Hybrid energy storage energy saving method and system for elevators

Abstract

The application relates to the technical field of energy recovery, in particular to a hybrid energy storage energy-saving method and system for an elevator, which comprises the following steps: collecting operation state parameters of the elevator in real time in response to a descending or braking instruction of the elevator; collecting control data of an elevator controller, obtaining a braking position of the elevator, and obtaining a braking distance according to the current position and the braking position of the elevator; inputting the obtained operation state parameters and the braking distance into a pre-constructed regenerative power peak-time prediction model to predict a regenerative power peak value and a duration; and the like. The hybrid energy storage energy-saving method for the elevator provided by the application can predict the peak value and the duration of the regenerative power of the elevator, release part of the capacity of the super capacitor in advance, make the super capacitor preferentially absorb energy in the power surge stage, and fully utilize the fast response characteristics of the super capacitor.

Description

Technical Field

[0001] This invention relates to the field of energy recovery technology, specifically to a hybrid energy storage and energy-saving method and system for elevators. Background Technology

[0002] When the elevator is descending (especially under heavy load) or braking and decelerating, the motor becomes a generator to produce regenerative electrical energy; when the elevator produces regenerative electrical energy, the energy storage system (lithium battery + supercapacitor) absorbs this energy through the DC bus of the frequency converter.

[0003] However, existing technologies typically inject regenerated electrical energy into the DC bus for absorption by energy storage units. However, the lack of a dynamic power distribution strategy between lithium batteries and supercapacitors results in supercapacitors failing to fully absorb instantaneous high-power energy, and lithium batteries being subjected to excessive pulse charge and discharge shocks, reducing their lifespan. Furthermore, while supercapacitors are suitable for absorbing high-power transient energy, their capacity is relatively small, and they are often idle in existing systems due to a lack of intelligent scheduling, failing to leverage their rapid charge and discharge advantages. Summary of the Invention

[0004] To address the aforementioned problems, this invention provides a hybrid energy storage and energy-saving method and system for elevators.

[0005] This invention adopts the following technical solution: a hybrid energy storage and energy-saving method for elevators, comprising:

[0006] In response to the descent or braking command of the elevator, the operating status parameters of the elevator are collected in real time;

[0007] Collect control data from the elevator controller, obtain the braking position of the elevator, and obtain the braking distance based on the current position and braking position of the elevator;

[0008] The acquired operating status parameters and braking distance are input into a pre-built peak-time regenerative power prediction model to predict the peak regenerative power and its duration.

[0009] Based on the predicted peak and duration of regeneration power and the current available margin of the supercapacitor, a supercapacitor regulation command is generated, and the voltage of the supercapacitor is regulated to the target voltage based on the regulation command.

[0010] In response to the descent or braking commands of the elevator, voltage and current sensors installed on the DC bus side of the elevator system are used to collect the DC bus voltage in real time. With DC bus Calculate the bus power ;

[0011] Calculate the fluctuation range of bus power between two consecutive sampling periods, and set the allocation weights for supercapacitors and lithium batteries based on the fluctuation range;

[0012] The regenerative power on the bus side is allocated according to the allocation weights of the supercapacitors and the lithium batteries to obtain the power allocated to the supercapacitors and the power allocated to the lithium batteries.

[0013] As a further description of the above technical solution: the operating status parameters of the elevator include the car load, which is obtained by a weighing sensor; the operating speed and acceleration are obtained by a motor encoder or speed sensor; the operating mode is determined by the controller status signal; the operating mode includes going up, going down, and braking.

[0014] As a further description of the above technical solution: the training method of the regenerative power peak-time prediction model includes:

[0015] H sets of training data are collected in advance, where H is a positive integer greater than 1. The H sets of training data include the elevator's operating status parameters and braking distance, as well as the corresponding peak regenerative power and duration.

[0016] The peak regenerative power and duration are constructed into a set and assigned a number to obtain updated training data, which consists of the elevator's operating status parameters and braking distance, as well as the corresponding number.

[0017] The operating status parameters and braking distance of the elevator are converted into a set of corresponding feature vectors. Each set of feature vectors is used as the input of the regenerative power peak-time prediction model. The regenerative power peak-time prediction model outputs the number corresponding to each set of operating status parameters and braking distance, uses the actual number corresponding to each set of operating status parameters and braking distance as the prediction target, and minimizes the loss function value of the regenerative power peak-time prediction model as the training objective. Training stops when the loss function value of the regenerative power peak-time prediction model is less than or equal to the preset target loss value. Based on the number predicted by the regenerative power peak-time prediction model, the regenerative power peak and duration in the corresponding set are obtained.

[0018] As a further description of the above technical solution: the method for generating the supercapacitor regulation command includes:

[0019] The required available supercapacitor capacity to be reserved is determined based on the obtained peak regeneration power and duration. ;

[0020] Calculate the available margin of the current supercapacitor. ;

[0021] when ≥ No adjustment instructions are generated;

[0022] when < Generate supercapacitor regulation commands.

[0023] As a further description of the above technical solution: the method for setting the supercapacitor allocation weight and the lithium battery allocation weight includes:

[0024] ;

[0025] ;

[0026] In the formula, Assign weights to supercapacitors. Assign weights to lithium batteries. As a balance factor, This refers to the fluctuation range.

[0027] As a further description of the above technical solution: the method also includes:

[0028] Obtain the charge value and health coefficient of the lithium battery, perform a weighted summation based on the charge value and health coefficient to generate the lithium battery state coefficient, and generate a lithium battery weight adjustment instruction based on a preset state coefficient threshold.

[0029] The allocation weights of lithium batteries are adjusted based on the lithium battery allocation weight adjustment instruction, and the allocation weights of supercapacitors are adjusted based on the adjusted lithium battery allocation weights.

[0030] As a further description of the above technical solution: the method for obtaining the charge value of the lithium battery includes: measuring the battery current in real time, integrating the battery current to obtain the charge change, and calculating the charge value by combining the initial SOC value and rated capacity of the battery.

[0031] As a further description of the above technical solution: the method for obtaining the health coefficient of the lithium battery includes:

[0032] Obtain battery health characteristic parameters, including AC impedance, battery temperature, and charge / discharge cycles;

[0033] A mapping relationship between battery health feature parameters and battery health coefficient is pre-constructed, and the battery health coefficient is obtained based on the acquired battery health feature parameters;

[0034] The method for constructing the mapping relationship between the battery health characteristic parameters and the battery health coefficient includes:

[0035] Experiments were conducted on the target type of lithium battery to cover the operating status under different capacity decay stages, forming a training dataset. The training dataset includes the feature parameter vector of each experimental sample and the corresponding actual battery health coefficient measurement value.

[0036] The model is trained by using the collected feature parameter vectors and the actual battery health coefficient measurements to establish the mapping relationship between battery health feature parameters and lithium battery health coefficient. The mapping model adopts a multiple regression model, and the model training process aims to minimize the prediction error. The loss function is the mean squared error.

[0037] As a further description of the above technical solution: the method for generating lithium battery allocation weight adjustment instructions based on a preset state coefficient threshold includes:

[0038] The lithium battery allocation weight adjustment instruction includes a first adjustment instruction and a second adjustment instruction, and the degree to which the first adjustment instruction and the second adjustment instruction reduce the lithium battery allocation weight increases sequentially.

[0039] Preset state coefficient thresholds BSSmax and BSSmin, where BSSmax > BSSmin;

[0040] It should be noted that the state coefficient threshold was determined by those skilled in the art based on a large number of experiments.

[0041] when If BSSmax is greater than or equal to BSSmax, no lithium battery allocation weight adjustment instruction will be generated; The state coefficient of a lithium battery;

[0042] When BSSmax > When BSSmin is reached, the first adjustment command is generated;

[0043] when When ≤BSSmin, generate a second adjustment instruction.

[0044] A hybrid energy storage and energy-saving system for elevators, used to implement the aforementioned hybrid energy storage and energy-saving method for elevators, the system comprising:

[0045] The first parameter acquisition module, in response to the descent or braking command of the elevator, acquires the operating status parameters of the elevator in real time.

[0046] The distance acquisition module collects control data from the elevator controller, obtains the braking position of the elevator, and calculates the braking distance based on the current position and braking position of the elevator.

[0047] The power-time prediction module inputs the acquired operating status parameters and braking distance into a pre-built regenerative power peak-time prediction model to predict the regenerative power peak and duration.

[0048] The capacitor adjustment module generates a supercapacitor adjustment command based on the predicted peak and duration of regeneration power and the available margin of the current supercapacitor, and adjusts the voltage of the supercapacitor to the target voltage based on the adjustment command.

[0049] The second parameter acquisition module, in response to the descent or braking command of the elevator, acquires the DC bus voltage in real time through voltage and current sensors installed on the DC bus side of the elevator system. With current Calculate the bus power ;

[0050] The weight allocation module calculates the fluctuation range of the bus power between two consecutive sampling periods, and sets the allocation weights for supercapacitors and lithium batteries based on the fluctuation range.

[0051] The power adjustment module allocates the regenerative power on the bus side according to the allocation weights of the supercapacitors and lithium batteries, thus obtaining the power allocated to the supercapacitors and the power allocated to the lithium batteries.

[0052] Beneficial effects:

[0053] The hybrid energy storage and energy-saving method for elevators provided by this invention predicts the peak value and duration of the elevator's regenerative power and releases part of the supercapacitor's capacity in advance, enabling it to preferentially absorb energy during power surges, thereby fully utilizing the supercapacitor's rapid response characteristics.

[0054] Secondly, this invention introduces a dynamic allocation strategy, setting weighting factors based on real-time power fluctuations to achieve coordinated energy distribution between the supercapacitor and the lithium battery. This dynamic adjustment avoids situations where the supercapacitor fails to fully absorb transient energy due to idleness, while simultaneously reducing the stress on the lithium battery caused by high-frequency pulse charging and discharging, effectively extending battery life. Attached Figure Description

[0055] The present invention will be further explained below with reference to the accompanying drawings and embodiments:

[0056] Figure 1 A flowchart of a hybrid energy storage and energy-saving method for elevators provided in Embodiment 1 of the present invention;

[0057] Figure 2 A flowchart of a hybrid energy storage and energy-saving method for elevators provided in Embodiment 2 of the present invention;

[0058] Figure 3 A flowchart of a method for obtaining the health coefficient of a lithium battery according to Embodiment 2 of the present invention;

[0059] Figure 4 This is a module connection diagram of the hybrid energy storage and energy-saving system for elevators provided in Embodiment 3 of the present invention. Detailed Implementation

[0060] To make the technical means, creative features, objectives, and effects of this invention readily understandable, the invention is further described below with reference to specific illustrations. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0061] Example 1

[0062] Please see Figure 1 This invention provides a technical solution: a hybrid energy storage and energy-saving method for elevators, comprising:

[0063] In response to the descent or braking command of the elevator, the operating status parameters of the elevator are collected in real time;

[0064] The elevator's operating status parameters include the car load, obtained by a weighing sensor; the operating speed and acceleration, obtained by a motor encoder or speed sensor; and the operating mode, determined by the controller's status signal. The operating modes include ascending, descending, and braking.

[0065] Collect control data from the elevator controller, obtain the braking position of the elevator, and obtain the braking distance based on the current position and braking position of the elevator;

[0066] The acquired operating status parameters and braking distance are input into a pre-built peak-time regenerative power prediction model to predict the peak regenerative power and its duration.

[0067] The training method for the regenerative power peak-time prediction model includes:

[0068] H sets of training data are collected in advance, where H is a positive integer greater than 1. The H sets of training data include elevator operating status parameters and braking distance, as well as the corresponding peak regenerative power and duration.

[0069] The peak regenerative power and duration are constructed into a set and assigned a number to obtain updated training data, which consists of elevator operating status parameters and braking distance, as well as the corresponding number.

[0070] The elevator's operating status parameters and braking distance are converted into a set of corresponding feature vectors. Each set of feature vectors is used as the input to the regenerative power peak-time prediction model. The regenerative power peak-time prediction model outputs the number corresponding to each set of operating status parameters and braking distance, uses the actual number corresponding to each set of operating status parameters and braking distance as the prediction target, and minimizes the loss function value of the regenerative power peak-time prediction model as the training objective. Training stops when the loss function value of the regenerative power peak-time prediction model is less than or equal to the preset target loss value. Based on the number predicted by the regenerative power peak-time prediction model, the regenerative power peak and duration within the corresponding set are obtained.

[0071] The peak-time prediction model for regenerative power can be one of the following: support vector machine regression, random forest regression, or neural network regression.

[0072] Based on the predicted peak and duration of regeneration power and the current available margin of the supercapacitor, a supercapacitor regulation command is generated, and the voltage of the supercapacitor is regulated to the target voltage based on the regulation command.

[0073] The method for generating the supercapacitor regulation command includes:

[0074] The required available supercapacitor capacity is calculated based on the obtained peak regeneration power and duration. ;

[0075] The formula for calculating the usable capacity of the supercapacitor is as follows: In the formula, The available capacity (J) of the supercapacitor that needs to be reserved. To predict peak regeneration power (W). To predict the peak duration (s). As a safety margin factor (between 0 and 1), optionally, The value is 0.8;

[0076] Calculate the available margin of the current supercapacitor. ;

[0077] The formula for calculating the available margin is: ;

[0078] In the formula, This represents the energy margin (J) that a current supercapacitor can immediately absorb. This is the highest safe voltage allowed for a supercapacitor. The current voltage of the supercapacitor. This is the equivalent capacitance of a supercapacitor. This represents the energy stored in a supercapacitor at a given voltage.

[0079] when ≥ No adjustment instructions are generated;

[0080] when < Generate supercapacitor adjustment commands;

[0081] The method for adjusting the voltage of a supercapacitor to the target voltage based on adjustment commands is as follows:

[0082] The difference between the available capacity of the supercapacitor that needs to be reserved and the energy margin that the current supercapacitor can immediately absorb is obtained to get the energy to be released. Based on the obtained energy to be released, the target voltage is calculated, and the supercapacitor is adjusted to the target voltage.

[0083] The method for calculating the target voltage is as follows: the target voltage is calculated based on the supercapacitor energy function, and the expression for the supercapacitor energy function is: ;

[0084] It should be noted that when releasing energy from the supercapacitor, the released energy is preferentially transferred to the lithium battery in a controlled manner to reduce energy waste; if the battery is currently unacceptable or for protection purposes, the energy is transferred to the braking resistor or energy dissipation device.

[0085] In this embodiment, short-term predictive scheduling of regenerative power can be achieved during elevator operation. By introducing a peak-time regenerative power prediction model, the system can predict the potential peak regenerative power and its duration based on operating status parameters such as car load, running speed, acceleration, and braking distance before the elevator enters a heavy-load descent or braking phase, thereby achieving advance control.

[0086] By calculating the required reserve capacity of the supercapacitor based on the prediction results and comparing it with the current available margin of the supercapacitor, the system can dynamically adjust the supercapacitor voltage to keep it within a reasonable range for energy absorption. This effectively avoids situations where the supercapacitor cannot absorb energy due to its voltage approaching its upper limit, ensuring that the supercapacitor can play a priority role during transient power surges, and improving the sufficiency and timeliness of energy recovery.

[0087] In response to the descent or braking commands of the elevator, voltage and current sensors installed on the DC bus side of the elevator system are used to collect the DC bus voltage in real time. With current Calculate the bus power ;

[0088] The formula for calculating the bus power is: In the formula, For bus power, This is the DC bus voltage. This refers to the DC bus current.

[0089] Calculate the fluctuation range of bus power between two consecutive sampling periods;

[0090] The formula for calculating the fluctuation range is: In the formula, For fluctuation range, The current bus power. The bus power at the previous sampling time. The sampling period;

[0091] Based on the fluctuation range, the allocation weights for supercapacitors and lithium batteries are set;

[0092] The methods for setting the supercapacitor allocation weights and lithium battery allocation weights include:

[0093] ;

[0094] ;

[0095] In the formula, Assign weights to supercapacitors. Assign weights to lithium batteries. This is a balancing factor used to prevent the supercapacitor weight from becoming zero during small fluctuations. This refers to the fluctuation range.

[0096] It should be noted that the above weighting settings ensure that when the fluctuation range is large, When the value is close to 1, supercapacitors preferentially absorb power; when the fluctuation amplitude is small and the power is stable, Lithium batteries are dominant and primarily provide energy.

[0097] The regenerative power on the bus side is allocated according to the supercapacitor allocation weight and the lithium battery allocation weight to obtain the power allocated to the supercapacitor and the power allocated to the lithium battery.

[0098] The formulas for calculating the power allocated to the supercapacitor and the power allocated to the lithium battery are as follows:

[0099] ;

[0100] ;

[0101] It should be noted that, For bus power, Assign weights to supercapacitors. Assign weights to lithium batteries. The power allocated to the supercapacitor, This refers to the power allocated to the lithium battery.

[0102] In this embodiment, by predicting the peak value and duration of the elevator's regenerative power, a portion of the supercapacitor's capacity is released in advance, enabling it to preferentially absorb energy during the power surge phase, thereby fully utilizing the supercapacitor's rapid response characteristics.

[0103] Secondly, this invention introduces a dynamic allocation strategy, setting weighting factors based on real-time power fluctuations to achieve coordinated energy distribution between the supercapacitor and the lithium battery. This dynamic adjustment avoids situations where the supercapacitor fails to fully absorb transient energy due to idleness, while simultaneously reducing the stress on the lithium battery caused by high-frequency pulse charging and discharging, effectively extending battery life.

[0104] Example 2,

[0105] Please see Figures 2-3 Based on Embodiment 1, this invention provides another technical solution:

[0106] Obtain the charge value and health coefficient of the lithium battery, perform a weighted summation based on the charge value and health coefficient to generate the lithium battery state coefficient, and generate a lithium battery weight adjustment instruction based on a preset state coefficient threshold.

[0107] The allocation weights of lithium batteries are adjusted based on the lithium battery allocation weight adjustment instruction, and the allocation weights of supercapacitors are adjusted based on the adjusted lithium battery allocation weights.

[0108] The method for obtaining the charge value of the lithium battery includes:

[0109] The change in charge is obtained by measuring the battery current in real time and integrating the battery current. The charge value is then calculated by combining the initial SOC value and the rated capacity of the battery.

[0110] The formula for calculating the charge value is: In the formula, The current state of charge, This represents the battery discharge / charge current; a positive value indicates discharge, and a negative value indicates charging. The variable is the integral variable, corresponding to the time point of current acquisition. This is the initial SOC value of the battery. This refers to the battery's rated capacity.

[0111] The method for obtaining the health coefficient of the lithium battery includes:

[0112] Obtain battery health characteristic parameters, including AC impedance, battery temperature, and charge / discharge cycles;

[0113] A mapping relationship between battery health feature parameters and battery health coefficient is pre-constructed, and the battery health coefficient is obtained based on the acquired battery health feature parameters;

[0114] The method for constructing the mapping relationship between the battery health characteristic parameters and the battery health coefficient includes:

[0115] Experiments were conducted on the target type of lithium battery to cover the operating status under different capacity decay stages, forming a training dataset. The training dataset includes the feature parameter vector of each experimental sample and the corresponding actual battery health coefficient measurement value.

[0116] A mapping model is trained using the collected feature parameter vectors and actual battery health coefficient measurements to establish a mapping relationship between battery health feature parameters and the health coefficient of lithium batteries. The mapping model adopts a multiple regression model, and the model training process aims to minimize the prediction error. The loss function is the mean squared error. The expression is:

[0117] ;

[0118] In the formula, The number of training samples. These are the model's predicted values. These are actual measured values.

[0119] Training stops when the loss function value of the mapping model is less than or equal to the preset target loss value.

[0120] Optionally, the expression for generating the lithium battery state coefficient by weighted summation based on the lithium battery's charge value and health coefficient is as follows: ;

[0121] In the formula, The state coefficient of a lithium battery. This is the normalized value of the lithium battery's charge. This is a normalized value for the battery health coefficient. and Weighting coefficients + =1;

[0122] To eliminate the influence of differences in dimensions and numerical ranges on the calculation of state coefficients, the state charge (SOC) and health coefficient (SOH) of lithium batteries are normalized to map them to the [0,1] interval. The minimum and maximum allowable values ​​of the corresponding parameters are used as normalization boundaries, and the currently measured actual values ​​are converted into dimensionless values ​​in the 0-1 interval through linear mapping.

[0123] It should be noted that the higher the state coefficient of a lithium battery, the better the battery's condition. The battery health coefficient represents the battery's health level or the percentage of its remaining lifespan. The higher the battery health coefficient, the longer the battery life. The charge value of a lithium battery represents the proportion of its remaining capacity to its rated capacity. The higher the charge value of a lithium battery, the greater its usable capacity.

[0124] The method for generating lithium battery allocation weight adjustment instructions based on a preset state coefficient threshold includes:

[0125] The lithium battery allocation weight adjustment instruction includes a first adjustment instruction and a second adjustment instruction, and the degree to which the first adjustment instruction and the second adjustment instruction reduce the lithium battery allocation weight increases sequentially.

[0126] Preset state coefficient thresholds BSSmax and BSSmin, where BSSmax > BSSmin;

[0127] It should be noted that the state coefficient threshold was determined by those skilled in the art based on a large number of experiments.

[0128] when If BSSmax is greater than or equal to BSSmax, no lithium battery allocation weight adjustment instruction will be generated;

[0129] When BSSmax > When BSSmin is reached, the first adjustment command is generated;

[0130] when When ≤BSSmin, generate a second adjustment instruction.

[0131] In this embodiment, the energy allocation process takes into account the charge value and health coefficient of the lithium battery, updates the allocation weight, and effectively delays the degradation of lithium battery life by dynamically adjusting the energy allocation ratio, thereby improving the stability and reliability of the system under long-term operation.

[0132] Example 3

[0133] Please see Figure 4 This invention provides a technical solution: a hybrid energy storage and energy-saving system for elevators, which is used to implement the hybrid energy storage and energy-saving method for elevators, the system comprising:

[0134] The first parameter acquisition module, in response to the descent or braking command of the elevator, acquires the operating status parameters of the elevator in real time.

[0135] The distance acquisition module collects control data from the elevator controller, obtains the braking position of the elevator, and calculates the braking distance based on the current position and braking position of the elevator.

[0136] The power-time prediction module inputs the acquired operating status parameters and braking distance into a pre-built regenerative power peak-time prediction model to predict the regenerative power peak and duration.

[0137] The capacitor adjustment module generates a supercapacitor adjustment command based on the predicted peak and duration of regeneration power and the available margin of the current supercapacitor, and adjusts the voltage of the supercapacitor to the target voltage based on the adjustment command.

[0138] The second parameter acquisition module, in response to the descent or braking command of the elevator, acquires the DC bus voltage in real time through voltage and current sensors installed on the DC bus side of the elevator system. With current Calculate the bus power ;

[0139] The weight allocation module calculates the fluctuation range of the bus power between two consecutive sampling periods, and sets the allocation weights for supercapacitors and lithium batteries based on the fluctuation range.

[0140] The power adjustment module allocates the regenerative power on the bus side according to the allocation weights of the supercapacitors and lithium batteries, thus obtaining the power allocated to the supercapacitors and the power allocated to the lithium batteries.

[0141] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A hybrid energy storage and energy-saving method for elevators, characterized in that, include: In response to the descent or braking command of the elevator, the operating status parameters of the elevator are collected in real time; Collect control data from the elevator controller, obtain the braking position of the elevator, and obtain the braking distance based on the current position and braking position of the elevator; The acquired operating status parameters and braking distance are input into a pre-built peak-time regenerative power prediction model to predict the peak regenerative power and its duration. Based on the predicted peak and duration of regeneration power and the current available margin of the supercapacitor, a supercapacitor regulation command is generated, and the voltage of the supercapacitor is regulated to the target voltage based on the regulation command. In response to the descent or braking command of the elevator, voltage and current sensors are installed on the DC bus side of the elevator system to collect the DC bus voltage and DC bus current in real time and calculate the bus power. Calculate the fluctuation range of bus power between two consecutive sampling periods, and set the allocation weights for supercapacitors and lithium batteries based on the fluctuation range; The regenerative power on the bus side is allocated according to the allocation weights of the supercapacitors and the lithium batteries to obtain the power allocated to the supercapacitors and the power allocated to the lithium batteries.

2. The hybrid energy storage and energy-saving method for elevators according to claim 1, characterized in that, The operating status parameters of the elevator include the car load, which is obtained by a weighing sensor; the operating speed and acceleration are obtained by a motor encoder or speed sensor; the operating mode is determined by the controller status signal; the operating mode includes ascending, descending and braking.

3. The hybrid energy storage and energy-saving method for elevators according to claim 1, characterized in that, The training method for the regenerative power peak-time prediction model includes: H sets of training data are collected in advance, where H is a positive integer greater than 1. The H sets of training data include the elevator's operating status parameters and braking distance, as well as the corresponding peak regenerative power and duration. The peak regenerative power and duration are constructed into a set and assigned a number to obtain updated training data, which consists of the elevator's operating status parameters and braking distance, as well as the corresponding number. The elevator's operating status parameters and braking distance are converted into a set of corresponding feature vectors. Each set of feature vectors is used as input to the regenerative power peak-time prediction model. The regenerative power peak-time prediction model outputs the number corresponding to each set of operating status parameters and braking distance, uses the actual number corresponding to each set of operating status parameters and braking distance as the prediction target, and minimizes the loss function value of the regenerative power peak-time prediction model as the training objective. Training stops when the loss function value of the regenerative power peak-time prediction model is less than or equal to the preset target loss value. Based on the number predicted by the regenerative power peak-time prediction model, the regenerative power peak and duration within the corresponding set are obtained.

4. The hybrid energy storage and energy-saving method for elevators according to claim 1, characterized in that, The method for generating the supercapacitor regulation command includes: The required available supercapacitor capacity is calculated based on the obtained peak regeneration power and duration. ; Calculate the available margin of the current supercapacitor. ; when ≥ No adjustment instructions are generated; when < Generate supercapacitor regulation commands.

5. The hybrid energy storage and energy-saving method for elevators according to claim 1, characterized in that, The methods for setting the supercapacitor allocation weights and lithium battery allocation weights include: ； ； In the formula, Assign weights to supercapacitors. Assign weights to lithium batteries. As a balance factor, This refers to the fluctuation range.

6. The hybrid energy storage and energy-saving method for elevators according to claim 1, characterized in that, Also includes: Obtain the charge value and health coefficient of the lithium battery, perform a weighted summation based on the charge value and health coefficient to generate the lithium battery state coefficient, and generate a lithium battery weight adjustment instruction based on a preset state coefficient threshold. The allocation weights of lithium batteries are adjusted based on the lithium battery allocation weight adjustment instruction, and the allocation weights of supercapacitors are adjusted based on the adjusted lithium battery allocation weights.

7. The hybrid energy storage and energy-saving method for elevators according to claim 6, characterized in that, The method for obtaining the charge value of the lithium battery includes: measuring the battery current in real time, integrating the battery current to obtain the charge change, and calculating the charge value by combining the initial SOC value and rated capacity of the battery.

8. The hybrid energy storage and energy-saving method for elevators according to claim 6, characterized in that, The method for obtaining the health coefficient of the lithium battery includes: Obtain battery health characteristic parameters, including AC impedance, battery temperature, and charge / discharge cycles; A mapping relationship between battery health feature parameters and battery health coefficient is pre-constructed, and the battery health coefficient is obtained based on the acquired battery health feature parameters; The method for constructing the mapping relationship between the battery health characteristic parameters and the battery health coefficient includes: Experiments were conducted on the target type of lithium battery to cover the operating status under different capacity decay stages, forming a training dataset. The training dataset includes the feature parameter vector of each experimental sample and the corresponding actual battery health coefficient measurement value. The model is trained by using the collected feature parameter vectors and the actual battery health coefficient measurements to establish the mapping relationship between battery health feature parameters and lithium battery health coefficient. The mapping model adopts a multiple regression model, and the model training process aims to minimize the prediction error. The loss function is the mean squared error.

9. The hybrid energy storage and energy-saving method for elevators according to claim 6, characterized in that, The method for generating lithium battery allocation weight adjustment instructions based on a preset state coefficient threshold includes: The lithium battery allocation weight adjustment instruction includes a first adjustment instruction and a second adjustment instruction, and the degree to which the first adjustment instruction and the second adjustment instruction reduce the lithium battery allocation weight increases sequentially. Preset state coefficient thresholds BSSmax and BSSmin, where BSSmax > BSSmin; when If BSSmax is greater than or equal to BSSmax, no lithium battery allocation weight adjustment instruction will be generated; The state coefficient of a lithium battery; When BSSmax > When BSSmin is reached, the first adjustment command is generated; when When ≤BSSmin, generate a second adjustment instruction.

10. A hybrid energy storage and energy-saving system for elevators, used to implement the hybrid energy storage and energy-saving method for elevators according to any one of claims 1-9, characterized in that, The system includes: The first parameter acquisition module, in response to the descent or braking command of the elevator, acquires the operating status parameters of the elevator in real time. The distance acquisition module collects control data from the elevator controller, obtains the braking position of the elevator, and calculates the braking distance based on the current position and braking position of the elevator. The power-time prediction module inputs the acquired operating status parameters and braking distance into a pre-built regenerative power peak-time prediction model to predict the regenerative power peak and duration. The capacitor adjustment module generates a supercapacitor adjustment command based on the predicted peak and duration of regeneration power and the available margin of the current supercapacitor, and adjusts the voltage of the supercapacitor to the target voltage based on the adjustment command. The second parameter acquisition module, in response to the descent or braking command of the elevator, acquires the DC bus voltage in real time through voltage and current sensors installed on the DC bus side of the elevator system. With current Calculate the bus power ; The weight allocation module calculates the fluctuation range of the bus power between two consecutive sampling periods, and sets the allocation weights for supercapacitors and lithium batteries based on the fluctuation range. The power adjustment module allocates the regenerative power on the bus side according to the allocation weights of the supercapacitors and lithium batteries, thus obtaining the power allocated to the supercapacitors and the power allocated to the lithium batteries.

Images