A Subgrade Construction Compaction Index Optimization System and Its Method

Abstract

The present invention relates to the technical field of subgrade construction. The present invention discloses a subgrade construction compaction index optimization system and method; it includes screening demand data from the first original data, extracting environmental data and construction data from the second original data, grouping the compaction indexes into index sets, optimizing a target set corresponding to the pre-data through a compaction index optimization model, parsing out the optimized indexes from the target set, and formulating the evaluation grades of the optimized indexes; compared with the prior art, the present invention can represent the user requirements, construction requirements and environmental conditions in the subgrade construction in multiple dimensions, avoiding the limitation problem of the optimized result of the compaction index. At the same time, through the artificial intelligence model, the compaction index can be quickly and accurately optimized, avoiding the inaccuracy caused by relying on manual experience and data query methods to optimize the compaction index, and improving the optimization effect of the subgrade construction compaction index.

Description

Technical Field

[0001] The present invention relates to the technical field of subgrade construction, and more specifically, to a subgrade construction compaction index optimization system and method thereof. Background Art

[0002] The compaction index of the subgrade is a key index for evaluating its bearing capacity and stability, and is also the basis for evaluating the quality of subgrade construction. Different types of compaction indexes correspond to the compaction quality evaluation effects of different dimensions of subgrade construction. In order to comprehensively and accurately evaluate the compaction quality of subgrade construction, it is necessary to select appropriate compaction indexes from numerous compaction indexes with the help of a compaction index optimization system according to different compaction quality analysis requirements.

[0003] The patent application with the reference publication number CN118607973A discloses a subgrade construction compaction index optimization method and system. It extracts features from the collected data, evaluates the correlation between the extracted features and the construction compaction effect index through statistical analysis or machine learning algorithms, and determines the weights of each feature index in combination with the analytic hierarchy process, so as to scientifically and objectively consider the weight distribution of intelligent subgrade construction compaction indexes. Based on the determined feature index weights and feature indexes, an optimization model is established, and the established optimization model is used for decision prediction and optimization, so as to develop an intelligent decision support system. According to the developed system, decision-making and optimization schemes can be generated, which can be applied to different actual environments and can meet the requirements of complex and changeable actual scenarios;

[0004] The existing compaction index optimization system collects construction information and external environment information in subgrade construction, and after a large amount of analysis and calculation of the collected information, relies on a model to perform the optimization operation of the compaction index. For example, in the above patent application, it collects and calculates a large amount of data related to soil characteristics, meteorological conditions and engineering equipment, and combines the optimization model to optimize the compaction index. This optimization method does not consider the personalized demand data of actual subgrade construction compaction evaluators, making the reference factors for subsequent compaction index optimization insufficiently comprehensive, resulting in limitations in the optimization results of the compaction index and unable to ensure the accuracy of the optimization results of the compaction index.

[0005] In view of this, the present invention proposes a subgrade construction compaction index optimization system and method to solve the above problems. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art and to achieve the above object, the present invention provides the following technical solution: A subgrade construction compaction index optimization system, which is applied to an optimization server, includes:

[0007] The first data acquisition module is used to receive the first raw data from the user side and screen out the required data from the first raw data. The required data includes calculation accuracy, effective optimization duration, and index optimization value.

[0008] The second data acquisition module is used to obtain the second raw data from the construction side and extract the environmental data and construction data from the second raw data.

[0009] The index set generation module is used to group the compaction indexes in the index library based on the grouping criteria to generate an index set.

[0010] The target set optimization module is used to aggregate the required data, environmental data, and construction data into pre-positioned data and optimize the target set corresponding to the pre-positioned data through the trained compaction index optimization model.

[0011] The target set analysis module is used to analyze the optimized indexes from the target set and formulate the evaluation levels of the optimized indexes. The evaluation levels include simple level, general level, and difficult level.

[0012] Furthermore, the screening methods for calculation accuracy, effective optimization duration, and index optimization value are as follows:

[0013] Taking the moment when the user side inputs data for the first time as the starting moment and the current moment as the ending moment, the time period between the starting moment and the ending moment is recorded as the required time period.

[0014] Query all the data input by the user side during the required time period through the database and arrange all the data in the order of input time to obtain the first raw data.

[0015] Identify the first semantics of all the first raw data one by one through natural language processing technology and split the first semantics into a text part and a digital part through word segmentation technology.

[0016] Respectively, the first semantics with the text parts of accuracy, duration, and quantity are recorded as accurate semantics, duration semantics, and quantity semantics, and the accurate semantics, duration semantics, and quantity semantics are sorted in the order of input time to obtain the first queue, the second queue, and the third queue.

[0017] Combine the accurate semantics at the last position in the first queue, the duration semantics at the last position in the second queue, and the quantity semantics at the last position in the third queue with the corresponding digital parts respectively to generate the calculation accuracy, effective optimization duration, and index optimization value.

[0018] Furthermore, the environmental data includes construction temperature value, construction humidity value, construction wind speed value, and soil type.

[0019] The extraction methods for construction temperature value, construction humidity value, construction wind speed value and soil type are as follows:

[0020] Query the construction period of subgrade construction through the database, and divide the construction period into A consecutive and equal-duration sub-periods;

[0021] Through temperature sensors, humidity sensors and wind speed sensors, respectively detect the temperature values, humidity values and wind speed values at the detection moments in A sub-periods, and obtain A real-time temperature values, A real-time humidity values and A real-time wind speed values;

[0022] Respectively eliminate the maximum and minimum values of the real-time temperature values, real-time humidity values and real-time wind speed values, and add up and average the remaining A real-time temperature values, A real-time humidity values and A real-time wind speed values one by one to obtain A sub-temperature values, A sub-humidity values and A sub-wind speed values;

[0023] Respectively add the maximum and minimum values of the sub-temperature values, sub-humidity values and sub-wind speed values and average them to obtain the construction temperature value, construction humidity value and construction wind speed value;

[0024] Query the soil data of subgrade construction through the database, identify the keywords of the soil data through natural language processing technology, and record the keywords as the soil type;

[0025] The soil type includes clay type, sand type, silt type and gravel type.

[0026] Furthermore, the construction data includes smooth wheel pressure fluctuation value, compaction times, compaction duration and compaction temperature;

[0027] The extraction method for the smooth wheel pressure fluctuation value is:

[0028] Query through the database the moments when the roller conducts smooth wheel compaction construction for the first and last times during the construction period, and record them as the compaction start moment and the compaction end moment respectively;

[0029] During the period from the compaction start moment to the compaction end moment, the smooth wheel pressure of the roller is detected in real time through a pressure sensor, and the moment when the smooth wheel pressure is greater than or equal to the preset pressure threshold for the first time is recorded as the effective start moment;

[0030] During the period from the effective start moment to the compaction end moment, the moment when the smooth wheel pressure is less than the preset pressure threshold for the last time is recorded as the effective end moment, and the period between the effective start moment and the effective end moment is recorded as the compaction period;

[0031] During the compaction period, randomly mark non - adjacent query times, and query the wheel pressures at the query times to obtain query pressures;

[0032] After successively subtracting the query pressure at the next query time from the query pressure at the previous query time, obtain pressure differences;

[0033] Sum up the pressure differences and then calculate the average to obtain the wheel pressure fluctuation value.

[0034] Furthermore, the grouping criterion is: there are no duplicate compaction indicators in the indicator set, and the number of compaction indicators in the indicator set is more than two.

[0035] Furthermore, the method for generating the indicator set is as follows:

[0036] Arrange all the compaction indicators in the indicator library in the order of storage time to generate an indicator queue;

[0037] Starting from the compaction indicator at the first position in the indicator queue and ending with the compaction indicator at the last position in the indicator queue, randomly select an indefinite number of compaction indicators from the indicator queue, and after combining the selected compaction indicators, obtain E verification sets;

[0038] Count the number of compaction indicators in each of the E verification sets one by one, and eliminate the verification sets with the number of compaction indicators less than 3 to obtain F identification sets;

[0039] Repeatedly identify the compaction indicators in each of the F identification sets one by one, eliminate the identification sets with duplicate compaction indicators, and record the remaining identification sets as the indicator set to obtain H indicator sets.

[0040] Furthermore, the training method of the compaction indicator optimization model is:

[0041] Pre - collect multiple groups of pre - data of subgrade construction and the corresponding indicator sets, and perform character conversion on the soil types in sequence. Convert the clay type to NT, the sand type to ST, the silt type to FT, and the gravel type to SS;

[0042] Mark each group of pre - data as training features, number the indicator sets of each group of training features, and record the numbered values of the indicator sets;

[0043] Use the training features as the input data of the compaction index optimization model, use the numbered values as the output data of the compaction index optimization model, divide the labeled training features into a training set and a test set, use the training set to train the compaction index optimization model, use the test set to test the compaction index optimization model, preset an error threshold, and obtain the compaction index optimization model when the mean of the prediction errors of all training features in the test set is less than the error threshold.

[0044] Further, output the optimized numbered values through the compaction index optimization model, and denote the index set with the same numbered values in the index set as the target set.

[0045] Further, the methods for formulating the easy level, general level, and difficult level are as follows:

[0046] Query the conventional collection duration and standard number of personnel of S optimized indicators one by one through the database, and calculate S levels of difficulty based on the S conventional collection durations and S standard numbers of personnel;

[0047] Count the number of optimized indicators with a difficulty level greater than the difficulty threshold, denote it as the difficulty quantity value, and denote one-fifth and one-half of the number of optimized indicators as the first critical value and the second critical value respectively;

[0048] When the difficulty quantity value is less than or equal to the first critical value, formulate the easy level;

[0049] When the first critical value is less than the difficulty quantity value and the difficulty quantity value is less than or equal to the second critical value, formulate the general level;

[0050] When the difficulty quantity value is greater than the second critical value, formulate the difficult level.

[0051] A method for optimizing subgrade construction compaction indicators, implemented based on the above subgrade construction compaction indicator optimization system, includes:

[0052] S1: Receive the first raw data from the user terminal, and screen out the required data from the first raw data. The required data includes calculation accuracy, effective optimization duration, and index optimization value;

[0053] S2: Obtain the second raw data from the construction terminal, and extract the environmental data and construction data from the second raw data;

[0054] S3: Group the compaction indicators in the index library based on the grouping criteria to generate an index set;

[0055] S4: Aggregate the required data, environmental data, and construction data into pre-data, and optimize the target set corresponding to the pre-data through the trained compaction index optimization model;

[0056] S5: Parse out the preferred indicators from the target set, and formulate the evaluation levels of the preferred indicators. The evaluation levels include simple level, general level, and difficult level.

[0057] The technical effects and advantages of a subgrade construction compaction index optimization system and method of the present invention:

[0058] By receiving the first raw data from the user terminal, screening out the demand data from the first raw data, obtaining the second raw data of the construction terminal, and extracting the environmental data and construction data from the second raw data, the present invention can comprehensively represent the user needs, construction requirements, and environmental conditions in subgrade construction in multiple dimensions. Thus, it can provide comprehensive and diverse data support for the evaluation and analysis of subgrade construction compaction quality, avoid the problem of limitations in the subsequent optimization results of compaction indicators, improve the accuracy of the optimization results of compaction indicators, and group the compaction indicators in the indicator library based on the grouping criteria to generate an indicator set, that is, perform an indefinite number of classification and combination operations on a large number of different types of compaction indicators, so that the indicator set can correspond and match any different quantities and different types of demand data, environmental data, and construction data, providing a screening basis for the subsequent optimization of compaction indicators. By summarizing the demand data, environmental data, and construction data into preposed data, and optimizing the target set corresponding to the preposed data through a trained compaction indicator optimization model, parsing out the preferred indicators from the target set, and formulating the evaluation levels of the preferred indicators, it can not only perform fast and accurate optimization operations on compaction indicators through an artificial intelligence model, avoid the inaccuracy caused by relying on manual experience and data query methods to optimize compaction indicators, but also represent the subsequent collection and evaluation difficulty levels of the optimized compaction indicators, and further provide comprehensive, accurate, and intuitive opinion guidance for the subgrade construction compaction indicator optimization operation, improving the optimization effect of subgrade construction compaction indicators. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a schematic structural diagram of a subgrade construction compaction index optimization system provided in Embodiment 1 of the present invention;

[0060] Figure 2 It is a schematic module diagram of a subgrade construction compaction index optimization system provided in Embodiment 1 of the present invention;

[0061] Figure 3 It is a schematic flow diagram of a subgrade construction compaction index optimization method provided in Embodiment 2 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0062] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0063] Embodiment 1: Please refer to Figure 1 and Figure 2 As shown, a subgrade construction compaction index optimization system described in this embodiment is applied to an optimization server and includes:

[0064] A first data acquisition module, which receives the first original data from the user terminal and filters out the required data from the first original data;

[0065] The user terminal refers to the using terminal that needs to use the system to optimize the subgrade construction compaction index, and serves as a data interaction and association platform between the system and the user. The first original data refers to all the basic data generated and collected in the user terminal, enabling the first original data to comprehensively represent the individual needs and situations of different user terminals and serving as the basis for subsequent filtering of the required data;

[0066] The required data refers to the requirements for unconventional personalized index optimization in the user terminal to achieve the optimization of the subgrade construction compaction index, serves as an accurate representation of the compaction index optimization in the user terminal, and is also one of the factors affecting the subsequent optimization result of the compaction index;

[0067] The required data includes calculation accuracy, effective optimization duration, and index optimization value;

[0068] The calculation accuracy refers to the minimum accuracy of data calculation in the compaction index when the user terminal optimizes the subgrade construction compaction index, and can serve as one of the reference factors for subsequent optimization of the compaction index. When the calculation accuracy is greater, it indicates that the user terminal has a higher requirement for the accuracy of data calculation in the compaction index.

[0069] The effective optimization duration refers to the maximum limit of the data acquisition duration related to the compaction index when the user terminal optimizes the subgrade construction compaction index, and can serve as one of the reference factors for subsequent optimization of the compaction index. When the effective optimization duration is greater, it indicates that the user terminal has a lower requirement for the data acquisition duration related to the compaction index.

[0070] The index optimization value refers to the representation of the specific quantity of the index in the compaction index when the user terminal optimizes the subgrade construction compaction index, and can serve as one of the reference factors for subsequent optimization of the compaction index. When the index optimization value is greater, it indicates that the user terminal has more requirements for the specific quantity in the compaction index.

[0071] The screening method for calculation accuracy, effective optimal duration, and index optimal value is as follows:

[0072] Taking the moment when the user terminal first inputs data as the starting moment and the current moment as the ending moment, the time period between the starting moment and the ending moment is recorded as the demand time period;

[0073] Query all the data input by the user terminal during the demand time period through the database, and arrange all the data in the order of input time to obtain the first original data; By arranging in chronological order, the orderliness of the data can be ensured, avoiding the phenomenon of chaotic and disorderly data, and thus facilitating subsequent data search, analysis, and use;

[0074] Use natural language processing technology to identify the first semantics of all the first original data one by one, and use word segmentation technology to split the first semantics into a text part and a digital part; The first semantics is a concise representation of the true meaning expressed by the first original data, which can achieve the simplification effect of complex data. The text part and the digital part are part of the first semantics, used to accurately represent the text and numbers of the first original data;

[0075] Respectively, the first semantics with the text parts of accuracy, duration, and quantity are recorded as accurate semantics, duration semantics, and quantity semantics, and the accurate semantics, duration semantics, and quantity semantics are sorted in the order of input time to obtain the first queue, the second queue, and the third queue;

[0076] Combine the accurate semantics at the last position in the first queue, the duration semantics at the last position in the second queue, and the quantity semantics at the last position in the third queue with the corresponding digital parts respectively to generate the calculation accuracy, effective optimal duration, and index optimal value.

[0077] It should be noted that since there may be input errors or different requirements at different times when the user terminal inputs relevant demand information, in order to most accurately represent the accurate data of the user terminal in compaction index optimization, the data with the latest input time needs to be used as the object.

[0078] The second data acquisition module obtains the second original data of the construction terminal and extracts the environmental data and construction data from the second original data;

[0079] The construction terminal refers to the user terminal that can provide the comprehensive subgrade construction data required for the compaction index optimization operation of the system, enabling the construction terminal to provide sufficient subgrade construction data for the system. The second original data refers to the data generated within the construction terminal that can comprehensively represent the specific situation of subgrade construction and serves as the extraction basis for subsequent environmental data and construction data;

[0080] Environmental data refers to the data of the environment where the subgrade construction is located when the construction side optimizes the compaction index of the subgrade construction. It is also one of the factors affecting the subsequent optimization results of the compaction index.

[0081] Environmental data includes construction temperature value, construction humidity value, construction wind speed value and soil type.

[0082] The construction temperature value refers to the environmental temperature at the location during the overall subgrade construction process. Different environmental temperatures will have different impacts on the compaction operation of the subgrade construction, thus affecting the subsequent optimization results of the compaction index.

[0083] The construction humidity value refers to the environmental humidity at the location during the overall subgrade construction process. Different environmental humidities will have different impacts on the compaction operation of the subgrade construction, thus affecting the subsequent optimization results of the compaction index.

[0084] The construction wind speed value refers to the environmental wind intensity at the location during the overall subgrade construction process. Different wind intensities will have different impacts on the water evaporation in the subgrade construction, thus affecting the subsequent optimization results of the compaction index.

[0085] The soil type refers to the type of soil at the location of the subgrade construction. Different soil types will have different impacts on the compaction degree of the subgrade construction.

[0086] The extraction methods of the construction temperature value, construction humidity value, construction wind speed value and soil type are as follows:

[0087] Query the construction period of the subgrade construction through the database, and divide the construction period into A consecutive and equal-duration sub-periods; the construction period refers to the period from the start of the subgrade construction to the end of the subgrade construction. By dividing the construction period into sub-periods, it can not only ensure the consistency of the duration of each sub-period, but also finely divide the period with a large duration span, reduce the quantity of data in each sub-period, and facilitate subsequent data query and calculation.

[0088] Detect the temperature value, humidity value and wind speed value at the detection moments in the A sub-periods through temperature sensors, humidity sensors and wind speed sensors respectively, and obtain real-time temperature values, real-time humidity values and real-time wind speed values;

[0089] Remove the maximum and minimum values of the real-time temperature values, real-time humidity values and real-time wind speed values respectively, and the remaining real-time temperature values, real-time humidity values and The real-time wind speed values are accumulated one by one and then averaged to obtain A sub-temperature values, A sub-humidity values, and A sub-wind speed values;

[0090] The expression for the sub-temperature value is:

[0091] ;

[0092] In the formula, is the sub-temperature value of the th sub-time period, = 1, 2... A, is the th real-time temperature value of the th sub-time period;

[0093] The expression for the sub-humidity value is:

[0094] ;

[0095] In the formula, is the sub-humidity value of the th sub-time period, is the th real-time humidity value of the th sub-time period;

[0096] The expression for the sub-wind speed value is:

[0097] ;

[0098] In the formula, is the sub-wind speed value of the th sub-time period, is the th real-time wind speed value of the th sub-time period;

[0099] The maximum and minimum values of the sub-temperature value, sub-humidity value, and sub-wind speed value are added and averaged respectively to obtain the construction temperature value, construction humidity value, and construction wind speed value;

[0100] The expression for the construction temperature value is:

[0101] ;

[0102] In the formula, is the construction temperature value, is the maximum value of the sub-temperature value, is the minimum value of the sub-temperature value;

[0103] The expression for the construction humidity value is:

[0104] ;

[0105] Wherein, is the construction humidity value, is the maximum value of the sub-humidity value, is the minimum value of the sub-humidity value;

[0106] The expression of the construction wind speed value is:

[0107] ;

[0108] Wherein, is the construction wind speed value, is the maximum value of the sub-wind speed value, is the minimum value of the sub-wind speed value;

[0109] Query the soil data of subgrade construction through the database, identify the keywords of the soil data through natural language processing technology, and record the keywords as the soil type.

[0110] Specifically, the soil type includes clay type, sandy soil type, silt type, and gravel type. Soils of each type will have different impacts on the subsequent compaction performance of subgrade construction, thereby affecting the results of subsequent compaction index optimization.

[0111] Construction data refers to the operation data of the equipment used for subgrade construction during the optimization of subgrade construction compaction indexes at the construction end, and is also one of the factors affecting the results of subsequent compaction index optimization;

[0112] Construction data includes the smooth wheel pressure fluctuation value, compaction times, compaction duration, and compaction temperature;

[0113] The smooth wheel pressure fluctuation value refers to the change range of the pressure of the smooth wheel of the roller on the subgrade during compaction construction, which can represent the compaction stability of the roller. When the smooth wheel pressure fluctuation value is larger, it indicates that the compaction stability of the roller is worse;

[0114] The extraction method of the smooth wheel pressure fluctuation value is:

[0115] Query the database to obtain the start and end times of the first and last smooth wheel compaction construction of the roller during the construction period, and record them as the compaction start time and the compaction end time respectively;

[0116] During the period from the compaction start time to the compaction end time, the smooth wheel pressure of the roller is detected in real time through a pressure sensor, and the time when the smooth wheel pressure is greater than or equal to the preset pressure threshold for the first time is recorded as the effective start time; the preset pressure threshold refers to the minimum value of the smooth wheel pressure that can keep the roller in a normal and stable compaction state, which can provide a basis for the acquisition range of the smooth wheel pressure to ensure that the collected smooth wheel pressure corresponds to the normal and stable compaction state of the roller;

[0117] During the effective start time to the compaction end time, the moment when the smooth wheel pressure is less than the preset pressure threshold for the last time is recorded as the effective end time, and the time period between the effective start time and the effective end time is recorded as the compaction time period;

[0118] During the compaction time period, randomly mark non - adjacent query times, and query the smooth wheel pressure at query times to obtain query pressures;

[0119] After successively subtracting the query pressure at the next query time from the query pressure at the previous query time, obtain pressure differences;

[0120] The expression of the pressure difference is:

[0121] ;

[0122] In the formula, is the th pressure difference, = 1, 2... , is the query pressure at the th query time, is the query pressure at the th query time;

[0123] After accumulating and averaging the pressure differences, obtain the smooth wheel pressure fluctuation value;

[0124] The expression of the smooth wheel pressure fluctuation value is:

[0125] ;

[0126] In the formula, is the smooth wheel pressure fluctuation value, is the th pressure difference.

[0127] The compaction times, compaction duration, and compaction temperature respectively refer to the number of times the smooth wheel of the roller compacts the roadbed, the length of the compaction time, and the level of the compaction temperature when the smooth wheel of the roller compacts the roadbed, so as to represent the specific compaction parameters of the roller; the compaction times, compaction duration, and compaction temperature are all obtained by querying the database.

[0128] The index set generation module groups the compaction indexes in the index library based on the grouping criteria to generate an index set;

[0129] The index library refers to a collection composed of diverse compaction indexes that can evaluate the compaction quality of subgrade construction, enabling the compaction indexes stored in the index library to meet the multi-dimensional index optimization requirements for subgrade construction compaction evaluation in various situations of different types, different requirements, different combinations, etc.;

[0130] Since the compaction indexes corresponding to different demand data, environmental data, and construction data for evaluating the compaction quality of subgrade construction are all different, and the number of compaction indexes used to evaluate the compaction quality of subgrade construction usually needs to be greater than 2. Therefore, in order to ensure that the compaction indexes in the index library can meet the corresponding compaction optimization requirements of a large amount of different demand data, environmental data, and construction data, it is necessary to classify and combine the compaction indexes in the index library according to different demand data, environmental data, and construction data, and combine the compaction indexes corresponding to each group of different demand data, environmental data, and construction data into an index set.

[0131] The compaction indexes include but are not limited to compaction degree, compaction value, elastic modulus, California bearing ratio, subgrade reaction coefficient, water content, porosity, relative density, shear strength, Benkelman deflection, etc., enabling a large number of compaction indexes to be grouped correspondingly in different quantities and different types;

[0132] When grouping the compaction indexes in the index library, grouping needs to be carried out under the restriction of grouping criteria to ensure that when the compaction indexes are grouped, the compaction indexes in each index set are in a unique state, and the number of compaction indexes in each index set is at least three;

[0133] The grouping criteria are: there are no duplicate compaction indexes in the index set, and the number of compaction indexes in the index set is more than two; this can ensure that the compaction indexes in the index set can maintain uniqueness, and the number of compaction indexes is at least three;

[0134] The generation method of the index set is as follows:

[0135] Arrange all the compaction indexes in the index library in sequence according to the chronological order of storage time to generate an index queue;

[0136] Starting from the compaction index at the first position in the index queue and ending with the compaction index at the last position in the index queue, randomly select an indefinite number of compaction indexes from the index queue, and after combining the selected compaction indexes, obtain E to-be-verified sets;

[0137] Count the number of compaction indexes in each of the E to-be-verified sets one by one, and eliminate the to-be-verified sets with the number of compaction indexes less than 3 to obtain F to-be-identified sets;

[0138] Repeatedly identify the compaction indicators in each of the F sets to be identified, eliminate the sets to be identified with repeated compaction indicators, and denote the remaining sets to be identified as indicator sets, obtaining H indicator sets.

[0139] A target set optimization module that aggregates the requirement data, environmental data, and construction data into preprocessed data, and optimizes the target set corresponding to the preprocessed data through a trained compaction indicator optimization model;

[0140] The preprocessed data is the overall representation after aggregating the requirement data, environmental data, and construction data, that is, it can orderly aggregate the scattered requirement data, environmental data, and construction data, and form the preprocessed data with an overall structure;

[0141] The compaction indicator optimization model refers to a machine learning model that can, based on the preprocessed data composed of requirement data, environmental data, and construction data, screen out the corresponding target set from the indicator sets. At this time, the target set is one of the indicator sets, and multiple compaction indicators in the target set can match the requirement data, environmental data, and construction data corresponding to the preprocessed data;

[0142] The training method of the compaction indicator optimization model is as follows:

[0143] Pre-collect multiple groups of preprocessed data for subgrade construction and the corresponding indicator sets, and perform character conversion on the soil types in sequence. Exemplarily, convert the clay type to NT, the sand type to ST, the silt type to FT, and the gravel type to SS;

[0144] Mark each group of preprocessed data as training features, number the indicator sets of each group of training features, and record the numbered values of the indicator sets. Exemplarily, when there are W groups of preprocessed data, then number the W corresponding indicator sets of the W groups of preprocessed data in ascending order one by one, so that the numbered values of the W indicator sets are 1, 2... W;

[0145] Use the training features as the input data of the compaction indicator optimization model, use the numbered values as the output data of the compaction indicator optimization model, divide the labeled training features into a training set and a test set, use 70% of the training features as the training set, use 30% of the training features as the test set, use the training set to train the compaction indicator optimization model, use the test set to test the compaction indicator optimization model, preset an error threshold, and when the mean of the prediction errors of all training features in the test set is less than the error threshold, obtain the compaction indicator optimization model.

[0146] Among them, the compaction indicator optimization model adopts any one of the support vector machine model or the random forest model.

[0147] After the compaction index optimization model is trained, the collected pre-data can be input into the compaction index optimization model, and the corresponding numbered value can be optimized. At this time, based on the optimized numbered value, the index set with the same numbered value as the optimized numbered value is screened out from the index set, which is used as the target set, and the compaction index in the target set is used as the compaction index corresponding to the pre-data of subgrade construction.

[0148] The target set analysis module analyzes the optimized indexes from the target set and formulates the evaluation levels of the optimized indexes.

[0149] The optimized indexes refer to all the compaction indexes in the target set optimized by the compaction index optimization model, so that the optimized indexes can be used as the corresponding index data for the user to evaluate and detect the compaction quality of subgrade construction, and provide guidance for subsequent evaluation and detection.

[0150] After obtaining the optimized indexes, it is necessary to give specific analysis opinions on the optimization results of the optimized indexes according to the specific quantity and type of the optimized indexes, so as to provide reasonable and accurate difficulty tips for the subsequent evaluation and collection of the optimized indexes.

[0151] The evaluation levels include easy level, general level and difficult level; the easy level means that the operation difficulty is relatively low when collecting and evaluating the optimized indexes subsequently, the general level means that the operation difficulty is medium when collecting and evaluating the optimized indexes subsequently, and the difficult level means that the operation difficulty is relatively high when collecting and evaluating the optimized indexes subsequently.

[0152] The formulation methods of the easy level, general level and difficult level are as follows:

[0153] By querying the database one by one, the conventional collection duration and standard number of personnel of S optimized indexes are obtained, and based on the S conventional collection durations and S standard numbers of personnel, S difficulty levels are calculated; the conventional collection duration refers to the duration that personnel need to spend when collecting the optimized indexes, which is used to represent the collection difficulty of the optimized indexes from one aspect, and the standard number of personnel refers to the number of personnel to be invested when collecting the optimized indexes, which is used to represent the collection difficulty of the optimized indexes from another aspect.

[0154] The expression of the difficulty level is:

[0155] ;

[0156] In the formula, is the difficulty level of the th optimized index, = 1, 2... S, is the conventional collection duration of the th optimized index, is the The standard number of personnel for a preferred indicator 、 is a weight factor greater than 0, and ;

[0157] Count the number of preferred indicators with a difficulty level greater than the difficulty threshold, denoted as the difficulty quantity value, and record one-fifth and one-half of the number of preferred indicators as the first critical value and the second critical value respectively; the difficulty threshold refers to the minimum value of the difficulty level corresponding to the preferred indicators that are evaluated as difficult to collect, so as to distinguish the specific difficulty levels of different preferred indicators;

[0158] When the difficulty quantity value is less than or equal to the first critical value, it indicates that the number of indicators with a large difficulty level in the collection and evaluation operations among the preferred indicators is small, then a simple level is formulated;

[0159] When the first critical value is less than the difficulty quantity value and the difficulty quantity value is less than or equal to the second critical value, it indicates that the number of indicators with a large difficulty level in the collection and evaluation operations among the preferred indicators is medium, then a general level is formulated;

[0160] When the difficulty quantity value is greater than the second critical value, it indicates that the number of indicators with a large difficulty level in the collection and evaluation operations among the preferred indicators is large, then a difficult level is formulated.

[0161] In this embodiment, by receiving the first original data from the user terminal, screening out the required data from the first original data, obtaining the second original data of the construction terminal, and extracting the environmental data and construction data from the second original data, it is possible to comprehensively represent the user requirements, construction requirements, and environmental conditions in subgrade construction in multiple dimensions. Thus, it can provide comprehensive and diverse data support for the evaluation and analysis of subgrade construction compaction quality, avoid the problem of limitations in the subsequent optimization results of compaction indicators, improve the accuracy of the optimization results of compaction indicators, and group the compaction indicators in the indicator library based on the grouping criteria to generate an indicator set. That is, it can perform indefinite number of classification and combination operations on a large number of different types of compaction indicators, enabling the indicator set to correspond and match any different quantities and different types of required data, environmental data, and construction data, providing a screening basis for the subsequent optimization of compaction indicators. By summarizing the required data, environmental data, and construction data into preposed data, and using the trained compaction indicator optimization model to optimize the target set corresponding to the preposed data, parsing out the optimized indicators from the target set, and formulating the evaluation grades of the optimized indicators, it can not only quickly and accurately optimize the compaction indicators through an artificial intelligence model, avoiding the inaccuracy caused by relying on manual experience and data query methods to optimize compaction indicators, but also represent the subsequent acquisition and evaluation difficulty levels of the optimized compaction indicators. Furthermore, it provides comprehensive, accurate, and intuitive opinion guidance for the subgrade construction compaction indicator optimization operation, improving the optimization effect of subgrade construction compaction indicators.

[0162] Embodiment 2: Please refer to Figure 3 As shown in the figure, the parts not described in detail in this embodiment can be seen in the description of Embodiment 1. A method for optimizing subgrade construction compaction indicators is provided, which is applied to an optimization server and implemented based on a subgrade construction compaction indicator optimization system, including:

[0163] S1: Receive the first original data from the user terminal, and screen out the required data from the first original data. The required data includes calculation accuracy, effective optimization duration, and indicator optimization value;

[0164] S2: Obtain the second original data of the construction terminal, and extract the environmental data and construction data from the second original data;

[0165] S3: Based on the grouping criteria, group the compaction indicators in the indicator library to generate an indicator set;

[0166] S4: Summarize the required data, environmental data, and construction data into preposed data, and use the trained compaction indicator optimization model to optimize the target set corresponding to the preposed data;

[0167] S5: Parse out the preferred indicators from the target set and formulate the evaluation levels for the preferred indicators. The evaluation levels include simple level, general level, and difficult level.

[0168] As described above, it is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention.

Claims

1. A subgrade construction compaction index optimization system, applied to an optimization server, characterized in that Including: A first data acquisition module, configured to receive first raw data from a user terminal and screen out required data from the first raw data, where the required data includes calculation accuracy, effective optimization duration, and index optimization value; The required data refers to the requirements for unconventional personalized index optimization to achieve subgrade construction compaction index optimization within the user terminal; The screening methods for calculation accuracy, effective optimization duration, and index optimization value are as follows: Taking the moment when the user terminal first inputs data as the starting moment and the current moment as the ending moment, the time period between the starting moment and the ending moment is recorded as the required time period; Querying all the data input by the user terminal during the required time period through a database and arranging all the data in the order of input time to obtain the first raw data; Identifying the first semantics of all the first raw data one by one through natural language processing technology and splitting the first semantics into a text part and a digital part through word segmentation technology; Respectively, the first semantics with the text parts being accuracy, duration, and quantity are recorded as accurate semantics, duration semantics, and quantity semantics, and the accurate semantics, duration semantics, and quantity semantics are sorted in the order of input time to obtain a first queue, a second queue, and a third queue; Combining the accurate semantics at the last position in the first queue, the duration semantics at the last position in the second queue, and the quantity semantics at the last position in the third queue with the corresponding digital parts respectively to generate calculation accuracy, effective optimization duration, and index optimization value; A second data acquisition module, configured to obtain second raw data from a construction terminal and extract environmental data and construction data from the second raw data; The environmental data includes construction temperature value, construction humidity value, construction wind speed value, and soil type, and the construction data includes smooth wheel pressure fluctuation value, compaction times, compaction duration, and compaction temperature; An index set generation module, configured to group the compaction indexes in an index library based on a grouping criterion to generate an index set; The generation method of the index set is as follows: Arranging all the compaction indexes in the index library in the order of storage time to generate an index queue; Taking the compaction index at the first position in the index queue as the starting point and the compaction index at the last position in the index queue as the ending point, randomly selecting an indefinite number of compaction indexes from the index queue, and combining the selected compaction indexes to obtain E to-be-verified sets; Counting the number of compaction indexes in each of the E to-be-verified sets one by one and excluding the to-be-verified sets with the number of compaction indexes less than 3 to obtain F to-be-identified sets; Repeatedly identifying the compaction indexes in each of the F to-be-identified sets one by one, excluding the to-be-identified sets with duplicate compaction indexes, and recording the remaining to-be-identified sets as the index set to obtain H index sets; A target set optimization module, configured to summarize the required data, environmental data, and construction data into pre-data and optimize a target set corresponding to the pre-data through a trained compaction index optimization model; A target set analysis module, configured to analyze out optimized indexes from the target set and formulate an evaluation level for the optimized indexes, where the evaluation level includes an easy level, a general level, and a difficult level; The methods for formulating the easy level, general level, and difficult level are as follows: Query the conventional acquisition durations and standard personnel numbers of S preferred indicators one by one through the database, and calculate S difficulty levels based on the S conventional acquisition durations and S standard personnel numbers; Count the number of preferred indicators with difficulty levels greater than the difficulty threshold, record it as the difficulty quantity value, and record one-fifth and one-half of the number of preferred indicators as the first critical value and the second critical value respectively; When the difficulty quantity value is less than or equal to the first critical value, formulate the easy level; When the first critical value is less than the difficulty quantity value and the difficulty quantity value is less than or equal to the second critical value, formulate the general level; When the difficulty quantity value is greater than the second critical value, formulate the difficult level.

2. The optimized system for subgrade construction compaction indicators according to claim 1, characterized in that The extraction methods for the construction temperature value, construction humidity value, construction wind speed value, and soil type are as follows: Query the construction period of subgrade construction through the database, and divide the construction period into A consecutive sub-periods with equal durations; The temperature values, humidity values, and wind speed values at detection moments in A sub-periods are respectively detected by a temperature sensor, a humidity sensor, and a wind speed sensor, obtaining real-time temperature values, real-time humidity values, and real-time wind speed values; ​​​​ Exclude the maximum and minimum values of the real-time temperature value, real-time humidity value, and real-time wind speed value respectively, and for the remaining real-time temperature values, real-time humidity values, and real-time wind speed values, accumulate them one by one and then calculate the average to obtain A sub-temperature values, A sub-humidity values, and A sub-wind speed values; Respectively add the maximum and minimum values of the sub-temperature value, sub-humidity value, and sub-wind speed value and take the average to obtain the construction temperature value, construction humidity value, and construction wind speed value; Query the soil data of subgrade construction through the database, identify the keywords of the soil data through natural language processing technology, and record the keywords as the soil type; The soil type includes clay type, sandy soil type, silt type, and gravel type.

3. The optimized system for subgrade construction compaction indicators according to claim 2, characterized in that The extraction method for the smooth wheel pressure fluctuation value is as follows: Query the moments when the roller performs smooth wheel compaction construction for the first time and the last time during the construction period through the database, and record them as the compaction start moment and the compaction end moment respectively; During the period from the compaction start moment to the compaction end moment, detect the smooth wheel pressure of the roller in real time through a pressure sensor, and record the moment when the smooth wheel pressure is greater than or equal to the preset pressure threshold for the first time as the effective start moment; During the period from the effective start moment to the compaction end moment, record the moment when the smooth wheel pressure is less than the preset pressure threshold for the last time as the effective end moment, and record the period between the effective start moment and the effective end moment as the compaction period; During the compaction period, randomly mark non - adjacent query times, and query the wheel pressures at the query times to obtain query pressures; After successively subtracting the query pressure at the next query moment from the query pressure at the previous query moment, obtain pressure differences; Accumulate the pressure differences and then calculate the average to obtain the value of the vibratory roller pressure fluctuation.

4. The preferred system for roadbed construction compaction index according to claim 3, wherein The training method for the compaction index optimization model is as follows: Pre-collect multiple groups of pre-data of subgrade construction and the corresponding index sets, and perform character conversion on the soil type in turn, convert the clay type to NT, the sandy soil type to ST, the silt type to FT, and the gravel type to SS; Mark each group of pre-data as training features, number the index sets of each group of training features, and record the numbered values of the index sets; Use the training features as the input data of the compaction index optimization model, use the numbered values as the output data of the compaction index optimization model, divide the labeled training features into a training set and a test set, use the training set to train the compaction index optimization model, use the test set to test the compaction index optimization model, preset an error threshold, and when the mean of the prediction errors of all training features in the test set is less than the error threshold, obtain the compaction index optimization model.

5. The optimized system for subgrade construction compaction index according to claim 4, characterized in that, Output the optimized serial number value through the compaction index optimization model, and denote the index set with the same serial number value as the optimized serial number value in the index set as the target set.

6. A method for optimizing subgrade construction compaction indicators, implemented based on the subgrade construction compaction indicator optimization system described in any one of claims 1-5, characterized in that, Including: S1: Receive the first original data from the user side, and screen out the required data from the first original data. The required data includes calculation accuracy, effective optimization duration, and index optimization value; S2: Obtain the second original data from the construction side, and extract the environmental data and construction data from the second original data; S3: Group the compaction indexes in the index library based on the grouping criteria to generate an index set; S4: Aggregate the required data, environmental data, and construction data into pre-data, and optimize the target set corresponding to the pre-data through the trained compaction index optimization model; S5: Parse out the optimized indexes from the target set, and formulate the evaluation levels of the optimized indexes. The evaluation levels include easy level, general level, and difficult level.

Images