Parameter adjustment method, system and device of elevator energy management system

Abstract

This invention discloses a parameter adjustment method, system, and device for an elevator energy management system, relating to the field of intelligent elevators. The method includes: acquiring the elevator's load data and running direction; calculating the current load rate and the load change rate within a preset short-term window based on the load data; determining the operating conditions based on the current load rate, load change rate, and running direction, including normal operating conditions and extreme switching conditions caused by drastic changes in load rate within a short period; invoking corresponding charging and discharging parameters according to the current operating conditions of the elevator to recover energy generated during elevator operation; and acquiring current and historical operating data of the elevator under the current operating conditions. This invention solves the problem that fixed parameters or simple threshold logic cannot promptly adjust the operating state of the energy storage system when facing such extreme conditions by real-time identification of extreme conditions based on load change rate and dynamic parameter adjustment oriented towards specific scenarios.

Description

Technical Field

[0001] This invention relates to the field of intelligent elevators, and more particularly to a parameter adjustment method, system, and device for an elevator energy management system. Background Technology

[0002] In elevator applications in factory settings, traditional elevator energy management algorithms generally employ fixed parameter control or simple threshold-triggered control logic, which has significant technical shortcomings. Firstly, they lack specific adaptation for extreme conditions in factory settings where elevator loads frequently and instantaneously switch (such as sudden increases / decreases in load due to cargo loading / unloading or continuous multi-floor transport). Faced with such extreme conditions, fixed parameters or simple threshold logic cannot adjust the energy storage system's operating status in a timely manner. Secondly, traditional solutions have weak scenario recognition capabilities, only able to roughly classify basic operating conditions such as no-load and full-load, lacking targeted analysis of specific load characteristics in factories. This results in control strategies that are too general and lack adaptability. These shortcomings directly lead to traditional solutions failing to balance energy recovery efficiency and energy storage device lifespan—under extreme conditions, either rigid parameters result in insufficient regenerated energy absorption, or the failure to match load switching characteristics causes the energy storage device to be subjected to overload shocks, accelerating device aging. This makes it difficult to meet the dual requirements of high efficiency and reliability for elevator energy management systems in factory settings. Summary of the Invention

[0003] The main objective of this invention is to provide a parameter adjustment method, system, and device for an elevator energy management system, which aims to solve the problem that fixed parameters or simple threshold logic cannot adjust the operating state of the energy storage system in a timely manner when facing such extreme working conditions.

[0004] To achieve the above objectives, this invention proposes a parameter adjustment method for an elevator energy management system, comprising: Obtain the elevator's load data and direction of travel; Based on the load data, calculate the current load rate and the load change rate within a preset short window; Based on the current load rate, load change rate and running direction, the current operating condition scenario of the elevator is identified. The operating condition scenario includes normal operating conditions and extreme switching conditions caused by a sudden change in load rate in a short period of time. Based on the current operating conditions of the elevator, the corresponding charging and discharging parameters are invoked to control the energy storage device to recover the electrical energy generated during the elevator operation. The system acquires current and historical operating data of the elevator under the current operating conditions, and optimizes and updates the charging and discharging parameters corresponding to the current operating conditions of the elevator based on the current and historical operating data.

[0005] Furthermore, the extreme switching conditions include a fully loaded downward switching condition and an unloaded upward switching condition; identifying the current elevator operating condition scenario based on the current load rate, load change rate, and operating direction includes: If the elevator is descending and the current load rate is higher than the full load threshold, then proceed to the descending scenario judgment: If the load change rate is lower than the first change rate threshold, it is identified as a normal full-load downlink condition; If the load change rate is higher than or equal to the first change rate threshold, it is identified as an extreme switching state from no load to full load, and a full load downlink condition. If the elevator is moving upwards and the current load rate is below the no-load threshold, then proceed to the upward movement scenario for judgment: If the load change rate is lower than the second change rate threshold, it is identified as a normal no-load uplink condition; If the load change rate is higher than or equal to the second change rate threshold, it is identified as an extreme switching state from full load to no load, and the no load uplink condition is identified. If the current load rate is between the no-load threshold and the full-load threshold, it is identified as a normal transitional operating condition. If the elevator descends and the current load rate is lower than or equal to the no-load threshold, it is identified as a normal no-load descending condition. If the elevator is moving upwards and the current load rate is higher than or equal to the full load threshold, it is identified as a normal full load upward operation.

[0006] Furthermore, the charging and discharging parameters include charging power and discharging power. The step of calling the corresponding charging and discharging parameters based on the current operating conditions of the elevator to recover energy generated during elevator operation includes: When the condition is identified as a normal full-load downlink operation, the energy storage device is controlled to recover energy at the first charging power. When the extreme switching state full-load downlink condition is identified, the energy storage device is controlled to recover energy with a second charging power, which is higher than the first charging power. When the condition is identified as a normal no-load uplink condition, the energy storage device is controlled to supply power at the first discharge power. When an extreme switching state with no-load uplink operation is identified, the energy storage device is controlled to supply power at a second discharge power, which is lower than the first discharge power.

[0007] Furthermore, the method also includes: When identified as a normal transitional operating condition, the energy storage device is controlled to recover energy at a third charging power, which is lower than the first charging power; and / or, the energy storage device is controlled to supply power at a third discharging power, which is lower than the first discharging power. When the condition is identified as a normal no-load downlink operating condition, the energy storage device is controlled to recover energy at a fourth charging power, which is lower than the first charging power. When the operation is identified as a normal full-load uplink condition, the energy storage device is controlled to supply power at a fourth discharge power, which is higher than the first discharge power.

[0008] Furthermore, the charging and discharging parameters also include charging duration, wherein: The charging time set for the extreme switching state full-load downlink condition is greater than the charging time set for the normal full-load downlink condition; The charging time set for normal full-load downlink operation is greater than the charging time set for normal transition operation during downlink; The charging time set during downlink in normal transition conditions is greater than or equal to the charging time set during normal no-load downlink conditions.

[0009] Furthermore, the charge / discharge parameters also include a state-of-charge protection threshold, wherein: The state of charge protection threshold set for extreme switching no-load uplink conditions is higher than the state of charge protection threshold set for normal no-load uplink conditions. The state-of-charge protection threshold set for the normal no-load uplink condition is higher than the state-of-charge protection threshold set for the normal transition condition during uplink. The state-of-charge protection threshold set during the normal transition operation is higher than or equal to the state-of-charge protection threshold set during the normal full-load uplink operation.

[0010] Furthermore, the optimization and updating of the charging and discharging parameters corresponding to the current operating conditions of the elevator based on current and historical operating data includes: The current operating data and historical operating data are input into the machine learning model to optimize the charging and discharging parameters corresponding to the current operating conditions of the elevator, and the optimized charging and discharging parameters are output. Update the optimized charging and discharging parameters to match the charging and discharging parameters corresponding to the current operating conditions of the elevator.

[0011] Furthermore, the current operating data includes at least one of the following: load data, running direction, running speed, state of charge of the energy storage device, temperature of the energy storage device, voltage and current of the energy storage device, real-time charging and discharging power of the energy storage device, DC bus voltage of the elevator inverter, and power grid interaction power.

[0012] The present invention also proposes a parameter adjustment device for an elevator energy management system, comprising: The system includes a memory, a processor, and a parameter adjustment program stored in the memory and executable on the processor, the parameter adjustment program being configured to implement the steps of a parameter adjustment method for an elevator energy management system.

[0013] The present invention also proposes an elevator energy management system, which includes an energy storage device, an elevator frequency converter, a data acquisition module and a central control module; The energy storage device includes a battery unit and a supercapacitor unit, which are connected in parallel to the DC bus of the elevator frequency converter. The data acquisition module is used to acquire the elevator's load data, running direction, and current running data; The central control module is communicatively connected to the data acquisition module, the energy storage device, and the elevator frequency converter.

[0014] This invention solves the problem that fixed parameters or simple threshold logic cannot adjust the operating state of the energy storage system in a timely manner when facing such extreme switching conditions by real-time identification of extreme operating conditions based on load change rate and dynamic parameter adjustment according to the scenario. This ensures that the instantaneous high-power regenerative energy is quickly and fully absorbed under extreme switching conditions. Attached Figure Description

[0015] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of the module structure of the elevator energy management system according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating the parameter adjustment method for the elevator energy management system according to an embodiment of the present invention.

[0018] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0019] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of the present invention and are not intended to limit the present invention.

[0020] To better understand the technical solution of the present invention, a detailed description will be provided below in conjunction with the accompanying drawings and specific embodiments.

[0021] This embodiment takes a material handling elevator in a factory as an example. The elevator has a rated load of 1000kg (M=1000kg) and is equipped with a lithium battery energy storage system with a rated capacity of 20kWh and a set of 100F supercapacitor modules as a collaborative energy storage unit.

[0022] refer to Figure 1 , Figure 1 This is a schematic diagram of the module structure of the elevator energy management system according to an embodiment of the present invention; like Figure 1 As shown, the elevator energy management system in this embodiment includes: a data acquisition module, a central control module, an elevator frequency converter, and an energy storage device. Data acquisition module 10: Specifically, it consists of a gravity sensor (range 0-1500kg) installed at the bottom of the car, a running direction signal provided by the elevator main control cabinet, and a battery management system (BMS) built into the energy storage converter (PCS). The gravity sensor collects load data at a frequency of 10Hz, and the BMS reports the SOC, voltage, and temperature of the energy storage system at a frequency of 1Hz.

[0023] The central control module 20 is implemented by an embedded industrial controller, and the program logic running within it includes a scene recognition module 21 and a dynamic adaptation parameter tuning module 22.

[0024] The scene recognition module 21 is used to calculate the current load rate and load change rate based on the load data. Based on the current load rate, load change rate and running direction, it identifies the current operating conditions of the elevator. The operating conditions include normal operating conditions and extreme switching conditions caused by a sudden change in load rate in a short period of time. The dynamic adaptation parameter adjustment module 22 is used to generate corresponding energy storage device charging and discharging control commands based on the operating conditions.

[0025] Energy storage device 30: includes a battery unit and a supercapacitor unit connected in parallel. This parallel combination is connected to the DC bus of the elevator inverter 40 to recover regenerative energy during elevator descent or to provide auxiliary power for elevator ascent.

[0026] Elevator frequency converter 40: Its DC bus terminal is connected to the energy storage device, and its AC terminal is connected to the elevator drive motor and the power grid. During elevator operation, the elevator frequency converter controls the operation of the elevator motor and feeds the regenerative energy generated by the elevator motor back to the DC bus, where it is absorbed by the energy storage device 30; when the elevator needs power, the energy storage device provides electrical energy to the frequency converter through the DC bus.

[0027] refer to Figure 2 , Figure 2The flowchart provided for the parameter adjustment method of the elevator energy management system in an embodiment of the present invention. In this embodiment, a typical handling task of the elevator is used as an example to explain the solution: it travels upward without load from the 1st floor to the 3rd floor, loads 800 kg of goods, and then travels downward fully loaded back to the 1st floor.

[0028] The parameter adjustment method of the elevator energy management system in an embodiment of the present invention includes steps S10 to S50: Step S10, obtain the load data and running direction of the elevator.

[0029] Specifically, the central control module continuously reads the gravity data Mt of the gravity sensor and the running direction of the elevator.

[0030] Step S20, based on the load data, calculate the current load rate and the load change rate within a preset short-time window.

[0031] Furthermore, the methods for calculating the load rate and the load change rate include: assuming the rated load of the elevator is M and the current load is Mt, the formula for calculating the current load rate L is: L = Mt / M; assuming the load of the elevator at time t is M0 and the load changes to M1 within the short-time window Δt (Δt ≤ 3 s), then the absolute load change amount Δm = ∣M0 M1∣, and the formula for calculating the relative load change rate c is: c = Δm / M0; set the load rate thresholds: the first load rate threshold L1 = 10%, the second load rate threshold L2 = 80%; set the relative load change rate threshold c0 = 70%.

[0032] In this embodiment, calculate the current load rate L: L = Mt / 1000.

[0033] Calculate the load change rate c: take 1 second as the time window (Δt = 1 s) and continuously record the load values. For example, when loading goods on the 3rd floor, record the load M0 = 50 kg (no load) at time t0 and the load M1 = 850 kg (fully loaded) at time t1. Then the absolute change amount Δm = |50 - 850| = 800 kg, and the load change rate c = 800 / 50 = 1600%.

[0034] Step S30, according to the current load rate, load change rate and running direction, identify the current operating condition scenario of the elevator.

[0035] Furthermore, identify and output the real-time operating condition scenario of the elevator. The classification of the operating condition scenarios is as follows: When the real-time load rate L ≤ L1, the load change rate c < c0 and the running direction is upward, the load state is no load, and the corresponding operating condition scenario is normal no-load upward; When the real-time load rate L ≤ L1, the load change rate c ≥ c0, and the running direction is upward, the load state is no-load (extreme switching), and the corresponding working condition scenario is no-load upward in the extreme switching state; When the real-time load rate L ≥ L2, the load change rate c < c0, and the running direction is downward, the load state is full-load, and the corresponding working condition scenario is normal full-load downward; When the real-time load rate L ≥ L2, the load change rate c ≥ c0, and the running direction is downward, the load state is full-load (extreme switching), and the corresponding working condition scenario is full-load downward in the extreme switching state; When L1 < L ≤ L2 (the load change rate is not used as a judgment condition), and the running direction is upward or downward, the load state is the transition state, and the corresponding working condition scenario is the normal transition condition; When the real-time load rate L ≤ L1 (the load change rate is not used as a judgment condition) and the running direction is downward, the load state is no-load, and the corresponding working condition scenario is normal no-load downward; When the real-time load rate L ≥ L2 (the load change rate is not used as a judgment condition) and the running direction is upward, the load state is full-load, and the corresponding working condition scenario is normal full-load upward.

[0036] In this embodiment, the dynamic adaptation parameter adjustment module is used to formulate an adapted charge-discharge strategy according to the working condition scenario output by the scenario recognition module, and complete parameter iteration optimization in combination with the execution feedback.

[0037] Among them, the inputs are: the working condition scenario output by the scenario recognition module, and the strategy execution result feedback by the instruction execution module; Among them, the outputs are: the charge-discharge strategy adapted to the current elevator working condition, including core parameters such as charge-discharge power, charge duration, SOC protection threshold, working mode, etc.

[0038] Furthermore, the specific parameters corresponding to each working condition scenario are as follows: Normal no-load upward working condition: The charging function is disabled, the discharge power is 5 - 8 kW, there is no actively set charge duration, the SOC protection threshold is 20%, and the working mode is the lithium battery independent mode; Extreme switching state no-load upward working condition: The charging function is disabled, the discharge power is 2 - 3 kW, there is no actively set charge duration, the SOC protection threshold is 30%, and the working mode is the lithium battery independent / grid priority switching mode; Normal full-load downward working condition: The charging power is 5 - 8 kW, the discharging function is disabled, the charge duration is 10 s, the SOC protection threshold is 20%, and the working mode is the lithium battery independent mode; Extreme switching state full-load downlink condition: The charging power is 8 - 12 kW (8 kW for elevators below 20 kW and 12 kW for elevators above 30 kW), the discharging function is disabled, the charging duration is 15 s, the SOC protection threshold is 20%, and the working mode is the collaborative mode of lithium battery + supercapacitor; Conventional transition condition: The charging power is 3 - 5 kW, the discharging power is 3 - 5 kW, the charging duration is 8 s, the SOC protection threshold is 20%, and the working mode is the independent mode of lithium battery; Conventional no-load downlink condition: The charging power is 2 - 3 kW, the discharging function is disabled, the charging duration is 5 s, the SOC protection threshold is 20%, and the working mode is the independent mode of lithium battery; Conventional full-load uplink condition: The charging function is disabled, the discharging power is 5 - 8 kW, there is no actively set charging duration, the SOC protection threshold is 20%, and the working mode is the independent mode of lithium battery.

[0039] In this embodiment, the no-load threshold L1 = 10%, the full-load threshold L2 = 80%, and the change rate threshold c0 = 70%.

[0040] When the elevator goes up空载 (Mt = 30 kg, L = 3% < L1) from the 1st floor, the load is stable (c = 0). Combining with the upward direction, the system identifies it as the "conventional no-load upward" condition.

[0041] Step S40, call the corresponding charge and discharge parameters according to the current operating condition scenario of the elevator, and control the energy storage device to recover the electric energy generated during the elevator operation.

[0042] Furthermore, the charge and discharge strategies for each condition are as follows: Conventional no-load upward: The core goal is to balance power supply stability and energy utilization rate and meet the basic power demand for no-load upward; adopt the independent mode of lithium battery and increase the basic discharge power from 3 - 5 kW to 5 - 8 kW.

[0043] Extreme switching state no-load upward: The core goal is to limit the discharge power and avoid deep discharge of the lithium battery; adopt the independent mode of lithium battery and lower the discharge power to 2 - 3 kW. When the SOC of the energy storage system < 30%, immediately switch to the power grid priority power supply mode.

[0044] Conventional full-load downward: The core goal is to balance the regeneration energy absorption efficiency and the temperature rise of the energy storage; adopt the independent mode of lithium battery and increase the charging power to 5 - 8 kW, and set the charging duration to 10 s to ensure full absorption of the regeneration energy.

[0045] Extreme switching state full load downlink: The core objective is to maximize the absorption of instantaneous high-power regenerative energy and avoid energy dissipation; the charging power is increased to 8-12kW and the charging time is set to 15s; when the charging power gap is ≥1kW, the supercapacitor is activated, and the lithium battery and supercapacitor collaborative mode is enabled, with the supercapacitor taking the lead in absorbing the instantaneous excess energy.

[0046] Normal transitional operating conditions: The core objective is to match the energy supply and demand of the transitional load and avoid system fluctuations; an independent lithium battery mode is adopted, with both charging and discharging power set to the base power of 3-5kW and charging time set to 8s.

[0047] Normal no-load downlink: The core objective is to efficiently absorb a small amount of regenerative energy and avoid ineffective charging; adopt the lithium battery independent mode, reduce the charging power to 2-3kW, and set the charging time to 5s.

[0048] Normal full-load uplink: The core objective is to provide sufficient discharge power to ensure the power requirements for full-load uplink; adopting an independent lithium battery mode, the discharge power is increased to 5-8kW.

[0049] In this embodiment, when loading goods on the 3rd floor, the elevator is in a stationary, ready-to-descend state. When it is detected that the load increases dramatically from 50kg to 850kg within 1 second (i.e., L=85%>L1, c=1600%>c0), and the elevator starts descending in the next moment, the system immediately identifies it as "extreme switching state full-load descent".

[0050] If the elevator descends from the 3rd floor fully loaded (Mt=850kg) to the 1st floor, unloads the cargo to 30kg, and then ascends, the system will recognize the "extreme switching state from full load to empty load, empty load ascending condition" at the moment the upward movement starts.

[0051] The central control module stores a preset strategy mapping table. Based on the identification results, the parameters are dynamically adjusted as follows: Identified as "normal no-load upward movement": Preset parameters are invoked, and a command is sent to the energy storage converter: the maximum discharge power is set to 7kW, and the SOC protection threshold is 20%. The energy storage system independently supplies power for the elevator's upward movement.

[0052] Identified as "Extreme switching state full load downlink": Invoke the enhanced parameters for this condition and immediately send a command to the energy storage converter to dynamically adjust the charging power limit from the usual 6kW to 10kW. Also extend the charging time to 15 seconds to ensure coverage of the energy recovery window throughout the downlink process.

[0053] Furthermore, the central control module estimates the regenerative energy power in real time. Since the elevator's rated power is 30kW, the instantaneous regenerative power during a fully loaded descent may reach 25kW, while the maximum instantaneous charging power of the lithium battery pack is limited to 12kW, resulting in a 13kW power shortfall. At this point, the supercapacitor controller is activated via RS485, connecting it in parallel with the lithium battery to jointly absorb this 25kW instantaneous peak power. Furthermore, the supercapacitor absorbs most of the peak energy within 2-3 seconds, and the lithium battery continues to absorb the remaining energy, perfectly avoiding the activation of the braking resistor.

[0054] Step S50: Obtain the current operating data and historical operating data of the elevator under the current operating conditions, and optimize and update the charging and discharging parameters corresponding to the current operating conditions of the elevator based on the current operating data and historical operating data.

[0055] Specifically, after the adaptive optimization process system has been running for a period of time, the adaptive optimization module starts the adaptive algorithm to complete the iterative optimization of parameters.

[0056] In this embodiment, four core adaptive algorithms are applied to optimize charging and discharging parameters iteratively using historical operating data, with the dual objectives of efficient energy utilization and ensuring the lifespan of energy storage equipment, for different elevator operating conditions. The core principles, applicable operating conditions, and parameter optimization rules of each algorithm are as follows: Specifically, this example proposes the Q-Learning reinforcement learning algorithm. Based on the trial-and-error learning mechanism of reinforcement learning, this algorithm uses different combinations of charging and discharging parameters of the energy storage device as "actions" and the comprehensive indicators of energy utilization rate and equipment loss of the energy storage device under specific elevator operating conditions as "reward values." By iteratively updating the value scoring Q-function, the optimal mapping relationship between the elevator operating conditions and the charging and discharging parameter combinations is obtained. The update formula for the Q-function is: Q(s t +1,a t +1)=Q(s t ,a t )+α[r t +γmaxaQ(s t +1,a) Q(s t ,a t )]; Where Q(s) t ,a t Q(s) represents the value score of the elevator's operating conditions and charging / discharging parameter combination at time t. t+1 ,a t+1 () represents the value score at time t+1.

[0057] Where s tLet t be the elevator operating condition at time t, corresponding to the seven operating condition states of this invention; a t The combination of charging and discharging parameters for the energy storage device at time t specifically includes two core parameters: charging power P and charging duration T. For example, under the full-load downlink condition of this invention, the basic settings are charging power P = 8kW and charging duration T = 15s; α is the learning rate, used to control the update amplitude of the charging and discharging parameters, and its value ranges from 0.1 to 0.3; γ is the discount factor, used to assign weight to the influence of future rewards on the current decision, and its value ranges from 0.7 to 0.9; r t The reward value at time t is calculated by weighting the energy utilization rate η of the energy storage device and relevant equipment loss indicators (such as battery temperature rise ΔT) using characteristic weighting factors. The specific calculation formula is r. t =k1×η+k2×△T, where k1 and k2 are feature weighting factors; a is the optimal combination of charging and discharging parameters corresponding to the next operating condition. For example, after correction, its value can be set as charging power P=10kW and charging time T=15s.

[0058] This algorithm is applicable to elevator operating conditions in this invention where the energy absorption rate needs to be improved while balancing equipment losses, specifically: extreme switching state from no load to full load downward operating condition and normal transition downward operating condition.

[0059] Furthermore, with the comprehensive reward objective of "maximizing the energy absorption rate of energy storage devices + minimizing equipment losses," the following rules are implemented: Furthermore, the reward value r t Based on the energy absorption rate η of the energy storage device and the temperature rise ΔT of the equipment, the specific formula is as follows:

[0060] Specifically, when the energy absorption rate η≥90%, the reward value increases by 0.1 for every 1% increase; when the equipment temperature rises by more than 5℃, the reward value is reduced proportionally for the over-temperature portion; when the energy absorption rate <90%, the reward value is reduced by -0.2 directly.

[0061] The initial values ​​are based on preset charge and discharge parameters: Full-load downlink operation: charging power P=8kW, charging time T=15s; Downward operating condition under transitional load: Charging power P=3 5kW, charging time T=8s; calculate the bonus value after each charge / discharge cycle. If the bonus value for three consecutive cycles of the same condition is ≥0.8, then lock the current parameters; otherwise, fine-tune the parameters in the direction of increasing the bonus value. The adjustment step size is: charging power step size 0.2kW, charging time step size 1s, and the formula is:

[0062] If the reward value r at time t+1 t+1 The reward value r is greater than time t. t This indicates the current charging power P t The value of P is not optimal, and increasing the power helps improve overall efficiency (increased energy absorption rate / reduced equipment loss). Therefore, the charging power P at time t+1 is... t+1 In P t The base is increased by 0.2kW, with a step size of 0.2kW representing the optimal fine-tuning amplitude calibrated experimentally, balancing the efficiency and stability of parameter adjustment; if the reward value r at time t+1... t+1 The reward value r less than time t t This indicates that the current charging power Pt value is not optimal. For example, excessive power may lead to excessive equipment temperature rise, or insufficient power may lead to inadequate energy recovery. Therefore, the charging power P at time t+1 should be adjusted accordingly. t+1 In P t Reduce the power by 0.2kW to decrease the power output and verify whether this can increase the reward value; if the reward value r at time t+1 is... t+1 The reward value r is greater than time t. t This indicates the current charging time T. t There is still room for optimization in the value of , and extending the duration will help improve the overall efficiency. Therefore, the charging duration T at time t+1 will be adjusted. t+1 In T t Adding 1 second to the base timeframe helps prevent parameter oscillations caused by excessively large duration adjustments; if the reward value r at time t+1... t+1 The reward value r less than time t t This indicates the current charging time T. t The value of T is unreasonable (e.g., too long a duration leads to energy waste, or too short a duration leads to insufficient energy recovery), therefore the charging duration T at time t+1 is changed. t+1 In T t Reduce the time by 1 second to shorten the duration and verify whether it can increase the reward value.

[0063] When the parameters are adjusted, if the following conditions are met for 5 consecutive identical operating conditions, the parameters are considered to have converged and the iteration is stopped: the energy absorption rate of the energy storage device fluctuates by ≤2%; and the equipment loss (such as temperature rise) is within the safe threshold range.

[0064] In this embodiment, a least squares support vector machine (LSSVM) algorithm for dynamic parameter adjustment of elevator energy management system is also provided. It is used to construct a nonlinear mapping model between elevator operating conditions and energy storage device charging and discharging power, so as to realize accurate prediction and dynamic correction of charging and discharging parameters with the goal of reducing battery discharge loss. Specifically, it includes core principles, parameter definitions, applicable operating conditions and charging and discharging parameter learning and calculation rules. In this embodiment, the algorithm simplifies the support vector calculation process using the least squares optimization criterion, constructs a nonlinear mapping model between input and output parameters, trains the model using historical operating data, and achieves accurate prediction of the optimal charging and discharging parameters for the energy storage device; the expression of the nonlinear mapping model is: ; Where x is the model input parameter, specifically the current operating condition of the elevator and the current state of charge (SOC) value of the energy storage system, corresponding to the seven operating conditions of this invention, and the current SOC value of the energy storage system is 30% for example; φ is the kernel function, used to map the input parameter x to a high-dimensional feature space, and the radial basis function is selected as the kernel function in this scenario; ω is the weight vector in the high-dimensional feature space, which exists in the form of a column vector, and its specific value is obtained by training the model based on the historical operating data of the elevator energy management system; b is the bias term of the mapping model, and its specific value is also obtained by training the model based on the historical operating data; y is the model output parameter, that is, the optimal discharge power P of the energy storage device, and the exemplary value is P=10kW.

[0065] Among them, the least squares optimization criterion is used to replace the inequality constraints of the traditional support vector machine, which reduces the complexity of model solution and adapts to the real-time parameter adjustment requirements of the elevator energy management system.

[0066] In this embodiment, the Least Squares Support Vector Machine (LSSVM) algorithm is applicable to the operating conditions in the elevator energy management system where it is important to reduce battery discharge loss and ensure operational stability, specifically: the extreme switching state from full load to no load during the no-load upward operation and the normal transition upward operation.

[0067] The Least Squares Support Vector Machine (LSSVM) algorithm aims to minimize the risk of battery discharge loss and executes the following rules for learning and adjusting charging and discharging parameters: The LSSVM model's input is configured as "current elevator operating condition and current state of charge (SOC) of the energy storage system," and its output is configured as "optimal discharge power of the energy storage device," thus achieving a directional mapping from operating condition and SOC state to optimal discharge power.

[0068] The data training logic collects at least 100 sets of historical operating condition data from the elevator energy management system. The historical data must cover the discharge power of the energy storage device and the corresponding battery loss data under different operating conditions and different SOC states. After normalizing the historical data, it is input into the LSSVM model for training to establish the mapping relationship between discharge power and battery degradation rate, and to solve the model parameters (ω, b).

[0069] Real-time data on the elevator's current operating conditions and the energy storage system's current State of Charge (SOC) value are collected and input into a trained LSSVM model to predict battery loss risk under different discharge powers. The discharge power with the lowest battery loss risk is selected as the initial optimal discharge power, with an initial value of 1-2kW under extreme no-load upward operation and 3-5kW under transitional load upward operation. If the predicted battery loss risk R... loss Exceeding the preset warning threshold R warn Then, the optimal discharge power is dynamically corrected, and the correction formula is:

[0070] Among them, P adj P is the corrected discharge power. opt The initial optimal discharge power predicted by the model is used as the basis for the above correction rule. When the battery loss risk exceeds the threshold, the discharge power is reduced by a step size of 0.3kW or 0.5kW to reduce battery discharge loss.

[0071] In this embodiment, the adaptive algorithm also includes a BP neural network algorithm for dynamic parameter adjustment of the elevator energy management system. This algorithm is used to fit the complex nonlinear relationship between the elevator operating conditions and the charging and discharging parameters of the energy storage device, so as to achieve accurate prediction and dynamic correction of the charging and discharging parameters, thereby balancing the stability of elevator power supply and power demand.

[0072] In this embodiment, the BP neural network algorithm adopts a multi-layer feedforward neural network structure. It iteratively adjusts the weight parameters of each layer of the network through the backpropagation algorithm to fit the complex relationship between "elevator operating condition parameters" and "energy storage device charging and discharging parameters". The network is trained using a large amount of historical operating data from the elevator energy management system, and the weight parameters are optimized to improve the prediction accuracy of charging and discharging parameters, adapting to the parameter adjustment requirements under normal operating conditions. It is suitable for operating conditions in the elevator energy management system that need to balance power supply stability and power demand, and where the charging and discharging parameter adjustment requirements are relatively smooth, specifically: normal no-load upward operating condition and normal full-load upward operating condition.

[0073] In this embodiment, the BP neural network algorithm aims to achieve the optimal elevator operation stability score and executes the following rules for learning and adjusting charging and discharging parameters: The structure of a BP neural network consists of an input layer, hidden layers, and an output layer. The input layer has 3 neurons, corresponding to the input parameters "current load value of the elevator, elevator running speed, and current state of charge (SOC value) of the energy storage system"; the hidden layer has 2 layers, each with 10 neurons, used to realize high-dimensional feature mapping of the input parameters; the output layer has 1 neuron, corresponding to the output parameter "optimal discharge power of the energy storage device".

[0074] Furthermore, historical operating data of the elevator energy management system is collected to form a training sample set containing "load value, operating speed, SOC value, discharge power, and operating stability score"; the operating stability score is used as the training objective, and the calculation formula for the operating stability score is: S=100 k×∣P i P0 | / P0; Where S is the operational stability score; P i P0 is the charging and discharging power predicted by the model; P0 is the actual power required by the elevator system; k is a compromise weighting factor for the "normal unloaded upward movement and normal fully loaded upward movement" conditions, with an example value of 5 (ensuring that the power deviation does not affect the operational stability while avoiding excessive deduction of scores for slight deviations); the training objective is to ensure that the operational stability score S corresponding to the model's predicted discharge power is not lower than the ideal threshold score S1 (S1 is 90 points for example). Furthermore, the weight parameters of each layer of the network are iteratively adjusted through the backpropagation algorithm, so that the operational stability score corresponding to the discharge power output by the model gradually approaches the target value, thus completing the network training.

[0075] The system collects the elevator's current load, operating speed, and SOC value of the energy storage system in real time. This data is then input into a trained BP neural network, which outputs the optimal discharge power of the energy storage device (the initial value for normal no-load upward operation is 5-8kW for example, and the initial value for normal full-load upward operation is also 5-8kW for example). If, during actual operation, the operational stability score S is lower than the first stability threshold score S0 (S0 is 85 points for example), the operational data is fed back to the BP neural network to fine-tune the network weight parameters, thereby correcting the discharge power parameters of the energy storage device and ensuring the elevator's operational stability.

[0076] In this embodiment, the adaptive algorithm also includes a model predictive control (MPC) algorithm for dynamic parameter adjustment of the elevator energy management system. It achieves advance adaptation of the charging and discharging parameters of the energy storage device through short-term load fluctuation prediction and rolling optimization, so as to ensure energy absorption efficiency and system operation stability.

[0077] In this embodiment, the Model Predictive Control (MPC) algorithm is based on the rolling optimization concept to establish a short-term elevator operating condition prediction model to predict the load change trend within a set time period. Combined with the operating constraints of the energy storage system, the optimal control strategy is solved to achieve advance matching and dynamic adjustment of charging and discharging parameters to the predicted operating conditions. Taking historical elevator operating condition data as input, the algorithm outputs the predicted load change trend and the corresponding predicted operating condition status, adapting to the energy management needs under load fluctuation scenarios. It is applicable to the operating conditions in the elevator energy management system that require advance prediction of load fluctuations and ensure energy absorption efficiency and system stability, specifically: normal no-load downward operating condition and normal full-load downward operating condition.

[0078] In this embodiment, the Model Predictive Control (MPC) algorithm aims to maximize energy absorption efficiency and executes the following rules for learning and adjusting charge and discharge parameters: The prediction model is based on the historical operating data of the elevator energy management system. An autoregressive integral moving average (ARIMA) prediction model is established to predict the elevator load change trend within the next second, thereby obtaining the corresponding predicted system operating conditions and providing a forward-looking basis for adjusting charging and discharging parameters.

[0079] Furthermore, the optimization target is set as "maximizing energy absorption efficiency" to improve the recovery and utilization rate of regenerative energy during elevator operation; The constraints include the maximum charge and discharge power of the energy storage system, the temperature rise of the energy storage battery ≤5℃, and the charging time ≤20s, to ensure that the adjustment of the charge and discharge parameters meets the safe operation threshold of the energy storage system.

[0080] Furthermore, the initial charging and discharging parameters are configured according to the elevator's operating conditions, wherein: the initial charging power under normal no-load downward operation is 2-3kW for example, and the initial charging time is 5s for example; the initial charging power under normal full-load downward operation is 5-8kW for example, and the initial charging time is 10s for example.

[0081] Furthermore, based on the load prediction results of the ARIMA model, the charging and discharging parameters are adjusted in advance: if the predicted load increases slightly, the charging power is increased by 0.5kW and the charging time is extended by 2s in advance to adapt to the increase in regenerative energy; if the predicted load is stable, the current charging and discharging parameters are maintained; if the predicted load decreases, the charging and discharging parameters are appropriately reduced to avoid ineffective energy consumption.

[0082] Furthermore, the specific formulas for forward adjustment of power and duration parameters are as follows:

[0083] Where: P adj Characterized by the adjusted charging power, T adjCharacterizes the adjusted charging time; P0 and T0 represent the initial charging power and initial charging time under the corresponding operating conditions, respectively; m t The m represents the load value at the current moment. t+1 Characterizes the predicted load value at the next time step.

[0084] The above are only some embodiments of the present invention and do not limit the patent scope of the present invention. All equivalent structural transformations made under the technical concept of the present invention using the contents of the present invention specification and drawings, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A parameter adjustment method for an elevator energy management system, characterized in that, include: Obtain the elevator's load data and direction of travel; Based on the load data, calculate the current load rate and the load change rate within a preset short window; Based on the current load rate, load change rate and running direction, the current operating condition scenario of the elevator is identified. The operating condition scenario includes normal operating conditions and extreme switching conditions caused by a sudden change in load rate in a short period of time. Based on the current operating conditions of the elevator, the corresponding charging and discharging parameters are invoked to control the energy storage device to recover the electrical energy generated during the elevator operation. The system acquires current and historical operating data of the elevator under the current operating conditions, and optimizes and updates the charging and discharging parameters corresponding to the current operating conditions of the elevator based on the current and historical operating data.

2. The parameter adjustment method for the elevator energy management system according to claim 1, characterized in that, The extreme switching conditions include the extreme switching state full-load downlink condition and the extreme switching state no-load uplink condition; The step of identifying the current operating condition scenario of the elevator based on the current load rate, load change rate, and running direction includes: If the elevator is descending and the current load rate is higher than the full load threshold, then proceed to the descending scenario judgment: If the load change rate is lower than the first change rate threshold, it is identified as a normal full-load downlink condition; If the load change rate is higher than or equal to the first change rate threshold, it is identified as an extreme switching state from no load to full load, and a full load downlink condition. If the elevator is moving upwards and the current load rate is below the no-load threshold, then proceed to the upward movement scenario for judgment: If the load change rate is lower than the second change rate threshold, it is identified as a normal no-load uplink condition; If the load change rate is higher than or equal to the second change rate threshold, it is identified as an extreme switching state from full load to no load, and the no load uplink condition is identified. If the current load rate is between the no-load threshold and the full-load threshold, it is identified as a normal transitional operating condition. If the elevator descends and the current load rate is lower than or equal to the no-load threshold, it is identified as a normal no-load descending condition. If the elevator is moving upwards and the current load rate is higher than or equal to the full load threshold, it is identified as a normal full load upward operation.

3. The method according to claim 2, characterized in that, The charging and discharging parameters include charging power and discharging power. The step of calling the corresponding charging and discharging parameters based on the current operating conditions of the elevator to recover energy generated during elevator operation includes: When the condition is identified as a normal full-load downlink operation, the energy storage device is controlled to recover energy at the first charging power. When the extreme switching state full-load downlink condition is identified, the energy storage device is controlled to recover energy with a second charging power, which is higher than the first charging power. When the condition is identified as a normal no-load uplink condition, the energy storage device is controlled to supply power at the first discharge power. When an extreme switching state with no-load uplink operation is identified, the energy storage device is controlled to supply power at a second discharge power, which is lower than the first discharge power.

4. The method according to claim 2, characterized in that, The method further includes: When identified as a normal transitional operating condition, the energy storage device is controlled to recover energy at a third charging power, which is lower than the first charging power; and / or, the energy storage device is controlled to supply power at a third discharging power, which is lower than the first discharging power. When the condition is identified as a normal no-load downlink operating condition, the energy storage device is controlled to recover energy at a fourth charging power, which is lower than the first charging power. When the operation is identified as a normal full-load uplink condition, the energy storage device is controlled to supply power at a fourth discharge power, which is higher than the first discharge power.

5. The method according to claim 2, characterized in that, The charging and discharging parameters also include the charging duration, wherein: The charging time set for the extreme switching state full-load downlink condition is greater than the charging time set for the normal full-load downlink condition; The charging time set for normal full-load downlink operation is greater than the charging time set for normal transition operation during downlink; The charging time set during downlink in normal transition conditions is greater than or equal to the charging time set during normal no-load downlink conditions.

6. The method according to claim 2, characterized in that, The charge / discharge parameters also include a state-of-charge protection threshold, wherein: The state of charge protection threshold set for extreme switching no-load uplink conditions is higher than the state of charge protection threshold set for normal no-load uplink conditions. The state-of-charge protection threshold set for the normal no-load uplink condition is higher than the state-of-charge protection threshold set for the normal transition condition during uplink. The state-of-charge protection threshold set during the normal transition operation is higher than or equal to the state-of-charge protection threshold set during the normal full-load uplink operation.

7. The method according to claim 1, characterized in that, The optimization and updating of charging and discharging parameters corresponding to the current operating conditions of the elevator based on current and historical operating data includes: The current operating data and historical operating data are input into the machine learning model to optimize the charging and discharging parameters corresponding to the current operating conditions of the elevator, and the optimized charging and discharging parameters are output. Update the optimized charging and discharging parameters to match the charging and discharging parameters corresponding to the current operating conditions of the elevator.

8. The method according to claim 1, characterized in that, The current operating data includes at least one of the following: load data, running direction, running speed, state of charge of the energy storage device, temperature of the energy storage device, voltage and current of the energy storage device, real-time charging and discharging power of the energy storage device, DC bus voltage of the elevator inverter, and power grid interaction power.

9. A parameter adjustment device for an elevator energy management system, characterized in that, include: The system includes a memory, a processor, and a parameter adjustment program stored in the memory and executable on the processor, the parameter adjustment program being configured to implement the steps of the parameter adjustment method of the elevator energy management system as described in any one of claims 1 to 8.

10. An elevator energy management system, characterized in that, The elevator energy management system includes an energy storage device, an elevator frequency converter, a data acquisition module, and a central control module. The energy storage device includes a battery unit and a supercapacitor unit, which are connected in parallel to the DC bus of the elevator frequency converter. The data acquisition module is used to acquire the elevator's load data, running direction, and current running data; The central control module is communicatively connected to the data acquisition module, the energy storage device, and the elevator frequency converter.

Images