Method for assisting in formulating design specification of elevator

Abstract

The invention relates to the technical field of big data, and discloses a method for assisting in formulating elevator design specifications, which comprises the following steps: constructing a feature selection model, training the feature selection model by adopting historical operation data of elevator parts and corresponding specification parameters, and obtaining the trained feature selection model; all specification parameters of the elevator are obtained, and the specification parameters are input into the feature selection model to calculate the contribution value of each specification parameter; judging whether the corresponding specification parameter is the most critical feature of the corresponding operation data according to the contribution value; and determining the distribution condition of the operation data according to the most critical features, and screening the most critical features to obtain a feature-index association table. According to the method, the characteristics most relevant to the historical data of the elevator are selected as the core indexes for defining the elevator model according to the specification parameters of the elevator parts, the elevator design efficiency is improved, and an important data basis is provided for design and manufacturing of the elevator.

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and in particular to a method for assisting in formulating elevator design specifications. Background Art

[0002] As a very important high-altitude transportation tool in our daily life, elevators need to be designed and manufactured very strictly. There are many types of parameters for elevator design. When developing a type of elevator model, market demand and manufacturing costs must be considered. The design parameters of the elevator include the design life (number of times and duration) of the entire elevator and components, rated speed and load, the power of the motor and inverter, and the design parameters of each component. If these complex and diverse parameters are combined according to the optimal solution, the elevator model with the best cost can be designed while ensuring market demand, service life and quality.

[0003] However, if you want to define a new type of elevator, you need to consider the elevator's safety, production cost, and the rationality of the production scale. There are hundreds of elevator specifications. How to select the most important parameters among these parameters as key design indicators and make selections or range definitions for these indicators? Currently, we can only rely on the experience of experts or the task requirements of the marketing department or customers. This is inefficient, and if the direction is not well grasped, there will be risks such as cost and non-compliance with market dynamics when the elevator is mass-produced. Summary of the invention

[0004] The present invention provides a method for assisting in the formulation of elevator design specifications, which binds the historical operation data of the elevator with the specification parameters of the elevator, selects the features most relevant to the historical data of the elevator for the specification parameters of the elevator components, and uses them as the core indicators for defining the elevator model, thereby improving the efficiency of elevator design and providing important data basis for the design and manufacture of elevators.

[0005] The present invention provides a method for assisting in formulating elevator design specifications, comprising:

[0006] Constructing a feature selection model, and using historical operation data of elevator components and their corresponding specification parameters to train the feature selection model, and obtaining a trained feature selection model;

[0007] Obtain all specification parameters of the elevator, and input the specification parameters into the feature selection model to calculate the contribution value of each specification parameter;

[0008] Determine whether the corresponding specification parameter is the most critical feature of the corresponding operating data based on the contribution value;

[0009] The distribution of the operating data is determined by the most critical feature, and then the most critical feature is screened to obtain a feature-indicator association table; wherein the indicator is an indicator of the operating data.

[0010] Furthermore, before the step of constructing a feature selection model, training the feature selection model using historical operation data of elevator components and their corresponding specification parameters, and obtaining a trained feature selection model, the step further includes:

[0011] Obtain multiple components of the elevator and extract historical operation data of one of the components one by one;

[0012] Associating the historical operation data of the components of the elevator with all the specification parameters of the elevator to obtain the association relationship between the historical operation data and the specification parameters;

[0013] Perform data cleaning on historical operation data to improve model operation efficiency.

[0014] Furthermore, in the step of performing data cleaning on the historical operation data to improve the model operation efficiency, the data cleaning includes:

[0015] Remove incomplete specification data; incomplete specification data refers to data in which a single specification field has empty values ​​accounting for more than half of the data.

[0016] Furthermore, in the step of performing data cleaning on the historical operation data to improve the model operation efficiency, the data cleaning also includes:

[0017] Remove descriptive information fields; wherein the descriptive information fields are data that have no impact on data distribution.

[0018] Furthermore, in the step of performing data cleaning on the historical operation data to improve the model operation efficiency, the data cleaning also includes:

[0019] Remove scattered specification values; scattered specification values ​​are data with no commonality, scattered grouping, and unclear characteristics.

[0020] Furthermore, the step of determining the distribution of the operating data based on the most critical feature, and then screening the most critical feature to obtain a feature-indicator association table includes:

[0021] Loading the operating data with the most critical feature, generating a distribution graph according to the most critical feature, and determining whether the distribution graph conforms to a normal distribution;

[0022] If the distribution graph conforms to the normal distribution and can separate the levels of each distribution range, it is determined that the effect of the feature selection is obvious;

[0023] If the distribution graph appears in a set form, the span of the sampled data is increased, or the characteristic calculation of the operating data index is removed, and only the index with obvious characteristic-data distribution is retained to establish a characteristic-index association table.

[0024] Furthermore, after the step of determining the distribution of the operating data based on the most critical feature and then screening the most critical feature to obtain a feature-indicator association table, the following step is further included:

[0025] Extract historical operating data of other components of the elevator and determine their most critical characteristics;

[0026] The feature-index association tables of all elevator components are combined together to form a mapping relationship between features and indicators.

[0027] Furthermore, after the step of combining the feature-index association tables of all elevator components to form a mapping relationship between features and indices, the following step is further included:

[0028] Determine the index of each component of the elevator, and use a clustering method to associate the indexes corresponding to the selected features into a wide table; optionally, the clustering method may be K-means;

[0029] The characteristics of each component of the elevator and the reference values ​​of the operating data corresponding to the characteristics are combined to form a list, which serves as the basis for elevator design.

[0030] The present invention also provides a device for assisting in formulating elevator design specifications, comprising:

[0031] A construction module, used for constructing a feature selection model, and training the feature selection model using historical operation data of elevator components and their corresponding specification parameters, and obtaining a trained feature selection model;

[0032] An acquisition module, used for acquiring all specification parameters of the elevator, and inputting the specification parameters into the feature selection model to calculate the contribution value of each specification parameter;

[0033] A judgment module is used to judge whether the corresponding specification parameter is the most critical feature of the corresponding operation data according to the contribution value;

[0034] The screening module is used to determine the distribution of the operating data based on the most critical features, and then screen the most critical features to obtain a feature-indicator association table; wherein the indicator is an indicator of the operating data.

[0035] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0036] The present invention also provides a computer-readable storage medium on which a computer program is stored, and the computer program implements the steps of the above method when executed by a processor.

[0037] The beneficial effects of the present invention are:

[0038] The present invention uses a preset data cleaning rule to clean the historical data accumulated from a large amount of elevator long-term operation, and uses a target selection model to calculate the contribution value of each specification parameter, and determines the specification parameter as the most critical feature of the corresponding operation data indicator according to the contribution value, and then uses the most critical feature to load the original operation data, and evaluates whether the selected feature is obvious according to the data distribution, so as to establish a feature-indicator association table, so as to select the features most relevant to the elevator historical data as the core indicators for defining the elevator model, improve the elevator design efficiency, and provide important data basis for the design and manufacture of elevators. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A flow chart of a method for assisting in elevator design specification formulation according to the present invention.

[0040] Figure 2 It is a schematic diagram of data loading in the present invention.

[0041] Figure 3 This is a schematic diagram of the calculation results of the contribution value in the present invention.

[0042] Figure 4 Schematic diagram of characteristic data distribution in the present invention.

[0043] Figure 5 A schematic diagram of the structure of the device formulated for the auxiliary elevator design specifications of the present invention.

[0044] Figure 6 It is a schematic diagram of the internal structure of the computer device of the present invention.

[0045] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0046] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0047] When designing an elevator, the optimal combination of design parameters must be achieved. The goal is to reduce design redundancy so that the parameters of the entire elevator or components basically meet the use requirements while ensuring quality and safety, and use this as the basis for the design of new elevator types. The historical data accumulated from the long-term operation of a large number of elevators can be used as a data source for simulating the most common operating conditions of elevators, that is, the historical operating data and the specifications of the elevators must be bound together. However, there are hundreds of categories of specifications for elevators and components, so feature selection is first required, that is, selecting the features that are most relevant to the elevator's historical data as the core indicators for defining the elevator model.

[0048] like Figure 1As shown, one embodiment of the present invention is: a method for assisting in formulating elevator design specifications, comprising:

[0049] S1. Construct a feature selection model, and use historical operation data of elevator components and their corresponding specification parameters to train the feature selection model, and obtain a trained feature selection model;

[0050] In some embodiments, before step S1, the method further includes:

[0051] S01. Acquire multiple components of the elevator, and extract historical operation data of one of the components one by one;

[0052] S02. Associating the historical operation data of the components of the elevator with all the specification parameters of the elevator to obtain an association relationship between the historical operation data and the specification parameters;

[0053] An elevator has hundreds of parts or specifications of the entire elevator. If you design a new elevator series, if you calculate and evaluate all these hundreds of parameters, the workload will be very large, and there will be a large number of combinations, and the design results will not converge. In order to prevent the parameter combination from being too discrete, it is necessary to select the key features of the elevator design.

[0054] The most obvious difference between elevator models is the core system, which includes the door system, traction system, and brake system. The components involved in these systems include traction motors, inverters, door machines, brakes, wire ropes, etc. The operating data of these components, that is, historical operating data, are collected and stored in the big data system, and used as a prerequisite for guiding the selection of the most critical features, that is, the training target of the feature selection model; with historical operating data as the target and elevator specification parameters as input features, the most critical features of each operating indicator are obtained by constructing and training a feature selection model.

[0055] Extract the historical operation data of a certain component of the elevator. For example, if you want to design the parameters related to the door, extract the historical operation data of the door system and associate the historical operation data with all the specification parameters of the elevator according to the elevator ID. Figure 2 The collected historical operation data of the elevator and the specification parameters of the elevator are shown in FIG. 1 , wherein the gray part is the historical operation data (excerpt display), and the white part is the specification parameters of the elevator (excerpt display). Figure 2 By correlating multiple data in the data set, a correlation table of historical operation data and specification parameters can be obtained. The historical operation data is used as the training target of the feature selection model, and the specification parameters are used as the input of the model. By continuously training the feature selection model and optimizing the parameters of the model, a feature selection model with accurate prediction effect is obtained. Preferably, the feature selection model in the present invention adopts the Lasso model.

[0056] S03. Perform data cleaning on historical operation data to improve model operation efficiency. Data cleaning specifically includes:

[0057] Remove incomplete specification data; incomplete specification data refers to data in which a single specification field has empty values ​​accounting for more than half of the data.

[0058] In some embodiments, data cleaning further includes removing descriptive information fields; wherein the descriptive information fields are data that have no impact on data distribution.

[0059] In some embodiments, data cleaning also includes removing scattered specification values; wherein the scattered specification values ​​are data that have no commonality, are scattered in grouping, and have unclear characteristics.

[0060] In some embodiments, data cleaning also includes removing data in which a single value accounts for more than half of the volume in a specification field. This data makes it impossible to further subdivide the sample and therefore needs to be removed.

[0061] In addition, historical data can be normalized. Normalization is a way to simplify calculations, giving all data attributes equal weights, making comparison and aggregation between attributes easier, and making data convergence conditions better, so that no attribute value overrides other attribute values. Normalization also helps prevent machine learning algorithms that use distance metrics between attributes from producing distorted results and improves the efficiency of data analysis.

[0062] The data normalization methods include Min-Max, Z-Score, Sigmoid, decimal calibration, etc. The data normalization method of one embodiment of the present invention is:

[0063] Perform a linear transformation on the original data and map the original data to [0, 1] using the following function without changing the original data distribution. In the following function, the quasi-normalized data is represented by x, x new Then it is the normalized data of x, Max and Min represent the maximum and minimum values ​​of the current data respectively.

[0064]

[0065] S2. Obtain all specification parameters of the elevator, and input the specification parameters into the feature selection model to calculate the contribution value of each specification parameter;

[0066] The feature selection model can use any neural network model with feature selection function. In one embodiment of the present invention, the feature selection model is a Lasso model. The Lasso model and the cleaned data are used to traverse the index column of the historical operation data. The index is used as the training target of the model, and the specification parameter is used as the input feature. The model runs to calculate the contribution value of the specification parameter to the training target, such as Figure 3 As shown, the model calculated the contribution values ​​of the top five features, from which it can be seen that kmxs, i.e., the door opening form, has the highest contribution value of 100, which should be regarded as the most critical feature.

[0067] S3. judging whether the corresponding specification parameter is the most critical feature of the corresponding operation data according to the contribution value;

[0068] A score threshold is set for the contribution value. When the contribution value calculated by the feature selection model reaches the preset score threshold, the specification parameter can be determined to be the most critical feature of the corresponding operation data indicator. If it does not reach the preset score threshold, the specification parameter is filtered and the next step of calculation is performed. Figure 3 The most critical feature of the operating data indicator is the one with a contribution score of 100.

[0069] S4. Determine the distribution of the operating data based on the most critical feature, and then screen the most critical feature to obtain a feature-indicator association table, wherein the indicator is the indicator of the operating data. Specifically, it includes:

[0070] S41, loading the operation data with the most critical feature, generating a distribution graph according to the most critical feature, and determining whether the distribution graph conforms to a normal distribution;

[0071] S42. If the distribution graph conforms to the normal distribution and can separate the levels of each distribution range, it is determined that the effect of the feature selection is obvious;

[0072] After calculating each indicator, the key specifications (features) are obtained. The original data is loaded, and the selected features are evaluated based on the distribution of the data. For example, for the indicator of the average holding current of the door closing (close_avg_electric), the key feature obtained is the door machine model (mjxh). The data in the following step S02 is used according to the specification of the door machine model to generate a distribution diagram of close_avg_electric, as shown in the figure below. Figure 4 As shown, the distribution conforms to the normal distribution and can separate the levels of each distribution range, indicating that the effect of feature selection is obvious.

[0073] S43. If the distribution graph appears in a set form (long-tail, multi-peak form), it means that the feature corresponding to the indicator is not obvious. Therefore, in some embodiments, the span of the sampling data is increased, or the feature calculation of the operating data indicator is removed, and only the indicators with obvious feature-data distribution are retained, and a feature-indicator association table is established.

[0074] That is, if the distribution diagram of the calculated feature at different values ​​for a certain operating data indicator has a long tail or multiple peaks, only the indicators with obvious feature-data distribution are retained, and a feature-indicator association table is established.

[0075] For example:

[0076] The most relevant characteristics of the door system are: kmxs (door opening form), mjxh (door operator model);

[0077] Features - The most obvious indicators of data distribution are:

[0078] kmxs (switching gate current, switching gate speed), mjxh (switching gate current, switching gate duration, switching gate holding current).

[0079] After step S4, the method further includes:

[0080] S5. extract historical operation data of other components of the elevator and determine the most critical features of each;

[0081] S6. Combine the feature-index association tables of all elevator components to form a mapping relationship between features and indicators.

[0082] As in steps S5-S6 above, in some embodiments, the same method is used to select other systems or components in the elevator according to their respective characteristics, and to associate data indicators with obvious distribution, and combine them together to form a mapping relationship between characteristics and indicators.

[0083] For example:

[0084]

[0085] S7. Determine the index of each component of the elevator, and use a clustering method to associate the indexes corresponding to the screened features into a wide table; preferably, the clustering method in the present invention uses K-means, specifically: use the K-Means model to divide the data into multiple clusters, and take the mean value of each feature of the member data of each cluster as the reference value of the feature.

[0086] As shown in step S7 above, the reference index of the newly designed ladder series in each design specification (component or whole ladder) is determined by clustering method. Since feature selection has been done, it is only necessary to associate all the operating data indicators corresponding to the selected features into a wide table. The dimension is greatly reduced compared with the data before feature selection.

[0087] For each of the above features, according to the mapped index field, load the operation data of all elevators, and cluster the data (preferably using the K-Means model) into multiple clusters. For each feature, the mean of the member data of each cluster is taken as the reference value of the feature.

[0088] For example: cluster the traction machine power,

[0089]

[0090] When subsequently designing the elevator, engineers use the reference data in the list and their experience to decide how the specifications of each component or the entire elevator should be defined, which greatly improves work efficiency.

[0091] S8. Combining the characteristics of each component of the elevator and the reference values ​​of the operating data corresponding to the characteristics to form a list, which serves as a basis for elevator design.

[0092] As in step S8 above, when planning a new ladder type, these lists can be referenced to combine them as the basis for the key design indicators of each system or component of the new ladder type.

[0093] The present invention uses preset data cleaning rules to clean the historical data accumulated from a large amount of long-term operation of elevators, and uses a target selection model to calculate the contribution value of each specification parameter, and determines the specification parameter as the most critical feature of the corresponding operation data indicator according to the contribution value, and then uses the most critical feature to load the original operation data, and evaluates whether the selected feature is obvious according to the data distribution, so as to establish a feature-indicator association table, so as to select the features most relevant to the elevator historical data as the core indicators for defining the elevator model, improve the elevator design efficiency, and provide important data basis for the design and manufacture of elevators.

[0094] like Figure 5 As shown, the present invention also provides a device for assisting in formulating elevator design specifications, comprising:

[0095] Construction module 1 is used to construct a feature selection model, and use the historical operation data of elevator components and their corresponding specification parameters to train the feature selection model, and obtain the trained feature selection model;

[0096] An acquisition module 2 is used to acquire all specification parameters of the elevator, and input the specification parameters into the feature selection model to calculate the contribution value of each specification parameter;

[0097] A judgment module 3 is used to judge whether the corresponding specification parameter is the most critical feature of the corresponding operation data according to the contribution value;

[0098] The screening module 4 is used to determine the distribution of the operating data based on the most critical feature, and then screen the most critical feature to obtain a feature-indicator association table; wherein the indicator is an indicator of the operating data.

[0099] In one embodiment, it further includes:

[0100] A historical operation data acquisition module is used to acquire multiple components of the elevator and extract the historical operation data of one of the components one by one;

[0101] An association module, used for associating the historical operation data of the elevator components with all the specification parameters of the elevator to obtain an association relationship between the historical operation data and the specification parameters;

[0102] The data cleaning module is used to clean the historical operation data to improve the model operation efficiency.

[0103] In one embodiment, the data cleaning module includes:

[0104] The first removal unit is used to remove incomplete specification data; wherein the incomplete specification data refers to data in which a single specification field has a null value accounting for more than half of the data.

[0105] In one embodiment, the data cleaning module further includes:

[0106] The second removing unit is used to remove the descriptive information field; wherein the descriptive information field is data that has no effect on data distribution.

[0107] In one embodiment, the data cleaning module further includes:

[0108] The third removal unit is used to remove scattered specification values; wherein the scattered specification values ​​are data with no commonality, scattered grouping, and unclear characteristics.

[0109] In one embodiment, the screening module 4 comprises:

[0110] A generating unit, configured to load the operating data with the most critical feature, generate a distribution graph according to the most critical feature, and determine whether the distribution graph conforms to a normal distribution;

[0111] A determination unit, used for determining that the effect of the feature selection is obvious when the distribution graph conforms to the normal distribution and can separate the levels of each distribution range;

[0112] The retaining unit is used to increase the span of the sampling data or remove the characteristic calculation of the running data index if the distribution diagram appears in a set form, retain only the index with obvious characteristic-data distribution, and establish a characteristic-index association table.

[0113] In one embodiment, it further includes:

[0114] An extraction module, used to extract historical operation data of other components of the elevator and determine the most critical features of each;

[0115] The combination module is used to combine the feature-index association tables of all elevator components to form a mapping relationship between features and indicators.

[0116] In one embodiment, it further includes:

[0117] An association module is used to determine the index of each component of the elevator, and associate the indexes corresponding to the selected features into a wide table by using a clustering method; optionally, the clustering method may be K-means;

[0118] The combination module is used to combine the characteristics of various components of the elevator and the reference values ​​of the operating data corresponding to the characteristics to form a list, which is used as the basis for elevator design.

[0119] The above modules and units are used to execute the corresponding steps in the method for formulating the auxiliary elevator design specifications. The specific implementation method thereof is described in the above method embodiment and will not be repeated here.

[0120] like Figure 6 As shown, the present invention also provides a computer device, which can be a server, and its internal structure can be as shown in Figure 6 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store all data required for the process of the method for assisting the formulation of elevator design specifications. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the method for assisting the formulation of elevator design specifications is implemented.

[0121] Those skilled in the art will understand that Figure 6 The structure shown in is merely a block diagram of a portion of the structure related to the present application solution and does not constitute a limitation on the computer device to which the present application solution is applied.

[0122] An embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, any one of the above-mentioned methods for assisting in formulating elevator design specifications is implemented.

[0123] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing related hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0124] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0125] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for assisting in formulating elevator design specifications, characterized in that: include: Constructing a feature selection model, and using historical operation data of elevator components and their corresponding specification parameters to train the feature selection model, and obtaining a trained feature selection model; Obtain all specification parameters of the elevator, and input the specification parameters into the feature selection model to calculate the contribution value of each specification parameter; Determine whether the corresponding specification parameter is the most critical feature of the corresponding operating data based on the contribution value; The distribution of the operating data is determined by the most critical feature, and then the most critical feature is screened to obtain a feature-indicator association table; wherein the indicator is an indicator of the operating data.

2. The method for formulating auxiliary elevator design specifications according to claim 1, characterized in that: Before the step of constructing the feature selection model, training the feature selection model using the historical operation data of the elevator components and their corresponding specification parameters, and obtaining the trained feature selection model, the method further includes: Obtain multiple components of the elevator and extract historical operation data of one of the components one by one; Associating the historical operation data of the components of the elevator with all the specification parameters of the elevator to obtain the association relationship between the historical operation data and the specification parameters; Perform data cleaning on historical operation data to improve model operation efficiency.

3. The method for formulating auxiliary elevator design specifications according to claim 2, characterized in that: In the step of performing data cleaning on the historical operation data to improve the model operation efficiency, the data cleaning includes: Remove incomplete specification data; incomplete specification data refers to data in which a single specification field has empty values ​​accounting for more than half of the data.

4. The method for formulating auxiliary elevator design specifications according to claim 2, characterized in that: In the step of performing data cleaning on the historical operation data to improve the model operation efficiency, the data cleaning also includes: Remove descriptive information fields; wherein the descriptive information fields are data that have no impact on data distribution.

5. The method for formulating auxiliary elevator design specifications according to claim 2, characterized in that: In the step of performing data cleaning on the historical operation data to improve the model operation efficiency, the data cleaning also includes: Remove scattered specification values; scattered specification values ​​are data with no commonality, scattered grouping, and unclear characteristics.

6. The method for formulating auxiliary elevator design specifications according to claim 1, characterized in that: The step of determining the distribution of the operating data based on the most critical feature, and then screening the most critical feature to obtain a feature-indicator association table includes: Loading the operating data with the most critical feature, generating a distribution graph according to the most critical feature, and determining whether the distribution graph conforms to a normal distribution; If the distribution graph conforms to the normal distribution and can separate the levels of each distribution range, it is determined that the effect of the feature selection is obvious; If the distribution graph appears in a set form, the span of the sampled data is increased, or the characteristic calculation of the operating data index is removed, and only the index with obvious characteristic-data distribution is retained to establish a characteristic-index association table.

7. The method for formulating auxiliary elevator design specifications according to claim 1, characterized in that: After the step of determining the distribution of the operating data based on the most critical feature and then screening the most critical feature to obtain a feature-indicator association table, the following step is also included: Extract historical operating data of other components of the elevator and determine their most critical characteristics; The feature-index association tables of all elevator components are combined together to form a mapping relationship between features and indicators.

8. The method for formulating auxiliary elevator design specifications according to claim 7, characterized in that: After the step of combining the feature-index association tables of all elevator components to form a mapping relationship between features and indices, the following steps are also included: Determine the indicators of each component of the elevator, and use clustering method to associate the indicators corresponding to the selected features into a wide table; The characteristics of each component of the elevator and the reference values ​​of the operating data corresponding to the characteristics are combined to form a list, which serves as the basis for elevator design.

9. A device for assisting in formulating elevator design specifications, characterized in that: include: A construction module, used for constructing a feature selection model, and training the feature selection model using historical operation data of elevator components and their corresponding specification parameters, and obtaining a trained feature selection model; An acquisition module, used for acquiring all specification parameters of the elevator, and inputting the specification parameters into the feature selection model to calculate the contribution value of each specification parameter; A judgment module is used to judge whether the corresponding specification parameter is the most critical feature of the corresponding operation data according to the contribution value; The screening module is used to determine the distribution of the operating data based on the most critical features, and then screen the most critical features to obtain a feature-indicator association table; wherein the indicator is an indicator of the operating data.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

Images