An elevator energy consumption optimization control method

Abstract

The present invention discloses an elevator energy consumption optimization control method, which relates to the technical field of elevator energy consumption optimization, and includes the following steps: During the operation of the elevator, the elevator operation data of the planned stop floors during this operation is obtained in real time, and the obtained elevator operation data is preprocessed to provide a basis for analyzing the usage pattern and operation efficiency of the elevator. Through the intelligent division of near, medium, and long-distance stops, the elevator system realizes refined speed and acceleration adjustment for different operation distances, effectively balances energy consumption and transportation efficiency, reduces the peak power demand and power waste. The near-distance stop reduces the instantaneous current impact and alleviates mechanical wear; the long-distance stop optimizes the acceleration, prevents excessive consumption under high load, and avoids the increase in equipment temperature. Overall, the intelligent control reduces equipment aging and heat dissipation requirements, significantly extends the elevator life, and reduces the long-term maintenance cost.

Description

Technical Field

[0001] The present invention relates to the technical field of elevator energy consumption optimization, and particularly to an elevator energy consumption optimization control method. Background Art

[0002] The energy consumption optimization of an elevator energy consumption optimization control system refers to the real-time monitoring of the elevator's operating status (such as load weight, floor stop frequency, acceleration and deceleration modes, etc.) through intelligent control algorithms and sensing technologies, analyzing the energy consumption characteristics of the elevator under different operating modes, and automatically adjusting the operating strategy, such as optimizing the acceleration and deceleration curves during startup and stop, reasonably arranging the elevator's operating route and stop times, or reducing no-load operation during off-peak hours, etc., so as to effectively reduce the electric energy consumption of the elevator in different usage scenarios, improve the overall energy efficiency, and reduce unnecessary energy waste.

[0003] The prior art has the following deficiencies:

[0004] In the prior art, elevators usually move up or down at a fixed speed. However, when operating between different floor distances, this method will significantly increase energy consumption, especially when frequently stopping at close floors or running long distances. For frequent stops at close floors, the elevator needs to start and brake frequently, and the fixed-speed acceleration and deceleration generate frequent peak power demands, resulting in an increase in instantaneous current, leading to power waste and aggravating equipment wear. When stopping at long-distance floors, although the frequent starts and stops are reduced, the fixed-speed operation prolongs the high-energy-consuming operation time, especially under heavy loads, resulting in obvious energy loss, accelerating equipment aging, and at the same time, the system temperature rises, increasing the energy consumption of the cooling equipment.

[0005] The above information disclosed in the background art section is only used to strengthen the understanding of the background of the present disclosure, and therefore it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0006] The object of the present invention is to provide an elevator energy consumption optimization control method. By intelligently dividing the close, medium, and long-distance stops, the elevator system realizes refined speed and acceleration adjustment for different operating distances, effectively balancing energy consumption and transportation efficiency, reducing peak power demands and power waste. Close stops reduce instantaneous current impact and relieve mechanical wear; long-distance stops optimize acceleration to prevent excessive consumption under high loads and avoid equipment temperature rise. Overall, the intelligent control reduces equipment aging and heat dissipation requirements, significantly extends the elevator's lifespan, and reduces long-term maintenance costs to solve the problems in the above background art.

[0007] To achieve the above object, the present invention provides the following technical solution: An elevator energy consumption optimization control method, comprising the following steps:

[0008] During the elevator operation, the elevator operation data of the planned stop floors during this operation is obtained in real time, and the obtained elevator operation data is preprocessed to provide a basis for analyzing the elevator usage pattern and operation efficiency;

[0009] Feature extraction is performed on the preprocessed elevator operation data. Based on the extracted features, the trained machine learning model is used to intelligently evaluate the changes in elevator stop floors, and identify potential patterns and trends in elevator operation;

[0010] According to the evaluation results of the machine learning model, the stop floors of the elevator are divided into short-distance stops, medium-distance stops, and long-distance stops;

[0011] For the case of medium-distance stops, combined with the load data of the elevator, the elevator operation speed is reset;

[0012] For short-distance stops and long-distance stops, according to the load data and the elevator operation speed reset for medium-distance stops, the acceleration and deceleration are intelligently adjusted.

[0013] Preferably, feature extraction is performed on the preprocessed elevator operation data. The extracted features include the distribution density between call requests on different floors and the offset distance of the elevator from the current floor to the farthest target floor. After obtaining them, under the detection window, after analyzing the distribution density between call requests on different floors and the offset distance of the elevator from the current floor to the farthest target floor, a call distribution sparsity index and a high-rise offset index are respectively generated. The call distribution sparsity index and the high-rise offset index are input into the pre-learned machine learning model, and a running span coefficient is generated through the machine learning model. The preprocessed elevator operation route is intelligently evaluated through the running span coefficient.

[0014] Preferably, under the detection window, the distribution density between call requests on different floors is analyzed to generate a call distribution sparsity index. The specific steps are as follows:

[0015] Under the detection window, collect the floors of all call requests and sort them in chronological order to form a call request floor sequence , where, , in the formula, is the floor number of the th call request, is the total number of call requests;

[0016] For each pair of call request floors in the call request floor sequence , calculate the distance between the floors and store these distances in the floor distance weight matrix . The calculation expression is as follows:

[0017] , where is the floor - to - floor distance weight, representing the element of the floor - to - floor distance weight matrix , representing the th and th call request, and the distance weight between them, is the floor number of the th call request, is the distance weight factor;

[0018] Based on the floor - to - floor distance weight , construct the dispersion increment matrix to reflect the distribution sparsity between call requests. The calculation expression is as follows:

[0019] , where is the dispersion increment value between the th and th call requests, is the call request distance influence index, is the time gap influence index, is a decimal to prevent division by zero;

[0020] Sum up all the dispersion increment values in the dispersion increment matrix to generate the call distribution sparsity index. The calculation expression is as follows:

[0021] , where is the call distribution sparsity index.

[0022] Preferably, under the detection window, analyze the offset distance of the elevator from the current floor to the farthest target floor to generate the high - rise offset index. The specific steps are as follows:

[0023] Under the detection window, first, set the starting floor of the elevator and calibrate the starting floor of the elevator as , and form a set of all target floors of the elevator in this operation. The set is , where is the first target floor in this elevator operation, is the total number of target floors in this elevator operation, and generate the offset distance. The calculation expression is as follows: , where represents the farthest target floor in this operation, represents the offset distance;

[0024] To improve the recognition of the offset distance in long - distance operations, introduce polynomial non - linear conversion for the offset distance Convert to a higher-order offset, define the offset distance after non-linear conversion, and the calculation expression is as follows:

[0025] , where is the offset distance after non-linear conversion, is the non-linear adjustment coefficient, is the coefficient of the quadratic offset term, used to control the influence of the term, is the coefficient of the 1.5th power offset term, used to adjust the influence of the term, and further increase the non-linear weight of long distances;

[0026] In actual operation, the load and time factors will affect the elevator energy consumption. Introduce these two correction factors, the load and time factors, and calculate the corrected offset distance. The calculation expression is as follows:

[0027] , where is the corrected offset distance, is the load, is the time factor, is the load influence coefficient, is the smoothing factor, is the time influence coefficient, is a decimal to prevent division by zero;

[0028] To further enhance the recognition of long-distance operation, apply exponential amplification to the corrected offset distance to generate the high-rise offset index. The calculation expression is as follows:

[0029] , where is the high-rise offset index, is the exponential amplification coefficient, is the smoothing factor of the denominator, is the adjustment factor of the offset distance.

[0030] Preferably, compare and analyze the running span coefficient generated after analyzing the elevator running route under the detection window with the preset running span coefficient reference range, and divide the obtained elevator running data. The specific division steps are as follows:

[0031] If the running span coefficient is less than the minimum value of the running span coefficient reference range, then divide the elevator running route into short-distance stops;

[0032] If the running span coefficient is within the running span coefficient reference range, then divide the elevator running route into medium-distance stops;

[0033] If the running span coefficient is greater than the maximum value of the reference range of the running span coefficient, the elevator running route is divided into long-distance stops.

[0034] Preferably, for short-distance stops and long-distance stops, according to the load data and the elevator running speed reset based on medium-distance stops, the specific steps for intelligently adjusting the acceleration and deceleration are as follows:

[0035] First, according to the real-time load of the elevator and the target running speed reset based on medium-distance stops , judge the reasonable target speed of the elevator at different stop distances. The expression is as follows:

[0036] , where is the target speed of the elevator running, is the weight coefficient, is the maximum load capacity of the elevator;

[0037] Based on the target speed of the elevator running and the short-distance stop and long-distance stop requirements, calculate the acceleration and deceleration. The calculation expressions are as follows:

[0038]

[0039] , where is the acceleration, is the default acceleration, is the current stop distance, is the maximum designed stop distance of the elevator, is the deceleration, is the default deceleration;

[0040] To improve the intelligent control effect, according to the sparse index of call distribution and the high-rise offset index , adjust the state update frequency of the elevator and adjust the node update interval. The calculation expression is as follows:

[0041] , where is the state update frequency, is the default update frequency of the elevator system, is the sparse index of call distribution adjustment coefficient, is the high-rise offset index adjustment coefficient;

[0042] According to the current acceleration of the elevator and deceleration The operating state under this condition is further optimized and adjusted through a feedback mechanism. The actual acceleration and deceleration should be adjusted in a timely manner according to the target speed and travel characteristics of the elevator operation. The expression is as follows: And make timely fine-tuning according to the travel characteristics, and the expression is as follows:

[0043] ,

[0044] , where, is the actual acceleration, is the current elevator operating speed, is the actual deceleration, is the feedback adjustment coefficient for adjusting the elevator acceleration, is the feedback adjustment coefficient of the deceleration;

[0045] Finally, perform the acceleration and deceleration control obtained from the above steps, and monitor the elevator operating state in real time to ensure that the control effect meets the expectations.

[0046] In the above technical solution, the technical effects and advantages provided by the present invention are as follows:

[0047] The present invention realizes refined speed adjustment for different operating distances by intelligently dividing the short-distance, medium-distance, and long-distance stops of the elevator. For medium-distance stops, the system dynamically adjusts the elevator's operating speed according to the load data to ensure a balance between energy consumption and transportation efficiency. Excessive speed will cause unnecessary energy consumption, while too slow speed will affect efficiency. The re-set speed can reduce power demand while ensuring passenger comfort, thereby optimizing energy efficiency. For short-distance stops, the adjustment of elevator acceleration and deceleration reduces the instantaneous current impact during frequent stops, reduces the peak power demand, and reduces power waste. At the same time, for long-distance stops, appropriate acceleration and deceleration are adjusted to avoid excessive consumption of equipment. Under the intelligent analysis and control of the system, the best speed and energy balance can be achieved for various stop distances, and the overall energy consumption of the elevator is reduced.

[0048] The present invention effectively alleviates the problems of frequent starts and stops and long-time high-load operation of the elevator at different stop distances through intelligent speed and acceleration control, and significantly reduces equipment wear. For short-distance stops, the system reduces the instantaneous high-current impact, reduces the load on the motor and transmission device, and avoids mechanical wear caused by frequent acceleration and sudden stops. For long-distance stops, intelligent control avoids excessive operation at high speed for a long time, especially in the case of heavy loads, preventing the equipment from overheating and accelerating wear. The system also reduces unnecessary energy loss through the optimization of heat dissipation equipment. These optimization strategies not only reduce the speed of equipment aging, but also reduce the additional load requirements of heat dissipation equipment, extend the operating life of the elevator, and reduce long-term maintenance and replacement costs at the same time. Brief Description of the Drawings

[0049] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0050] Figure 1 It is a flowchart of a method for optimizing elevator energy consumption control according to the present invention. Specific embodiments

[0051] Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that this disclosure will be more comprehensive and complete, and will fully convey the concept of the example embodiments to those skilled in the art.

[0052] The present invention provides a method for optimizing elevator energy consumption control as shown in Figure 1 and includes the following steps:

[0053] During the operation of the elevator, the elevator operation data of the planned stop floors during this operation is obtained in real time, and the obtained elevator operation data is preprocessed to provide a basis for analyzing the usage pattern and operation efficiency of the elevator.

[0054] During the start of the elevator operation, the elevator operation data of the planned stop floors during this operation is obtained in real time, including the call requests pressed by passengers on each floor and the target floors selected inside the elevator. By obtaining this information, the system can comprehensively understand the route of this elevator operation and provide the necessary data basis for subsequent analysis and optimization.

[0055] Feature extraction is performed on the preprocessed elevator operation data. Based on the extracted features, the trained machine learning model is used to intelligently evaluate the changes in elevator stop floors, and potential patterns and trends in elevator operation are identified.

[0056] Feature extraction is performed on the preprocessed elevator operation data. The extracted features include the distribution density between call requests on different floors and the offset distance of the elevator from the current floor to the farthest target floor. After obtaining them, under the detection window, the distribution density between call requests on different floors and the offset distance of the elevator from the current floor to the farthest target floor are analyzed, and a call distribution sparsity index and a high-rise offset index are respectively generated. The call distribution sparsity index and the high-rise offset index are input into the pre-learned machine learning model, and a running span coefficient is generated through the machine learning model. The preprocessed elevator operation route is intelligently evaluated through the running span coefficient.

[0057] The distribution density among call requests for different floors is sparse, which usually indicates that the elevator is operating in a long-distance and large-span state. A sparse distribution of call requests means that the floor distances between individual call requests are large, and the elevator needs to cross multiple intermediate floors without frequent stops. This distribution often occurs in high-rise buildings. For example, during peak hours in an office building, the elevator may need to directly cross multiple floors from a lower floor and stop at a designated floor on a higher floor, or during off-peak hours, the destination floors of passengers are more dispersed, resulting in the elevator having to travel to a relatively distant target floor after departing from the first floor. A dense distribution of call requests usually appears in areas where passengers gather to get on and off. In this case, the elevator stops at adjacent or closely spaced floors multiple times.

[0058] Under the detection window, analyze the distribution density among call requests for different floors to generate a call distribution sparsity index. The specific steps are as follows:

[0059] Under the detection window, collect the floors of all call requests and sort them in chronological order to form a call request floor sequence , where , in the formula, is the floor number of the th call request, is the total number of call requests;

[0060] For each pair of call request floors in the call request floor sequence , calculate the distance between the floors and store these distances in the floor distance weight matrix . The calculation expression is as follows:

[0061] , in the formula, is the floor distance weight, representing the element of the floor distance weight matrix , representing the distance weight between the th and the th call requests, is the floor number of the th call request, is the distance weight factor;

[0062] The floor distance weight refers to the distance value between each pair of call request floors, which is weighted through a power transformation (i.e., raised to a certain exponent) so that the farther the floor distance between a pair of requests, the greater the impact on the sparsity index. In the formula, this weight is represented by , where is the absolute distance between floors, raised to After taking the power as the weight. The purpose of this design is to emphasize those call requests with larger intervals, because large-span requests usually indicate that the elevator may be in a long-distance running state, and giving them a higher weight can more accurately reflect the characteristics of the elevator's long-distance running.

[0063] Distance weight factor is a hyperparameter that adjusts the influence degree of the distance between floors when calculating the sparsity index. By selecting a suitable value (usually greater than 1), the weight of long-distance requests can be amplified, making its contribution to the sparsity index more significant. A higher value will exacerbate the influence of long-distance call requests on the sparsity index, thus making the long-distance running mode more prominent in the sparsity index. In this way, the distance weight factor plays a key regulatory role in the index calculation, controlling the recognition sensitivity of the elevator's long-distance running state.

[0064] refers to all possible integer pairs from 1 to such that is satisfied. For these cases, the combinations between the th call request and the th call request and the th call request need to be calculated or considered.

[0065] Based on the distance weight between floors construct a dispersion increment matrix to reflect the distribution sparsity between call requests. The calculation expression is as follows:

[0066] , where is the dispersion increment value between the th and the th call requests, is the call request distance influence index that adjusts the contribution of long distance to sparsity, is the time gap influence index that increases the influence of time span on the sparsity index, is a decimal number to prevent division by zero, usually taking a value close to 0;

[0067] Call request distance influence index represents the influence degree of the distance between floors on sparsity when calculating the call distribution sparsity index. In the formula, it amplifies or weakens the influence of different floor distances on the sparsity index by raising the distance weight between floors to the th power. A higher value will significantly increase the contribution of a farther floor distance to the sparsity index, thus more prominently identifying the long-distance running characteristics; while a lower The value will weaken the influence of the floor spacing, making the sparsity index pay more attention to the nearby call requests. The influence index of call request distance here is to highlight the weight of floor spacing on sparsity, making the long-distance call requests more influential in the sparsity calculation, so as to more accurately reflect the running span of the elevator.

[0068] Time gap influence index Indicates the degree of influence of the time interval between different call requests on sparsity when calculating the call distribution sparsity index. It adjusts the influence of the time interval by performing a power transformation on the time difference between requests. A higher value will weaken the role of the time interval in the sparsity index, highlighting the spatial sparsity; while a lower value will amplify the role of the time gap, making the distribution sparsity index pay more attention to the concentration or dispersion in time. In this formula, the role of the time gap influence index is to adjust the temporal density of call requests, helping to identify those large-span operation demands caused by the proximity of call requests, so as to optimize the elevator dispatching.

[0069] Sum up all the dispersion increment values in the dispersion increment matrix to generate the call distribution sparsity index. The calculation expression is as follows:

[0070] , where is the call distribution sparsity index;

[0071] The larger the value of the call distribution sparsity index generated after analyzing the distribution density between call requests on different floors under the detection window, the more it usually indicates that the elevator is operating in a long-distance and large-span state. This index determines the running span of the elevator by analyzing the distribution density of call requests between different floors under the monitoring window. When the call requests are sparsely distributed, that is, the index value is large, it means that the floor distance between call requests is far, and the elevator needs to cross multiple floors without frequent stops during this operation, which is typical of long-distance operation characteristics. When the call distribution sparsity index is small, that is, the distribution density is high, it means that the elevator mainly stops between adjacent or nearby floors, and usually does not belong to large-span operation.

[0072] A rapid increase in the offset distance of the elevator from the current floor to the farthest target floor usually indicates that the elevator is in a long-distance and large-span operation state. The offset distance reflects the degree of floor span during this operation of the elevator. A rapidly increasing offset distance means that the elevator needs to cross a large number of intermediate floors on the way from the starting floor to the target floor. This situation usually occurs when the floor distribution is uneven or the passenger demand is mainly concentrated on floors that are far apart, such as running directly from a low floor to a high floor area. Long-distance operation not only requires the elevator to continuously run at high speed to optimize time efficiency, but also brings higher energy consumption due to large-span stops. The rapid increase in the offset distance forces the elevator to go through a series of actions such as acceleration, stable high speed, and deceleration in a short period of time, increasing the load and power consumption of the motor. Therefore, the rapid increase in the offset distance indicates a special mode of elevator operation. At this time, the elevator is in a long-distance and large-span operation state, and the speed and energy consumption need to be optimized to improve the operation efficiency and reduce the overall power consumption.

[0073] Under the detection window, analyze the offset distance of the elevator from the current floor to the farthest target floor to generate a high-rise offset index. The specific steps are as follows:

[0074] Under the detection window, first, set the starting floor of the elevator, that is, the current floor where the elevator is located, and calibrate the starting floor of the elevator as , form a set of all target floors in this operation of the elevator. The set is , where is the 1st target floor in this operation of the elevator, is the total number of target floors in this operation of the elevator. Generate the offset distance, and the calculation expression is as follows: , in the formula, represents the farthest target floor in this operation, represents the offset distance, which is the starting floor and the farthest target floor the absolute value of the floor difference between them;

[0075] is the last floor in the set of target floors, that is, the last target floor selected by the passengers in the elevator.

[0076] The set of target floors in the elevator is determined according to the target floors selected by the passengers after getting on the elevator at different floors, and is not arranged in the order of elevator operation. For example, if a passenger gets on the elevator on the 1st floor and selects floor, and another passenger gets on the elevator on floor and selects floor, the set of target floors may be recorded as , rather than in the order of elevator operation arranged.

[0077] Represents the set of target floors The maximum value in, that is, the target floor farthest from the starting floor in this run. It is the floor with the largest value selected from all target floors.

[0078] To improve the recognition of the offset distance in long-distance operation, a polynomial non-linear transformation is introduced to transform the offset distance Into a higher-order offset. Define the offset distance after non-linear transformation, and the calculation expression is as follows:

[0079] , where Is the offset distance after non-linear transformation, Is the non-linear adjustment coefficient, used to control the exponential amplification degree of the offset distance , Is the coefficient of the quadratic offset term, used to control The influence of the term, mainly used to strengthen the influence of the square of the offset distance, Is the coefficient of the 1.5th power offset term, used to adjust The influence of the term, further increasing the non-linear weight of long distances;

[0080] Non-linear adjustment coefficient Is a parameter used to adjust the exponential amplification effect of the offset distance . It acts directly on In the term, so that the offset distance shows a non-linear amplification effect in different sizes. A larger Value will significantly increase the contribution of the offset distance to the result, thus enhancing its importance during long-distance operation, while having less impact during short-distance operation. Therefore, The role of is to make this term more significantly reflect the effect of long distances on the system when the elevator is operating over long distances by non-linearly amplifying the influence of the offset distance.

[0081] Coefficient of the quadratic offset term Controls The weight of the term, which is a square term, mainly used to further strengthen the recognition of long distances. Since the square term can amplify the influence of the distance value more than the first power or 1.5th power, the contribution of this term to the result will increase significantly during long-distance operation. And The role is to adjust the influence of this square term, so that when the elevator runs long distances across multiple floors, the effect of the offset distance can be amplified, further enhancing the recognition of long-distance operation.

[0082] Coefficient of the 1.5th power offset term Adjusts The weight of the term, which lies between the first power and the second power, provides an intermediate degree of non - linear amplification. Compared with the sharp growth of the square term, the 1.5 - power growth is more gradual, enabling the offset distance to have a moderate impact on the result when operating at medium or long distances. Therefore, its function is to provide a relatively smooth increment, ensuring that the system's perception of long - distance operation is not overly extreme, but rather increases the contribution of the offset distance to the result in a moderate manner, thereby ensuring the stability and sensitivity of long - distance recognition.

[0083] In actual operation, the load (total passenger weight or number of passengers) and the time factor (estimated running time) will affect the elevator energy consumption. By introducing these two correction factors, the load and the time factor, the corrected offset distance is calculated. The calculation formula is as follows:

[0084] , where, is the corrected offset distance, is the load, is the time factor (estimated running time), is the load influence coefficient, used to adjust the degree of influence of the load on the offset distance, is the smoothing factor, which acts on the denominator of the load correction term to avoid result fluctuations caused by too small a denominator when the load is small, is the time influence coefficient, which adjusts the influence of the time factor (estimated running time) on the offset distance, is a decimal to prevent division by zero, usually taking a value close to 0;

[0085] The load influence coefficient is a parameter used to adjust the influence of the elevator load on the offset distance correction. During elevator operation, the load directly affects the power demand and energy consumption of the motor, especially during long - distance and large - span operation, where heavy loads significantly increase the system's load. By introducing the load influence coefficient , the weight of the load on the offset distance correction can be controlled. When has a large value, the contribution of the load to the corrected offset distance increases, and the system will pay more attention to the role of the load during long - distance operation, thereby reflecting a higher energy consumption demand under heavy - load conditions.

[0086] The time influence coefficient is a parameter that adjusts the influence of the elevator's expected running time on the offset distance correction. When the elevator operates for a long time, it is necessary to maintain the motor power, which gradually increases the energy consumption, especially more significantly during long - distance operation. The time influence coefficient ensures that the expected running time has an appropriate weight on the offset distance correction, such that when the running time is long, the system will increase the corrected offset distance Value, emphasizing the contribution of long - term operation to elevator energy consumption. Therefore, Its existence helps the system to more accurately reflect the actual impact of time factors on elevator energy efficiency in the context of long - distance and long - term operation.

[0087] To further enhance the recognition of long - distance operation, apply exponential amplification to the corrected offset distance to generate the high - rise offset index. The calculation expression is as follows:

[0088] , where is the high - rise offset index, is the exponential amplification coefficient, used to enhance the impact of the degree of offset, is the smoothing factor of the denominator, avoiding too large an index due to too small a denominator, is the adjustment factor of the offset distance, used to control the amplification effect of long - distance;

[0089] Under the detection window, the larger the performance value of the high - rise offset index generated after analyzing the offset distance of the elevator from the current floor to the farthest target floor, the more it indicates that the elevator is in a state of long - distance and large - span operation. The high - rise offset index evaluates the span of this operation by calculating the offset distance of the elevator from the current floor to the farthest target floor within the monitoring window. When the index value is large, it means that the elevator needs to cross a large floor interval, running directly from a lower floor to a higher floor (or vice versa), which belongs to a typical long - distance operation state, usually accompanied by long - term continuous high - speed operation, high energy consumption and equipment load. While when the high - rise offset index value is small, it indicates that the floor spacing of the elevator stops is short, belonging to a medium - short - distance operation scenario.

[0090] The machine - learning model is not limited here. Any machine - learning model that can realize the comprehensive analysis of the call distribution sparse index and the high - rise offset index to generate the operation span coefficient is acceptable. To implement the technical solution of the present invention, the present invention provides a specific implementation method;

[0091] The operation span coefficient is generated by the following formula:

[0092] , where , are the preset proportional coefficients of the call distribution sparse index and the high - rise offset index respectively, and , are both greater than 0.

[0093] As can be seen from the calculation expression of the running span coefficient, under the detection window, the larger the performance value of the call distribution sparsity index generated by analyzing the distribution density between different floor call requests, and the larger the performance value of the high-rise offset index generated by analyzing the offset distance of the elevator from the current floor to the farthest target floor, it indicates that the performance value of the running span coefficient generated by analyzing the elevator running route under the detection window is larger, indicating that the elevator is in a long-distance and large-span running state. On the contrary, it means that the floor spacing where the elevator stops is shorter, belonging to the running scenario of medium and short distances.

[0094] According to the evaluation results of the machine learning model, the stop floors of the elevator are divided into short-distance stops, medium-distance stops, and long-distance stops;

[0095] Compare and analyze the running span coefficient generated by analyzing the elevator running route under the detection window with the pre-set reference range of the running span coefficient, and divide the obtained elevator running data. The specific division steps are as follows:

[0096] If the running span coefficient is less than the minimum value of the reference range of the running span coefficient, the elevator running route is divided into short-distance stops;

[0097] If the running span coefficient is within the reference range of the running span coefficient, the elevator running route is divided into medium-distance stops;

[0098] If the running span coefficient is greater than the maximum value of the reference range of the running span coefficient, the elevator running route is divided into long-distance stops;

[0099] Short-distance stops refer to adjacent floors or floors with a difference of one or two floors. Medium-distance stops refer to stops with a moderate floor spacing, while long-distance stops refer to stops with a large floor spacing.

[0100] For the case of medium-distance stops, combine the load data of the elevator and re-set the elevator running speed;

[0101] For medium-distance floor stops, set a new running speed according to the load data and running efficiency. Medium-distance operation requires a balance between speed and energy consumption: too high a speed will increase energy consumption, and too low a speed will affect transportation efficiency. By optimizing the speed setting, unnecessary energy consumption can be reduced while ensuring passenger comfort.

[0102] For short-distance stops and long-distance stops, intelligently adjust the acceleration and deceleration according to the load data and the elevator running speed re-set for medium-distance stops;

[0103] For short-distance stops and long-distance stops, the specific steps to intelligently adjust the acceleration and deceleration according to the load data and the elevator running speed re-set for medium-distance stops are as follows:

[0104] First, based on the real-time load of the elevator and the target running speed reset according to the medium-distance stop , judge the reasonable target speed of the elevator at different stop distances, and the expression is as follows:

[0105] , where is the target speed of the elevator operation, is the weight coefficient, which is used to balance the speed change under different loads, is the maximum load capacity of the elevator;

[0106] The greater the load, the elevator should appropriately reduce the running speed to reduce the burden on the motor; when the load is light, a higher speed can be maintained to improve the operation efficiency.

[0107] Based on the target speed of the elevator operation and the short-distance stop and long-distance stop requirements, calculate the acceleration and deceleration, and the calculation expressions are as follows:

[0108]

[0109] , where is the acceleration, is the default acceleration, is the current stop distance (the distance between short-distance or long-distance floors), is the maximum designed stop distance of the elevator, is the deceleration, is the default deceleration;

[0110] When the stop distance is short (such as short-distance stop), by reducing the acceleration and deceleration, reduce the peak power demand and energy consumption; for long-distance stops, the acceleration and deceleration can be appropriately increased to speed up the journey and shorten the operation time.

[0111] The default acceleration and deceleration are usually determined by the design specifications and safety standards of the elevator, and are usually set as the maximum acceleration and maximum deceleration in the elevator control system to ensure safety and riding comfort. The specific values can be obtained through the model parameters of the elevator, which are usually preset by the manufacturer when the elevator leaves the factory and are indicated in the operation manual or technical specifications. For example, the default acceleration of a typical passenger elevator is usually between 0.5 - 1.0 m / s², and the deceleration range is similar to avoid discomfort to passengers when the elevator starts or stops. At the same time, these default values need to comply with relevant elevator safety standards (such as ISO8100) and local regulations to ensure safety and meet technical requirements.

[0112] To improve the intelligent control effect, according to the sparse index of call distribution and the high-rise offset index , adjust the status update frequency of the elevator, adjust the node update interval, and the calculation formula is as follows:

[0113] , where is the status update frequency, that is, the frequency at which the system adjusts and optimizes the elevator operation parameters (such as speed, acceleration, etc.) during operation, is the default update frequency of the elevator system, generally the basic frequency set by the manufacturer or management system during elevator installation and setting to ensure that the system has sufficient response speed under normal circumstances, is the call distribution sparsity index Adjustment coefficient, used to control the call distribution sparsity index The influence degree on the status update frequency, is the high-rise offset index Adjustment coefficient, used to control the high-rise offset index The influence degree on the status update frequency;

[0114] According to the current acceleration of the elevator and deceleration The operating state under, through the feedback mechanism for further optimization and adjustment, the actual acceleration and deceleration should be adjusted timely according to the target speed of the elevator operation and the stroke characteristics, and the formula is as follows:

[0115] ,

[0116] , where is the actual acceleration, is the current elevator operation speed, is the actual deceleration, is the feedback adjustment coefficient for adjusting the elevator acceleration, reflecting the influence degree of the gap between the actual speed and the target speed of the elevator on the acceleration adjustment, is the feedback adjustment coefficient of the deceleration, controlling the adjustment effect of the gap between the current speed and the target speed of the elevator on the deceleration;

[0117] When the gap between the actual speed and the target speed of the elevator is large, through amplifying the feedback adjustment, ensure that the elevator approaches the target speed quickly with appropriate acceleration or deceleration, so as to improve the operation smoothness and efficiency.

[0118] Finally, perform the acceleration and deceleration regulation obtained from the above steps, and monitor the elevator operation status in real time to ensure that the regulation effect meets the expectations. If situations such as the power peak exceeding the set range or unstable operation are found during the execution process, the system automatically adjusts various parameters to achieve dynamic optimization. The key parameters for monitoring include power consumption, operation time, and equipment temperature, so as to ensure the high efficiency and safety of the elevator operation under different stopping scenarios.

[0119] Through the intelligent division of short-distance, medium-distance, and long-distance stops of the elevator, the present invention realizes refined speed adjustment for different running distances. For medium-distance stops, the system dynamically adjusts the running speed of the elevator according to the load data to ensure a balance between energy consumption and transportation efficiency. Excessive speed will bring unnecessary energy consumption, while too slow speed will affect efficiency. The re-set speed can reduce the power demand while ensuring passenger comfort, thereby optimizing energy efficiency. For short-distance stops, the adjustment of elevator acceleration and deceleration reduces the instantaneous current impact during frequent stops, lowers the peak power demand, and reduces power waste. At the same time, for long-distance stops, appropriate acceleration and deceleration are adjusted to avoid excessive consumption of equipment. Under the intelligent analysis and control of the system, the best speed and energy balance can be achieved for various stop distances, overall reducing the energy consumption of the elevator.

[0120] Through the intelligent speed and acceleration control, the present invention effectively alleviates the problems of frequent start-stop and long-time high-load operation of the elevator at different stop distances, and significantly reduces equipment wear. For short-distance stops, the system reduces the instantaneous high-current impact, lowers the load on the motor and transmission device, and avoids mechanical wear caused by frequent acceleration and sudden stop. For long-distance stops, intelligent control avoids excessive operation at high speed for a long time, especially in the case of heavy loads, preventing the equipment temperature from being too high and wear from intensifying. The system also reduces unnecessary energy loss through the optimization of heat dissipation equipment. These optimization strategies not only reduce the speed of equipment aging, but also reduce the additional load requirements of heat dissipation equipment, extend the operation life of the elevator, and at the same time reduce the long-term maintenance and replacement costs.

[0121] Only some exemplary embodiments of the present invention have been described by way of illustration above. Undoubtedly, for those of ordinary skill in the art, various different ways can be used to modify the described embodiments without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the protection scope of the claims of the present invention.

Claims

1. An elevator energy consumption optimization control method, characterized in that: The following steps are involved: During the operation of the elevator, the elevator operation data of the planned stop floors in this operation is obtained in real time, and the obtained elevator operation data is pre-processed to provide a basis for analyzing the use mode and operation efficiency of the elevator; Extract features from pre-processed elevator operation data, and use the trained machine learning model to intelligently evaluate changes in elevator stop floors based on the extracted features, identifying potential patterns and trends in elevator operation; According to the evaluation results of the machine learning model, the elevator's stop floors are divided into short-distance stops, medium-distance stops, and long-distance stops; For mid-distance stops, the elevator speed is reset based on the elevator load data; For short-distance and long-distance stops, the acceleration and deceleration are adjusted intelligently according to the load data and the elevator running speed reset for medium-distance stops; The preprocessed elevator operation data is subjected to feature extraction. The extracted features include the distribution density of elevator call requests at different floors and the offset distance of the elevator from the current floor to the farthest target floor. After acquisition, the distribution density of elevator call requests at different floors and the offset distance of the elevator from the current floor to the farthest target floor are analyzed under the detection window, and then a call distribution sparse index and a high-level offset index are generated respectively. The call distribution sparse index and the high-level offset index are input into a pre-learned machine learning model, and an operation span coefficient is generated by the machine learning model. The preprocessed elevator operation route is intelligently evaluated by the operation span coefficient.

2. The elevator energy consumption optimization control method according to claim 1, characterized in that: Under the detection window, the distribution density of elevator call requests on different floors is analyzed to generate an elevator call distribution sparse index. The specific steps are as follows: In the detection window, all floors with elevator call requests are collected and sorted in chronological order to form an elevator call request floor sequence. ,in, , where It is The number of floors requested by the elevator call, is the total number of elevator call requests; Floor sequence for elevator call request For each pair of elevator request floors, calculate the distance between floors and store these distances in the inter-floor distance weight matrix The calculation expression is as follows: , where is the distance weight between floors, representing the distance weight matrix between floors The element of and The distance weight between elevator calls is It is The number of floors requested by the elevator call, is the distance weight factor; Based on the weight of the distance between floors , construct the discrete increment matrix To reflect the sparse distribution of elevator call requests, the calculation expression is as follows: , where It is and The discrete increment between elevator calls is is the elevator call distance impact index, is the time gap impact index, is a decimal number that prevents division by zero; The discrete increment matrix All discrete increment values ​​in The cumulative sum is used to generate the call distribution sparse index. The calculation expression is as follows: , where is the call distribution sparsity index.

3. An elevator energy consumption optimization control method according to claim 2, characterized in that: In the detection window, the offset distance of the elevator from the current floor to the farthest target floor is analyzed to generate a high-rise offset index. The specific steps are as follows: In the detection window, first, set the starting floor of the elevator and mark the starting floor of the elevator as , all the target floors of the elevator in this operation are grouped into a set, the set is ,in, It is the first target floor in this elevator operation. is the total number of target floors in this elevator operation, generating the offset distance. The calculation expression is as follows: , where It indicates the farthest target floor in this run. It indicates the offset distance; In order to improve the recognition of the offset distance in long-distance operation, a polynomial nonlinear transformation is introduced to convert the offset distance Convert to a higher order offset and define the offset distance after nonlinear conversion. The calculation expression is as follows: , where is the offset distance after nonlinear transformation, is the nonlinear adjustment coefficient, is the coefficient of the quadratic offset term, used to control The impact of the item, Is the coefficient of the 1.5-power offset term, used to adjust The influence of the term further increases the nonlinear weight of long distances; In actual operation, the load and time factors will affect the energy consumption of the elevator. The load and time factors are introduced as correction factors to calculate the corrected offset distance. The calculation expression is as follows: , where is the corrected offset distance, It is load-bearing. is the time factor, is the load influence coefficient, is the smoothing factor, is the time influence coefficient, is a decimal number that prevents division by zero; Further enhance the recognition of long distance runs, corrected offset distance Applying exponential amplification, the high-level offset index generated is calculated as follows: , where is the high-level deviation index, is the exponential magnification factor, is the smoothing factor of the denominator, is the adjustment factor for the offset distance.

4. The elevator energy consumption optimization control method according to claim 1, characterized in that: The operation span coefficient generated after analyzing the elevator operation route under the detection window is compared and analyzed with the preset reference range of the operation span coefficient, and the obtained elevator operation data is divided. The specific division steps are as follows: If the operating span coefficient is less than the minimum value of the operating span coefficient reference range, the elevator operation route is divided into short-distance stops; If the operating span coefficient is within the operating span coefficient reference range, the elevator operation route is divided into medium-distance stops; If the running span coefficient is greater than the maximum value of the running span coefficient reference range, the elevator running route is divided into long-distance stops.

5. The elevator energy consumption optimization control method according to claim 4, characterized in that: For short-distance and long-distance stops, the specific steps for intelligently adjusting acceleration and deceleration based on load data and the reset elevator running speed for medium-distance stops are as follows: First, according to the real-time load of the elevator and mid-distance stops reset target running speed , to determine the reasonable target speed of the elevator at different stopping distances, the expression is as follows: , where is the target speed of the elevator. is the weight coefficient, is the maximum load capacity of the elevator; Target speed based on elevator operation As well as the short-distance and long-distance docking requirements, calculate the acceleration and deceleration. The calculation expression is as follows: , where is the acceleration, is the default acceleration, is the current stop distance, is the maximum stopping distance of the elevator design. is the deceleration, is the default deceleration; In order to improve the effect of intelligent control, according to the sparse index of elevator call distribution and high-level deviation index , adjust the elevator status update frequency, adjust the node update interval, and the calculation expression is as follows: , where is the state update frequency, is the default update frequency of the elevator system, is the call distribution sparseness index Adjustment factor, is the high-level deviation index Adjustment factor; According to the current acceleration of the elevator and deceleration The actual acceleration and deceleration should be adjusted according to the target speed of the elevator. Make timely fine-tuning with the stroke characteristics, the expression is as follows: ， , where is the actual acceleration, is the current elevator running speed, is the actual deceleration, It is the feedback adjustment coefficient used to adjust the elevator acceleration. is the feedback adjustment coefficient of deceleration; Finally, execute the acceleration and deceleration control obtained in the above steps, and monitor the elevator operation status in real time to ensure that the control effect reaches the expected result.

Images