Self-adaptive shunting method for combined layer strategy elevator powered by emergency power supply

Abstract

The invention belongs to the technical field of self-adaptive dynamic planning, and provides a self-adaptive shunting method for a merging layer strategy elevator powered by an emergency power supply, which specifically comprises the following steps of: firstly, identifying a power supply mode of an energy storage power supply by monitoring voltage fluctuation of the elevator; the continuous high-demand areas are collected through target floor data to form a combined floor, then whether load overloading occurs in the service execution range of the elevator is judged according to the elevator load, and finally the service range is redistributed for the elevator group when load overloading occurs. The merging quality of the merging layer is improved through the method of adaptively establishing the merging layer according to the time period, so that the no-load rate can be effectively reduced while the stopping rate is reduced in the application process of the merging layer strategy. And service range adjustment is carried out on the problem of excessive load of individual elevators in the application process of the merging layer strategy, and power consumption balance between the elevators enables cooperation between energy storage power supplies connected in parallel to be more stable.

Description

Technical Field

[0001] The invention belongs to the technical field of adaptive dynamic programming, and in particular relates to an adaptive flow diversion method for an elevator with a merged layer strategy powered by an emergency power supply. Background Art

[0002] In the normal operation of the elevator system, the elevator usually relies on a direct connection to the power grid to obtain a stable power supply. However, in the face of power outages, system failures or other emergencies, the elevator must switch to energy storage batteries for power supply, which puts higher requirements on the energy scheduling and optimization of the diversion system. In the existing diversion strategy, by setting fixed floors as the boundaries of the service area, the elevator divides the functions between different floors, thereby effectively simplifying the complexity of scheduling decisions. On this basis, the merging layer strategy can greatly improve the durability of the elevator power supply. The merging layer strategy refers to merging several consecutive target floors into one target floor, and transporting passengers going to these floors to one of the merged floors, which not only optimizes the scheduling efficiency, but also enhances the continuity and stability of power supply when there is a lack of external power input. However, on the one hand, the existing merging layer strategy is usually implemented in a fixed proportion, for example, every two floors are merged into one merged layer. The disadvantage of this traditional merging method is that the parking rate is still relatively high, so the power consumption is limited, and this solidified mode cannot be adapted according to the actual elevator demand, which makes some merged layers have serious problems of empty transportation. On the other hand, there is a load imbalance problem between elevators in the merged layer strategy. Although merged layer transportation effectively reduces the stop rate, the weight density of each elevator transported per unit time increases accordingly. Especially when the elevator service range is in the floor interval with high transportation demand or during peak hours, the load pressure is huge. This long-term high-load operation requires the battery to provide higher power output, causing the battery temperature to rise and accelerate its degradation and attenuation. As the discharge depth of the battery gradually increases, the potential difference of the battery gradually decreases, resulting in the elevator being unable to maintain stable power output under high load, thereby affecting the operating stability of the elevator and the stability of the microgrid. Summary of the invention

[0003] The purpose of the present invention is to propose an adaptive diversion method for elevators with a merged floor strategy powered by an emergency power supply to solve one or more technical problems existing in the prior art and at least provide a beneficial choice or create conditions.

[0004] In order to achieve the above object, according to one aspect of the present invention, a method for adaptive traffic diversion of a combined floor strategy elevator powered by an emergency power supply is provided, the method comprising the following steps: The energy storage power supply mode is identified by monitoring the voltage fluctuations of the elevator. After the energy storage power supply mode is triggered, each time the elevator stops at the first floor, the continuous high-demand areas are dynamically aggregated to form a merged layer through the destination floor data. The elevator load is used to determine whether the service scope is overloaded. If overload occurs, the service scope is reallocated for the elevator group.

[0005] Furthermore, a method for identifying the energy storage power supply mode by monitoring the elevator voltage fluctuation is: using a voltage sensor to collect the elevator voltage data in real time, the voltage data measurement frequency range is 1 to 50 milliseconds / time; setting the dynamic window to 2 to 10 minutes, and the default value is 5 minutes; obtaining the average value of each voltage data in the dynamic window at each moment and recording it as the window voltage, fitting each window voltage through a curve and obtaining the change rate of the curve at each moment and recording it as the voltage-averaging change rate, wherein the change rate of the curve is obtained by the first-order derivative; if the voltage-averaging change rate is greater than the preset voltage-changing threshold, it is recognized that the energy storage power supply mode is turned on.

[0006] This method uses a dynamic window to reduce the instantaneous interference that may exist in a single sampling, while retaining short-term trend changes and reducing the risk of false touches. Voltage data generally serves to monitor the power supply status of elevators for a long time, and no additional test equipment is required. Usually, the value range of the power change threshold is 2.5V / s-10V / s; the power change threshold selection process should consider that if it is just a normal UPS switching, it is set to 2.5V / s-5V / s, and the sudden switching of supercapacitor protection is set to 5.0 V / s-7.0 V / s. If it is directly connected to the diesel generator mode, it is 5.0V / s-10.0V / s.

[0007] Furthermore, alternatively, a method for identifying the energy storage power supply mode by monitoring the elevator voltage fluctuation is: using a voltage sensor to collect the elevator voltage data in real time, setting the dynamic window to 2 to 10 minutes, and when the voltage data fluctuation range within the dynamic window exceeds ±5% to ±10%, it is identified that the energy storage power supply mode is turned on.

[0008] This method is suitable for simple PLC circuits in devices with limited computing resources.

[0009] Further, when the energy storage power supply mode is triggered, the method for aggregating continuous high-demand areas to form a merged layer through destination floor data is: intercepting a destination floor historical data set from the destination floor data in real time; wherein the destination floor historical data set is composed of a number of destination floor historical data records, each destination floor historical data record includes a real-time timestamp distance and a destination floor composition, and the real-time timestamp distance value range is 10 to 60 minutes; The DBSCAN clustering method is used to cluster the historical data set of the destination floors. The clustering results are filtered out by the LOF algorithm to obtain several clusters. The cluster with the largest number of elements is the target cluster. All target floors are identified from the target cluster, and any set of consecutive target floors is recorded as a merged layer.

[0010] The destination floor data here are data records consisting of a timestamp and a destination floor, and these data records are limited to destination floor data records starting from the first floor or the base floor. The base floor usually refers to the main floor where the elevator takes external passengers to different floors in the building.

[0011] The real-time timestamp distance refers to the time difference between the timestamp in the destination floor data record and the current time; the timestamp reading range is 24 hours; the time difference includes positive and negative values.

[0012] The LOF algorithm is a local outlier factor identification algorithm; The upper limit of the merged floors is preset to 2~6 floors. When the number of merged floors exceeds the upper limit, the merged floor will be divided into 2 merged floors. In the default state, the elevator will transport passengers from the first floor or the basic floor to the highest target floor among the merged floors.

[0013] The merged layer is to merge adjacent high-demand floors into one parking layer, which effectively reduces the number of elevator starts and stops and reduces energy consumption. The reduction in the number of starts and stops is conducive to maintaining the long-term operation status of the energy storage power supply and extending the recovery time of the energy storage power supply.

[0014] Passengers starting from the first floor select their target floor in the elevator lobby, and the display in the elevator lobby diverts the passengers, that is, guides the passengers to the corresponding elevator at the target floor.

[0015] The construction of the merged layer is based on the density-based clustering method, which effectively extracts regularity from the target floor demands of passengers in a specific period of time, reduces the sensitivity of floors with low density of stop demands, and then reasonably allocates different passengers to the corresponding elevators, making the operation path of each elevator more concentrated, reducing the stop rate while also suppressing the empty rate.

[0016] Furthermore, a method for judging whether the service scope of the elevator is overloaded according to the elevator load is as follows: the feedback interval is preset to 2 seconds to 20 seconds, and the time point corresponding to each feedback interval is recorded as a monitoring point. The elevator load includes the power consumption value and power peak value obtained once at each monitoring point, wherein the power consumption value is the average power consumption in the dynamic window corresponding to the monitoring point.

[0017] The dynamic window corresponding to the monitoring point refers to the time period in the reverse time direction of the monitoring point, with a value range of 2 to 5 minutes; The average power consumption is calculated as the ratio of the total power consumption in the dynamic window to the time size of the dynamic window.

[0018] Further, the method for judging whether the service range of the elevator is overloaded according to the elevator load is: For the power consumption value and power peak data set of the monitoring point, dynamic normalization processing is performed through a multi-dimensional quantile discretization framework. The specific process is as follows: The parameter space of power consumption and power peak is divided into four interval units by using the quantile-driven discretization strategy, and the probability density distribution of each interval is ensured to meet the non-uniformity constraint; the four interval units are not necessarily symmetrical; the interval unit is abbreviated as unit; Define the two extreme boundaries of any interval unit to extract its internal power consumption value as the extreme difference amplitude; The discrete intensity of the power consumption fluctuation within the unit is quantified by the range amplitude, which is the difference between the maximum power consumption observation value and the minimum power consumption observation value within the unit; the identified two extreme boundaries refer to the maximum and minimum values, so the range amplitude is used as a scalar measure of the disorder of energy consumption within the unit ∆ j ; Perform bimodal statistics extraction on interval units: the median value of the power consumption value is used as the robust central estimator E of the energy consumption distribution within the unit j ; The median value of the power peak is used as a typical representation of the load intensity within the unit P j ; The probability mass function of power consumption weight is constructed based on the global contribution ratio of the range amplitude: the power consumption weight of a unit is the normalized ratio of its range amplitude to the sum of the range amplitudes of all units; that is, the power consumption weight of a unit is the ratio of its range amplitude to the sum of the power consumption weights of all interval units, and the ratios obtained for each interval unit are normalized; The calculation process of the power consumption weight of the required unit is actually to accurately capture the load fluctuation trend. When the elevator starts and stops frequently or the load density fluctuates greatly during operation, the electrochemical reaction inside the battery shows rapid periodic changes. The power consumption weight captures the range of energy consumption in the interval through the quantile discretization technology. This range represents the rapid change degree of the elevator load from low to high. When the power consumption weight is high, it implies that the elevator has experienced a large power fluctuation in a short period of time. This load fluctuation is directly reflected in the instantaneous power output of the battery. Frequent and drastic power fluctuations lead to accelerated chemical reaction rates inside the battery, and the diffusion and exchange rates of electrolyte ions on the electrode surface change frequently, causing an increase in internal resistance and generating additional Joule heat, causing the local temperature of the battery to rise rapidly. For a fixed four-story trip condition, the energy consumption of the elevator may vary from 14kJ to 270kJ under different conditions. The non-uniform performance of the energy consumption distribution of the elevator further proves that the internal structure of the battery deteriorates rapidly and the battery capacity decays rapidly under time accumulation. Therefore, this trend needs to be involved in identifying the risk of battery degradation through the curve function constructed later.

[0019] At the same time, the construction of power weight depends on the relative strength of the median power peak value within the unit, that is, the ratio of the median power peak value of each unit to the sum of the median power peak values ​​of all units, so as to map the heterogeneous impact of power distribution on system load; The current fluctuation of the motor during operation is directly affected by the load. Generally speaking, when the load of the elevator increases, the rotor induced current increases, the torque output increases and the power increases. Therefore, when individual elevators are in the high power range for a long time while the load of other elevators is relatively low, it means that there is a high-low power load difference in the operation process, which further causes the power supply to reach the critical state of deep discharge of the battery in advance, causing the power supply potential to be significantly reduced and unable to stably output the required power. Therefore, it is necessary to use the power weight to construct the curve function to further identify the risk of battery degradation caused by the existence of this load imbalance.

[0020] The hyperbolic tangent function is constructed by power weight and power consumption weight to obtain the imbalance factor. Through the dynamic normalization mechanism driven by the segmented median value, this method can eliminate the dimensional difference while retaining the local nonlinear characteristics of the elevator load. Based on the design of the hyperbolic tangent function, the implicit interference of the data of high-power consumption and low-power equipment is suppressed, and the system's data sensitivity to load mutation scenarios is enhanced.

[0021] The core principle of using the hyperbolic tangent function to quantify the degree of deviation between the power weight and the power consumption weight is that the hyperbolic tangent function changes little in the area close to 0, but changes steeply at extreme values, which helps to avoid misjudgment within the normal range. When the power weight or power consumption weight exceeds the set threshold, the abnormal signal can be quickly amplified, making the impact of load abnormality or dust accumulation easier to identify. The set of all imbalance factors between the last elevator maintenance and the current one is recorded as the imbalance factor set. If the imbalance factors of all elevators meet the Z-Score ≥ 2.5 in the corresponding previous imbalance factor set, it is determined that the elevator is overloaded, and the overloaded elevator is pointed to the elevator corresponding to the maximum value in the imbalance factor set. The dynamic window method of the imbalance factor set is to limit its data history to the current 48 hours, so as to prevent the problem of sensitivity failure caused by too long maintenance cycle.

[0022] The current moment inherits the imbalance factor of the current monitoring point.

[0023] Here, the imbalance factor is measured through dynamic monitoring of the elevator load, so it can effectively quantify the excess characteristics of the elevator load process in actual operation. However, since the quantification process is directly related to the interval unit in which it is located, the analysis of the timing angle is defective, especially the data with a sudden increase in timing in a short period of time cannot be identified with high sensitivity. In order to solve this defect and reduce the sensitivity problem caused by this defect in timing angle analysis, another method for measuring the imbalance factor is provided.

[0024] The Z-Score threshold recognition method recommends the use of a dynamic window method because it can avoid data failure caused by battery capacity degradation during long-term operation. That is, the longer the battery is used, the greater the chance of judging that the elevator is overloaded. If the battery is a rechargeable working type, such as diesel power generation or solar power storage to supplement the power supply, a dynamic window must be defined. If the battery cannot be dynamically replenished, there is no need for dynamic recognition because its short operating cycle has little effect on the sensitivity of abnormal recognition.

[0025] Further, the method for judging whether the service range of the elevator is overloaded according to the elevator load is: In order to further deepen the real-time observation and judgment mechanism of overload phenomenon in the elevator loading process, this study proposed a load discrimination model based on time dynamic changes.

[0026] The independent variables σ are power consumption and peak power. T , σ P Based on, define the parameter load discrete level R TP , the parameter load discrete level is used to characterize the intrinsic reflection of the synergistic relationship between power consumption and power peak on the dynamic sensitivity of system load. In particular, by examining the change behavior of load discrete level over time, the first load discrimination model is constructed: ; The first load discrimination model combines the time derivative of the power consumption value and the power peak to express the degree of coupling effect of the nonlinear dynamic characteristics of the interaction between power consumption and power peak in the time dimension. Its principle is to capture the influence of the mismatch or synergy between power consumption and power peak in the elevator system on the load dynamics; where σ P 2 Ensure R TP The rate of change of is related to the absolute size of the power peak, avoiding scale misalignment caused by large differences in power peak amplitudes; dσ T / dt and σ P When the change is in the same direction, the mathematical contribution is positive, which means that the load may face increasing pressure; while σ T and dσ P When / dt changes in the opposite direction, the mathematical contribution is positive, which is a mitigation mechanism for the increasing degree of pressure and is a manifestation of the ability to maintain high-load operation. T >0) Power consumption increases rapidly while the peak power remains high (σ P If the value is larger, the problem is that the elevator system is under abnormal load, causing R TP The rate of change increases, otherwise it may reflect that the system has entered an inefficient operating state, which also affects the load dynamics. The manifestation of this coupling effect in the time dimension is the source of the nonlinear dynamic characteristics in the operation of the elevator system.

[0027] Based on the model constructed with the above parameters, by introducing the dynamic evolution behavior of the imbalance factor VIDX, a core differential equation describing the influence of the elevator load array on the imbalance factor is constructed, which is recorded as the second load discrimination model: ； The quantization parameter k 1 It represents the direct sensitivity caused by the power consumption value. Its intensity is determined by the specific dependence of the elevator group system on the change of power consumption value, combined with dσ T / dt is used to reflect the short-term sudden increase in power consumption data, including the power consumption changes caused by heavy load and frequent start and stop in the elevator. Therefore, it can be used to capture the direct contribution of power consumption changes to elevator load imbalance. The quantitative parameter k 2 It represents the indirect disturbance caused by the power peak change rate, reflecting the fatigue strength of the system under different power peak conditions. P / dt reflects whether the dynamic demand for elevator system power tends to be stable. When the value is larger, it indicates that the system operation state is unstable, which is easy to cause fatigue accumulation of mechanical parts. The quantitative parameter k 3 It represents the composite impact of load discrete level on the system, and is used to describe the weight effect of power consumption and power peak coupling dynamics. TP / dt captures the compound influence of dynamic changes in the coordinated calculation of power consumption and power peak on the imbalance factor. The larger the value, the stronger the unevenness of load distribution in the coupled dynamics and the worse the stability of the elevator load on the power demand.

[0028] k 1 , k 2 and k 3 It is the preset weight value, the value range is (0,1], and the default value is 1.

[0029] The set of all imbalance factors between the last elevator maintenance and the current one is recorded as the imbalance factor set. If the imbalance factors of all elevators meet the Z-Score ≥ 2.5 in the corresponding previous imbalance factor set, it is determined that there is an overloaded elevator, and the overloaded elevator is pointed to the elevator corresponding to the maximum value in the imbalance factor set.

[0030] Furthermore, when overload occurs, the method for reallocating the service range for the elevator group is: select and define the elevator that is overloaded as a re-planned elevator, whose service range is several consecutive floors, which are recorded as a service floor set, and remove the highest floor in the service floor set.

[0031] If the removed floor is not in the service range of any elevator after the removal, and there are other elevators whose service range is above the removed floor, the removed floor will be added to the service elevator corresponding to the upper floor.

[0032] Preferably, all undefined variables in the present invention, if not clearly defined, can be manually set thresholds.

[0033] The present invention also provides an emergency power supply-powered elevator adaptive diversion system with a merged layer strategy, the emergency power supply-powered elevator adaptive diversion system with a merged layer strategy comprises: a processor, a memory, and a computer program stored in the memory and executable on the processor, the processor implements the steps in the emergency power supply-powered elevator adaptive diversion method with a merged layer strategy when executing the computer program, the emergency power supply-powered elevator adaptive diversion system with a merged layer strategy can be run in computing devices such as desktop computers, laptop computers, PDAs, and cloud data centers, and the executable system may include, but is not limited to, a processor, a memory, and a server cluster, and the processor executes the computer program to run in the following system units: An energy storage power supply identification unit, used to identify the energy storage power supply mode by monitoring the elevator voltage fluctuation; The merging layer planning unit is used to form a merging layer by grouping continuous high-demand areas through destination floor data when the energy storage power supply mode is triggered; An overload identification unit is used to determine whether the service range of the elevator is overloaded according to the elevator load; The service range adjustment unit is used to reallocate the service range for the elevator group when overload occurs.

[0034] The beneficial effects of the present invention are as follows: the present invention provides an adaptive diversion method for elevators with a merged layer strategy powered by an emergency power supply. On the one hand, the merge quality of the merged layer is improved by adaptively establishing a merged layer according to a time period, so that the parking rate is reduced during the application of the merged layer strategy and the idle rate can also be effectively reduced. On the other hand, the service range is adjusted for the problem of excessive load on individual elevators during the application of the merged layer strategy, thereby reducing the working pressure of high-load elevators. The balanced power consumption between elevators makes the cooperation between parallel energy storage power supplies more stable, reduces the probability of sudden load shocks, and thus enhances the stability and continuity of the power supply of the microgrid. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The above and other features of the present invention will become more obvious by describing in detail the embodiments shown in the accompanying drawings. The same reference numerals in the accompanying drawings of the present invention represent the same or similar elements. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other accompanying drawings can be obtained based on these accompanying drawings without creative work. In the accompanying drawings: Figure 1 The figure shows a flow chart of an adaptive diversion method for elevators with a combined floor strategy powered by an emergency power supply; Figure 2 The figure shows a structural diagram of an elevator adaptive diversion system with a combined floor strategy powered by an emergency power supply. DETAILED DESCRIPTION

[0036] The following will be combined with the embodiments and drawings to clearly and completely describe the concept, specific structure and technical effects of the present invention, so as to fully understand the purpose, scheme and effect of the present invention. It should be noted that the embodiments and features in the embodiments of this application can be combined with each other without conflict.

[0037] like Figure 1 The figure shows a flow chart of an adaptive diversion method for elevators with a combined floor strategy powered by an emergency power supply. Figure 1 The present invention will be described in detail with reference to an embodiment of the present invention, wherein an adaptive traffic diversion method for an elevator powered by a combined layer strategy and an emergency power supply is provided. The method comprises the following steps: The energy storage power supply mode is identified by monitoring the voltage fluctuations of the elevator. After the energy storage power supply mode is triggered, each time the elevator stops at the first floor, the continuous high-demand areas are dynamically aggregated to form a merged layer through the destination floor data. The elevator load is used to determine whether the service scope is overloaded. If overload occurs, the service scope is reallocated for the elevator group.

[0038] Furthermore, a method for identifying the energy storage power supply mode by monitoring the elevator voltage fluctuation is: using a voltage sensor to collect the elevator voltage data in real time, and the voltage data measurement frequency range is 50 milliseconds / time; setting the dynamic window to 5 minutes; obtaining the average value of each voltage data in the dynamic window at each moment and recording it as the window voltage, fitting each window voltage through a curve and obtaining the rate of change of the curve at each moment and recording it as the voltage-averaging rate, wherein the rate of change of the curve is obtained by the first-order derivative; if the voltage-averaging rate is greater than the preset voltage-changing threshold, it is recognized that the energy storage power supply mode is turned on.

[0039] The window voltage is fitted with the least square method to obtain the curve V(t), then the voltage-balanced change rate is: V'(t) = dV / dt, the voltage-balanced change rate represents the rate of change of voltage over time, and the voltage-change threshold is 5.0 V / s; Furthermore, when the energy storage power supply mode is triggered, the method for aggregating continuous high-demand areas to form a merged layer through the destination floor data is: intercepting the destination floor historical data set from the destination floor data in real time; wherein the destination floor historical data set is composed of a number of destination floor historical data records, each destination floor historical data record includes a real-time timestamp distance and a destination floor composition, and the real-time timestamp distance is set to 20 minutes; The DBSCAN clustering method is used to cluster the historical data set of the destination floors. The clustering results are filtered out by the LOF algorithm to obtain several clusters. The cluster with the largest number of elements is the target cluster. All target floors are identified from the target cluster, and any set of consecutive target floors is recorded as a merged layer.

[0040] The upper limit of merged layers is preset to 4 layers. When the number of merged layers exceeds the upper limit, the merged layer will be split into 2 merged layers.

[0041] Passengers starting from the first floor select their target floor in the elevator lobby, and the display in the elevator lobby diverts the passengers, that is, guides the passengers to the corresponding elevator at the target floor.

[0042] In the embodiment, F plus a number represents a floor, the total number of floors is 21, and the elevator recognizes the merged floors as F{4, 5}, F{9, 10}, and F{18, 19, 20} respectively; F{4, 5} represents that the 4th and 5th floors are constructed as a merged floor, and so on for F{9, 10} and F{18, 19, 20}; therefore, when passengers select the 4th floor, they will actually be taken to the 5th floor, and then reach the 4th floor by stairs; when passengers select the 9th floor, they will actually be taken to the 10th floor, and then reach the 9th floor by stairs; when passengers select the 18th and 19th floors, they will actually be taken to the 20th floor, and then reach the 18th and 19th floors by stairs.

[0043] Furthermore, a method for judging whether the service scope of the elevator is overloaded according to the elevator load is as follows: the feedback interval is preset to 5 seconds, and the time point corresponding to each feedback interval is recorded as a monitoring point. The elevator load includes the power consumption value and power peak value obtained once at each monitoring point, wherein the power consumption value is the average power consumption in the dynamic window corresponding to the monitoring point.

[0044] The dynamic window corresponding to the monitoring point refers to the monitoring point in the reverse time direction, which is a time period of 2 minutes; the calculation method of the average power consumption is the ratio of the total power consumption in the dynamic window to the time size of the dynamic window.

[0045] Furthermore, the method for judging whether the service range of the elevator is overloaded is as follows: for the power consumption value and power peak data set of the monitoring point, dynamic normalization processing is performed through a multi-dimensional quantile discretization framework. The specific process is as follows: The parameter space of power consumption and power peak is divided into four interval units by using the quantile-driven discretization strategy, and the probability density distribution of each interval is ensured to meet the non-uniformity constraint; the four interval units are not necessarily symmetrical; the interval unit is abbreviated as unit; Define the two extreme boundaries of any interval unit to extract its internal power consumption value as the extreme difference amplitude; The discrete intensity of the power consumption fluctuation within the unit is quantified by the range amplitude, that is, the difference between the maximum power consumption observation value and the minimum power consumption observation value within the unit; the identified two extreme boundaries refer to the maximum and minimum values; Perform bimodal statistics extraction on interval units: the median value of the power consumption value is used as the robust central estimator E of the energy consumption distribution within the unit j ; The median value of the power peak is used as a typical representation of the load intensity within the unit P j ; The range amplitude is used as a scalar measure of the disorder of energy consumption within the unit ∆ j ; The probability mass function of power consumption weight is constructed based on the global contribution ratio of the range amplitude: the power consumption weight of a unit is the normalized ratio of its range amplitude to the sum of the range amplitudes of all units; At the same time, the construction of power weight depends on the relative strength of the median power peak value within the unit, that is, the ratio of the median power peak value of each unit to the sum of the median power peak values ​​of all units; The imbalance factor is obtained by the hyperbolic tangent function, where the mathematical expression of the imbalance factor can be: ; Among them, i is the monitoring point number; j is the interval unit number; i(j) represents the interval unit corresponding to monitoring point i; n is the number of monitoring points; tanh(·) is the hyperbolic tangent function, which is used to compress the influence of extreme values. Through the dynamic normalization mechanism driven by the segmented median value, this method retains the local nonlinear characteristics of the elevator load while eliminating the dimension difference; The calculation principle of the imbalance factor here is that the hyperbolic tangent function in the numerator is the power consumption term, and the rest of the numerator is the power term. The power consumption term is used to amplify the sensitivity of load mutation. The larger the value, the stronger the degree of change of local power consumption. The power term compares the local relative deviation value with the global contribution weight. The larger the power term, the greater the deviation of power consumption from the normal working range in the short term. The right side of the formula is the global energy efficiency factor, which drives the VIDX value to the steady-state value of 1 through the correlation between total energy consumption and total power. When the normal load VIDX value tends to be stable, the overload VIDX value significantly overflows the steady state, thereby passing the VIDX abnormal monitoring.

[0046] The set of all imbalance factors between the last elevator maintenance and the current one is recorded as the imbalance factor set. If the imbalance factors of all elevators meet the Z-Score ≥ 2.5 in the corresponding previous imbalance factor set, it is determined that there is an overloaded elevator, and the overloaded elevator is pointed to the elevator corresponding to the maximum value in the imbalance factor set.

[0047] The current moment inherits the imbalance factor of the current monitoring point.

[0048] Further, the method for judging whether the service range of the elevator is overloaded according to the elevator load is: In order to further deepen the real-time observation and judgment mechanism of overload phenomenon in the elevator loading process, this study proposed a load discrimination model based on time dynamic changes.

[0049] The independent variables σ are power consumption and peak power. T , σ P Based on, define the parameter load discrete level R TP , the parameter load discrete level is used to characterize the intrinsic reflection of the synergistic relationship between power consumption and power peak on the dynamic sensitivity of system load. In particular, by examining the change behavior of load discrete level over time, the first load discrimination model is constructed: ； Based on the model constructed with the above parameters, by introducing the dynamic evolution behavior of the imbalance factor VIDX, a core differential equation describing the influence of the elevator load array on the imbalance factor is constructed, which is recorded as the second load discrimination model: ； The quantization parameter k 1 The direct sensitivity caused by the power consumption value is determined by the specific dependence of the elevator group system on the change of power consumption value; the quantitative parameter k 2 It represents the indirect disturbance caused by the power peak change rate, reflecting the fatigue strength of the system under different power peak conditions; the quantitative parameter k 3 It represents the composite impact of the load discrete level on the system, and is used to describe the weight effect of the dynamic coupling between power consumption and power peak. The values ​​of k1, k2 and k3 are 1; The set of all imbalance factors between the last elevator maintenance and the current one is recorded as the imbalance factor set. If the imbalance factors of all elevators meet the Z-Score ≥ 2.5 in the corresponding previous imbalance factor set, it is determined that there is an overloaded elevator, and the overloaded elevator is pointed to the elevator corresponding to the maximum value in the imbalance factor set.

[0050] In order to further reduce the impact of the system recovery capability gradually attenuating with the fluctuation intensity, the second load discrimination model can be optimized to the third load discrimination model: ; The self-recovery term in the third load discrimination model -VIDXe -I It is used to describe the behavior of the system's recovery ability gradually decaying with the intensity of fluctuation under high fluctuation conditions; where I is a scalar quantitative parameter that represents the influence intensity of the system under the combined effect of changes in power consumption, power peak, and the coupling relationship between the two, and is specifically defined as: It is worth noting that as the load discrete level fluctuation becomes more severe, the rate of change of the imbalance factor VIDX increases significantly. This indicates that the self-recovery ability of the system shows a significant downward trend, and the sensitivity of the imbalance factor to the combined floor elevator demand gradually increases.

[0051] Furthermore, when overload occurs, the method for reallocating the service range for the elevator group is: select and define the elevator that is overloaded as a re-planned elevator, whose service range is several consecutive floors, which are recorded as a service floor set, and remove the highest floor in the service floor set.

[0052] In the embodiment, F plus a number represents a floor, and the service floor set of elevator A is a set consisting of F2-F12, where F2-F12 refers to each consecutive floor between F2 and F12; the service range of elevator B is a set consisting of F10-F21; when elevator A is marked as a re-planning elevator, F12 is removed as the highest floor in the service floor set, so that the service floor set of elevator A is updated to F2-F11; when elevator B is marked as a re-planning elevator, F21 is removed as the highest floor in the service floor set, so that the service floor set of elevator A is updated to F10-F20; after F21 is removed, passengers who need to reach the F21 floor can only be sent to the F20 floor and reach the 21st floor via stairs, wherein the probability of overload will converge with the increase in the number of times the elevator is re-planned during the power supply operation process, so that the service range of each elevator in the elevator group tends to be stable.

[0053] If the removed floor is not in the service range of any elevator after the removal behavior, and there are other elevators whose service range is above the removed floor, the removed floor will be added to the service elevator corresponding to the previous floor. Assuming that the service floor set of elevator A is a set consisting of F2-F10, and the service range of elevator B is a set consisting of F11-F21, when elevator A is marked as a re-planned elevator, F10 is removed as the highest floor in the service floor set, so that the service floor set of elevator A is updated to F2-F9, and the service range of elevator B is changed to F10-F21.

[0054] An embodiment of the present invention provides an emergency power supply-powered elevator adaptive diversion system with a merged layer strategy, such as Figure 2The figure shows a structural diagram of an adaptive diversion system for an elevator with a merged layer strategy powered by an emergency power supply according to the present invention. The adaptive diversion system for an elevator with a merged layer strategy powered by an emergency power supply according to this embodiment comprises: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the adaptive diversion method embodiment of an elevator with a merged layer strategy powered by an emergency power supply are implemented.

[0055] The system comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to run in the following units of the system: An energy storage power supply identification unit, used to identify the energy storage power supply mode by monitoring the elevator voltage fluctuation; The merging layer planning unit is used to form a merging layer by grouping continuous high-demand areas through destination floor data when the energy storage power supply mode is triggered; An overload identification unit is used to determine whether the service range of the elevator is overloaded according to the elevator load; The service range adjustment unit is used to reallocate the service range for the elevator group when overload occurs.

[0056] The said adaptive diversion system for elevators with a merged layer strategy powered by an emergency power supply can be run in computing devices such as desktop computers, laptops, PDAs and cloud servers. The said adaptive diversion system for elevators with a merged layer strategy powered by an emergency power supply can run on systems that may include, but are not limited to, processors and memories. Those skilled in the art will appreciate that the said example is merely an example of an adaptive diversion system for elevators with a merged layer strategy powered by an emergency power supply, and does not constitute a limitation on an adaptive diversion system for elevators with a merged layer strategy powered by an emergency power supply, and may include more or fewer components than the example, or a combination of certain components, or different components, for example, the said adaptive diversion system for elevators with a merged layer strategy powered by an emergency power supply may also include input and output devices, network access devices, buses, etc.

[0057] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the operating system of the elevator adaptive diversion system with combined layer strategy powered by an emergency power supply, and uses various interfaces and lines to connect the various parts of the operating system of the elevator adaptive diversion system with combined layer strategy powered by an emergency power supply.

[0058] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the emergency power supply-powered merging layer strategy elevator adaptive diversion system by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of a mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0059] Although the description of the present invention has been quite detailed and has been described in particular with respect to several described embodiments, it is not intended to be limited to any of these details or embodiments or any particular embodiment, so as to effectively cover the intended scope of the present invention. In addition, the present invention is described above with the embodiments foreseeable by the inventors, and its purpose is to provide a useful description, and those non-substantial changes to the present invention that are not currently foreseen may still represent equivalent changes of the present invention.

Claims

1. An adaptive traffic diversion method for elevators with a merged floor strategy powered by an emergency power supply, characterized in that: The method comprises the following steps: identifying the energy storage power supply mode by monitoring the voltage fluctuation of the elevator; after the energy storage power supply mode is triggered, each time the elevator stops at the first floor, dynamically forming a merged layer by aggregating continuous high-demand areas through destination floor data; judging whether the service scope is overloaded according to the elevator load; and reallocating the service scope for the elevator group when overload occurs.

2. The method for adaptively splitting elevator traffic using a combined floor strategy with emergency power supply according to claim 1 is characterized in that: The method of identifying the energy storage power supply mode by monitoring the elevator voltage fluctuation is: use a voltage sensor to collect the elevator voltage data in real time, set the dynamic window to 2 to 10 minutes; obtain the average value of each voltage data in the dynamic window at each moment and record it as the window voltage, fit each window voltage through a curve and obtain the change rate of the curve at each moment and record it as the voltage-sharing change rate; if the voltage-sharing change rate is greater than the preset voltage-changing threshold, it is identified that the energy storage power supply mode is turned on.

3. The method for adaptively splitting elevator traffic using a combined floor strategy and powered by an emergency power supply according to claim 1 is characterized in that: The method of identifying the energy storage power supply mode by monitoring the elevator voltage fluctuation is: use a voltage sensor to collect the elevator voltage data in real time, set the dynamic window to 2~10 minutes, and when the voltage data fluctuation range in the dynamic window exceeds ±5%~±10%, it is identified that the energy storage power supply mode is turned on.

4. The method for adaptively splitting elevator traffic using a combined floor strategy powered by an emergency power supply according to claim 1 is characterized in that: When the energy storage power supply mode is triggered, the method of aggregating the continuous high-demand areas through the destination floor data to form a merged layer is as follows: the destination floor historical data set is intercepted from the destination floor data in real time; The destination floor historical data set is composed of several destination floor historical data records, each destination floor historical data record includes a real-time timestamp distance and a destination floor composition, and the real-time timestamp distance value range is 10 to 60 minutes; The DBSCAN clustering method is used to cluster the historical data set of the destination floors. The clustering results are filtered out by the LOF algorithm to obtain several clusters. The cluster with the largest number of elements is the target cluster. All target floors are identified from the target cluster, and any set of consecutive target floors is recorded as a merged layer.

5. The method for adaptively splitting elevator traffic using a combined floor strategy and powered by an emergency power supply according to claim 1 is characterized in that: The method for judging whether the service scope of an elevator is overloaded according to its load is as follows: the feedback interval is preset to be 2 seconds to 20 seconds, and the time point corresponding to each feedback interval is recorded as a monitoring point. The elevator load includes the power consumption value and the power peak value obtained once at each monitoring point, wherein the power consumption value is the average power consumption in the dynamic window corresponding to the monitoring point. The imbalance factor set is obtained according to the power consumption value and the power peak value. If the imbalance factor of each elevator meets the Z-Score ≥ 2.5 in its imbalance factor set, it is determined that the elevator is overloaded, and the overloaded elevator is pointed to the elevator corresponding to the maximum value in the imbalance factor set.

6. The method for adaptively splitting elevator traffic using a combined floor strategy and powered by an emergency power supply according to claim 5, characterized in that: The method for obtaining the imbalance factor set according to the power consumption value and the power peak value is as follows: when judging whether the service scope of the elevator is overloaded according to the elevator load, a quantile-driven discretization strategy is used to divide the parameter space of the power consumption and the power peak into four interval units respectively, and ensure that the probability density distribution of each interval satisfies the non-uniformity constraint; the range of the power consumption value in the interval unit is the range amplitude, and the median value of the power consumption value is the robust central estimator; the median value of the power peak value is the typical characterization quantity; the power consumption weight of the interval unit is the normalized ratio of its range amplitude to the sum of the range amplitudes of all interval units; the power weight of the interval unit is the ratio of its typical characterization quantity to the sum of the typical characterization quantities of all interval units; a hyperbolic tangent function is constructed through the power weight and the range amplitude to obtain the imbalance factor; the set of all imbalance factors between the last maintenance of the elevator and the current one is recorded as the imbalance factor set.

7. The method for adaptively splitting elevator traffic using a combined floor strategy powered by an emergency power supply according to claim 5, characterized in that: The method of obtaining the imbalance factor set according to the power consumption value and the power peak value is as follows: the independent variables σ of the power consumption value and the power peak value are T , σ P Based on, define the parameter load discrete level R TP The parameter load discrete level is used to characterize the intrinsic reflection of the synergistic relationship between power consumption and power peak on the dynamic sensitivity of the system load; the first load discrimination model is constructed by examining the change behavior of the load discrete level over time; the second load discrimination model describing the elevator load is constructed according to the first load discrimination model to obtain the imbalance factor; the set of all imbalance factors between the last elevator maintenance and the current one is recorded as the imbalance factor set.

8. The method for adaptively splitting traffic of an elevator with a merged floor strategy powered by an emergency power supply according to claim 1 is characterized in that: The method for reallocating the service range for the elevator group when overload occurs is: select and define the elevator when overload occurs as the re-planned elevator, and its service range is a floor range composed of several consecutive floors, which is recorded as a service floor set, and the highest floor in the service floor set is eliminated.

9. An emergency power supply-powered elevator adaptive diversion system with merged floor strategy, characterized in that: The adaptive diversion system for elevators with a merged layer strategy powered by an emergency power supply comprises: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the adaptive diversion method for elevators with a merged layer strategy powered by an emergency power supply are implemented as described in any one of claims 1 to 8. The adaptive diversion system for elevators with a merged layer strategy powered by an emergency power supply runs on computing devices such as desktop computers, laptop computers, PDAs, and cloud data centers.

Images