characterized in that: The calculation formula of the maximum allowable loss power is as follows: In the formula, is the maximum allowed power dissipation, is the maximum allowed junction temperature of the power chip, is the real-time ambient temperature, is the real-time junction-to-ambient equivalent thermal resistance.
7. The elevator group management method based on thermal safety boundary prediction and cooperative scheduling according to claim 1, characterized in that: The generation of the load reduction protection strategy according to the real-time loss power of the power module and the maximum allowable loss power comprises: calculating safety margin according to the real-time loss power of the power module and the maximum allowable loss power; In the formula, is a safety margin, is the maximum allowed power loss, is the real-time power loss of the power module; According to the safety margin, a risk level is generated; if the risk level is low risk, and the load shedding protection strategy includes reducing the maximum speed and acceleration of the elevator, and prolonging the opening and closing door time of the elevator; if the risk level is high risk, and the load shedding protection strategy includes adjusting the task allocation priority of each elevator. 8.The elevator group management method based on thermal safety boundary prediction and cooperative scheduling according to claim 7, characterized in that: adjusting task allocation priority of each elevator comprises: obtaining cooperative scheduling data; the cooperative scheduling data includes current running state of each elevator, current floor of each elevator, floor of a call request and number of waiting passengers of each floor; analyzing comprehensive cost generated by allocating the call request to any elevator according to the cooperative scheduling data and the deviation ratio calculation result; generating task allocation priority according to the comprehensive cost corresponding to each elevator.