Engineering Planning and Design Data Processing and Management System Based on Multidimensional Data

Abstract

The present invention discloses an engineering planning and design data processing and management system based on multidimensional data, which belongs to the technical field of engineering planning and design data processing, and comprises a data source management module, a demand module and a design management module; the data source management module is used for determining each target source corresponding to the engineering planning and design; the demand module is used for data collection according to the design requirements, and data collection is performed in each target source according to the design requirements of the user, and the collected data is obtained, and the collected data is processed to obtain target auxiliary data; the design management module is used for managing the user's design completion data, obtaining the user's design completion data, determining the completion target according to the design completion data, and collecting reference data from each target source according to the completion target; evaluating the design completion data through each reference data, and obtaining the corresponding reference value; integrating each parameter data and the reference value into verification recommendation data; and recommending the verification recommendation data to the user.

Description

Technical Field

[0001] The present invention belongs to the technical field of engineering planning and design data processing, and in particular is an engineering planning and design data processing and management system based on multidimensional data. Background Art

[0002] With the rapid development of information technology and the continuous deepening of engineering planning and design, traditional data processing methods can no longer meet the complex and ever-changing needs of engineering planning. At present, the amount of data involved in the engineering planning and design process is huge, the types are diverse, and the relationships between the data are complex, which makes data processing a key factor restricting the efficiency and accuracy of engineering planning and design.

[0003] There are many deficiencies in the application of existing data processing systems in the field of engineering planning and design. First, most systems only focus on data processing in a single or a small number of dimensions, lack the ability to fully integrate and analyze multidimensional data, and find it difficult to fully tap the potential value of data. Second, existing systems often lack intelligent and automated support in the data processing process, resulting in low data processing efficiency and prone to errors and omissions. In addition, data interaction and sharing between systems also face many challenges. Data islands are common, making it difficult to achieve effective use of data and collaborative work.

[0004] Based on this, the present invention provides an engineering planning and design data processing and management system based on multidimensional data. Summary of the invention

[0005] In order to solve the problems existing in the above solutions, the present invention provides an engineering planning and design data processing and management system based on multi-dimensional data.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] Engineering planning and design data processing and management system based on multi-dimensional data, including data source management module, demand module and design management module;

[0008] The data source management module is used to determine each target source corresponding to the engineering planning and design.

[0009] Furthermore, the method for determining the target source includes:

[0010] Determine the design goal according to the engineering planning and design, and collect the corresponding engineering description data based on the design goal; set the engineering planning process according to the engineering description data; determine the corresponding demand data according to the engineering description data and the engineering planning process;

[0011] Determine each candidate data source according to the demand data, screen each candidate data source to obtain each target source; and establish a data channel with each target source.

[0012] Furthermore, the method for obtaining demand data includes:

[0013] Obtain historical planning data, identify the engineering planning process corresponding to each historical planning data, and set the historical demand data corresponding to each engineering planning process according to the historical planning data; classify each engineering planning process according to each historical demand data, and obtain each process classification and the historical demand data corresponding to each process classification;

[0014] Establish a demand matching library based on each process classification and historical demand data;

[0015] Match the corresponding demand data from the demand matching library according to the engineering planning process and engineering description data.

[0016] Further, the methods for classifying each engineering planning process include:

[0017] Step SA1: Identify the same engineering planning processes and integrate the same engineering planning processes into a first category; identify the historical demand data corresponding to the engineering planning processes in the first category, integrate the engineering planning processes with the same historical demand data into a second category, and divide the first category into multiple second categories;

[0018] Step SA2: Determine the reasons for the differences in historical demand data between the second categories, mark them as difference reasons, such as environmental differences, enterprise differences, personnel differences, etc., and perform an equivalence analysis on the difference reasons, that is, determine whether the difference reason will appear in another second category. If it will appear, then the reasons are determined to be equivalent; otherwise, the reasons are determined to be different; obtain the corresponding equivalence analysis results. When the equivalence analysis result is a difference in reasons, do not perform the corresponding operation; when the equivalence analysis result is that the reasons are equivalent, merge the corresponding two second categories to obtain a new second category, and merge the historical demands corresponding to the two second categories, which is equivalent to a union;

[0019] Step SA3: Loop step SA2 until the second classifications cannot be merged; mark each second classification as a process classification.

[0020] Furthermore, the method for screening each candidate data source includes:

[0021] Evaluate each candidate data source to obtain the corresponding true value and timeliness value; mark the obtained true value and timeliness value as SA and HA respectively;

[0022] Calculate the corresponding screening value according to the formula PA = b1 × SA + b2 × HA;

[0023] Where: PA is the screening value; b1 and b2 are both proportional coefficients, with a value range of 0 <b1≤1,0<b2≤1;

[0024] Each candidate data source whose screening value is greater than the threshold value X1 is marked as a target source.

[0025] The demand module is used to collect data according to design requirements to obtain user design requirements; collect data in each target source according to the design requirements to obtain corresponding collected data, process the collected data, and obtain target auxiliary data corresponding to the design requirements.

[0026] Furthermore, the method for processing the collected data includes:

[0027] De-duplicate the collected data to obtain initial data; perform feature recognition on each initial data to obtain each feature item data corresponding to the initial data, and integrate each feature item data into feature group data;

[0028] Each feature group data is equivalently classified to obtain several initial classifications, and representative data corresponding to each initial classification is set; conflicts are identified based on each representative data to obtain each conflict classification data; each conflict analysis data is verified to obtain the real data corresponding to the conflict analysis data, and each real data and each remaining representative data are integrated into target auxiliary data.

[0029] Furthermore, the method for verifying each conflict analysis data includes:

[0030] Obtain the verification points corresponding to each conflict analysis data and the weight coefficients corresponding to each verification point, and establish corresponding verification models according to each verification point;

[0031] The expression of the calibration model is

[0032] Where: i represents the corresponding check point, i = 1, 2, ..., n, n is a positive integer; si represents the input data corresponding to the corresponding check point; the output data is the check point value 1, 0 or -1;

[0033] Based on the verification model, each pair of representative data in the conflict analysis data is analyzed to determine the real data.

[0034] Furthermore, the method for analyzing each pair of representative data in the conflict analysis data based on the verification model includes:

[0035] According to the formula Calculate the corresponding verification evaluation value;

[0036] Where: RP is the calibration evaluation value; δi represents the weight coefficient of the corresponding calibration point; HP(si) is the calibration model;

[0037] The representative data with the verification evaluation value lower than the threshold value X2 are eliminated, and the remaining representative data are compared again in pairs; and so on, until the real data is obtained.

[0038] The design management module is used to manage the design completion data of users, obtain the design completion data of users, determine the completion target according to the design completion data, and collect reference data from each target source according to the completion target; evaluate the design completion data through each reference data to obtain the corresponding reference value; integrate each parameter data and the reference value into the verification recommendation data; and recommend the verification recommendation data to the user.

[0039] Further, the calculation method of the reference value includes:

[0040] Compare and analyze the design completion data with each reference data to obtain the project quantity change value, cost change value, and construction period change value corresponding to each reference data;

[0041] Calculate the corresponding reference value according to the formula CQ = b3×GL + b4×BL + b5×QL;

[0042] In the formula: CQ is the reference value; b3, b4, and b5 are all proportionality coefficients, and the value range is 0 < b3 ≤ 1, 0 < b4 ≤ 1, 0 < b5 ≤ 1; GL is the project quantity change value; BL is the cost change value; QL is the construction period change value.

[0043] Compared with the prior art, the beneficial effects of the present invention are:

[0044] Through the mutual cooperation among the data source management module, the requirement module, and the design management module, the intelligent processing of engineering planning and design data is realized; through the mutual cooperation between the data source management module and the requirement module, the intelligent acquisition of multi-dimensional data according to the design requirements is realized, and the multi-dimensional data is processed and verified to ensure the authenticity of the data; and effective assistance is provided for the engineering planning and design of users. By setting the data source management module, each target source is determined according to the actual requirements of engineering planning and design, realizing the multi-dimensional data collection in engineering planning and design, breaking the data island, providing data support for engineering planning and design, and improving the design efficiency; at the same time, screening each candidate data source during the process of determining each target source to ensure the authenticity and timeliness of the data of each target source. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0046] Figure 1 It is the principle block diagram of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0047] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] like Figure 1 As shown, the engineering planning and design data processing and management system based on multidimensional data includes a data source management module, a demand module and a design management module;

[0049] The data source management module is used to determine the target sources corresponding to the engineering planning and design, and the target sources are the selected data sources; the detailed process is as follows:

[0050] Set corresponding design goals according to the requirements of engineering planning and design, such as what kind of engineering to design; obtain engineering description data of the target engineering, which includes relevant data such as the region where the engineering is located and the surrounding environment, and is used to determine what data may be needed for the engineering planning and design based on the engineering description data, such as local rules and regulations, equipment, manpower, geology and other related conditions; set the engineering planning process and the corresponding demand data for each process node according to the engineering description data;

[0051] Determine each candidate data source according to the demand data, the candidate data source is the data source that can obtain the corresponding demand data; screen each candidate data source to obtain each target source; establish a data channel with each target source for subsequent access to each target source.

[0052] For the engineering planning process, the corresponding engineering planning process can be directly determined based on the engineering description data and the existing engineering planning process; it can also be directly set by the staff.

[0053] For demand data, a large amount of historical engineering planning and design data is obtained and marked as historical planning data; the corresponding engineering planning process is marked for each historical planning data, and the corresponding historical demand data is marked at each process node in the engineering planning process according to the corresponding required data recorded in the historical planning data, that is, the data required for each process node is identified according to the corresponding historical data, and marked as historical demand data; each engineering planning process is classified to obtain the classification of each process, that is, the same engineering planning process is first classified to identify whether the historical demand data corresponding to the same engineering planning process are the same. If not, determine the reason why the historical demand data is different. If the reason does not appear in another engineering planning process, it is reclassified into one category. Otherwise, it is regarded as one category, and the historical demand data are integrated; and so on to complete the classification of each engineering planning process.

[0054] Identify the historical demand data corresponding to each process classification, and establish a corresponding demand matching library based on each process classification and historical demand data;

[0055] According to the current engineering planning process and engineering description data, the corresponding process classification is matched from the demand matching library, and the historical demand data of the corresponding process classification is marked as demand data.

[0056] Each candidate data source is screened. In one embodiment, each candidate data source can be screened based on the existing technology. In another embodiment, the following method can be used for screening. Each candidate data source is evaluated to obtain the corresponding true value and timeliness value. The value range of the true value and the timeliness value is [1, 10]. The judgment criteria of the true value and the timeliness value are preset. For example, for the true value, the possible data sources are counted and sorted in order of credibility from high to low. The highest and lowest correspond to 10 and 0 respectively. The others are set accordingly according to the difference. Subsequently, the corresponding similarity matching is performed. Set it according to the closest case; for the timeliness value, obtain the possible timeliness range, that is, the time interval between the data of the data source and the corresponding data of the latest version; divide different timeliness intervals, and set corresponding timeliness values ​​for each timeliness interval; all are the same as the latest version, and the timeliness value is 10; specifically, a corresponding data source analysis model can be established based on a CNN network or a DNN network, and a corresponding training set can be established manually for training. The training set includes input data and output data, the input data is data source related data, and the output data is the true value and timeliness value; analysis is performed through the data source analysis model after successful training;

[0057] The obtained true value and time-effective value are marked as SA and HA respectively;

[0058] Calculate the corresponding screening value according to the formula PA = b1 × SA + b2 × HA;

[0059] Where: PA is the screening value; b1 and b2 are both proportional coefficients, with a value range of 0 <b1≤1,0<b2≤1;

[0060] Each candidate data source whose screening value is greater than the threshold value X1 is marked as a target source.

[0061] By setting up the data source management module, each target source is determined according to the actual needs of engineering planning and design, so as to realize multi-dimensional data collection in engineering planning and design, break the data silos, provide data support for engineering planning and design, and improve design efficiency; at the same time, in the process of determining each target source, the candidate data sources are screened to ensure the authenticity and timeliness of each target source data.

[0062] The demand module is used to collect data according to design requirements and obtain user design requirements, such as specifications, cases, possible design problems and other related design requirements of a certain design content; collect data from each target source according to the obtained design requirements, obtain corresponding collected data, process the collected data, and obtain target auxiliary data corresponding to the design requirements.

[0063] The collected data is processed, that is, duplicates are first removed, conflicting data are identified, each conflicting data is evaluated, conflicts that do not meet the evaluation requirements are eliminated, each real data is obtained, the real data is screened, and auxiliary data is obtained; the detailed process is as follows:

[0064] De-duplicate the collected data to obtain initial data; preset feature templates, which include feature identification items corresponding to various data types. For example, for design planning, it corresponds to a group of feature identification items, and then data identification is performed according to the group of feature identification items, such as version, implementation standard, name and other related data; the specific settings are summarized by professionals; feature identification is performed on each initial data through the preset feature template to obtain each feature item data corresponding to the initial data, and each feature item data is integrated into feature group data; in other embodiments, a corresponding feature recognition model can also be established in combination with existing technologies, and intelligent recognition is performed through the feature recognition model;

[0065] Perform equivalent classification on each feature group data to obtain several initial classifications, such as feature group data of A includes feature item data of 1, 2, 3, and 4; feature group data of B includes feature item data of 1, 3, and 4; feature group data of C includes feature item data of 1, 2, 4, and 5; among which, each feature item data of B is included in A, so AB is an initial classification. Since C has 5 which is not in AB, it does not belong to the initial classification corresponding to AB. At the same time, if feature group data of D includes feature item data of 1 and 4; then D belongs to the initial classification corresponding to AB and C respectively; that is, perform equivalent classification according to the inclusion relationship between each feature group data; screen each initial classification, and select the most complete feature group data as the representative data of the initial classification; such as A in the AB initial classification;

[0066] Conflict identification is performed based on each representative data to obtain each conflict classification data; conflict identification is to identify whether the corresponding feature item data in the representative data conflict, such as the same specification, but some of its feature item data are different and conflict with each other, then it is identified as conflict data to form conflict classification data; identification is performed based on existing data analysis technology;

[0067] Each conflict analysis data is checked to obtain the real data corresponding to the conflict analysis data, and each real data and each remaining representative data are integrated into target auxiliary data.

[0068] Methods for checking various conflict analysis data include:

[0069] Preset the verification points corresponding to different types of data and the weight coefficients corresponding to each verification point, such as data source comparison, data release time comparison, and the number of corresponding data; the weight coefficients corresponding to different verification points are different. The greater the impact on authenticity judgment, the greater the weight coefficient, such as source is greater than release time; specifically, according to the needs of verifying the type of data, the verification points and weight coefficients are set manually;

[0070] Get the verification points corresponding to each conflict analysis data, and establish the corresponding verification model according to each verification point. The verification model is used to compare the data corresponding to each verification point to determine which one is more real, such as the file release time, the closer to the current time, the more real; the data source, the higher the credibility, the more real; specifically, it is trained based on a large amount of historical conflict data, and the expression is Where: i represents the corresponding check point, i=1, 2, ..., n, n is a positive integer; si represents the input data corresponding to the corresponding check point, that is, the combined data of the two conflicting data at the check point, which is used to compare which one is more real; if A is compared with B, si is real, which means that AB is more real at the check point; there are also many cases with the same or similar properties, which are considered equal; the same applies to others; the output data is the check point value 1, 0 or -1;

[0071] Based on the verification model, each pair of representative data in the conflict analysis data is analyzed to determine the real data.

[0072] The method for analyzing each pair of representative data in the conflict analysis data based on the verification model includes:

[0073] According to the formula Calculate the corresponding calibration evaluation value; where: RP is the calibration evaluation value; δi represents the weight coefficient of the corresponding calibration point; HP(si) is the calibration model;

[0074] Eliminate the data with verification evaluation values ​​lower than the threshold value X2, and compare the remaining representative data again in pairs; and so on, until the real data is obtained, that is, the corresponding representative data without conflict.

[0075] Through the cooperation between the data source management module and the demand module, it is possible to intelligently acquire multidimensional data according to design requirements, and process and verify the multidimensional data to ensure the authenticity of the data, providing effective assistance for users' engineering planning and design.

[0076] The design management module is used to manage the design completion data of users, obtain the design completion data of users, determine the completion goal according to the design completion data, that is, what standard content is designed by the design completion data; collect reference data from each target source according to the completion goal, and the reference data is the design scheme that can achieve the completion goal. Retrieve according to the design cases in each target source. Existing retrieval technologies can be used for retrieval, or intelligent models can be established by combining technologies such as neural networks to intelligently analyze whether it can be used as reference data;

[0077] Evaluate the design completion data through each reference data to obtain the corresponding reference value; integrate each parameter data and the reference value into the verification and recommendation data; recommend the verification and recommendation data to the user.

[0078] The method for evaluating the design completion data through the reference data includes:

[0079] Mainly conduct comparative evaluations in three directions: engineering quantity, cost, and construction period. That is, estimate the engineering planning according to the reference data method and compare it with the user's design, and analyze the change values of the engineering quantity, cost, and construction period. For example, if the engineering quantity decreases by U, the change value is U divided by the corresponding engineering quantity of the user's design. A decrease is a positive number, and an increase is a negative number. A negative sign will be added before the calculated value; the same applies to the other two; the specific estimation method can be estimated according to the existing technology. Therefore, it is used as a reference for verification. Therefore, there can be a relatively large estimation error; Exemplarily, establish a corresponding estimation model based on the CNN network or DNN network, and establish a corresponding training set through artificial means for training. The training set includes input data and output data. The input data is the design completion data and reference data set according to the historical engineering planning data; the output data is the corresponding change values of the engineering quantity, cost, and construction period;

[0080] Calculate the corresponding reference value according to the formula CQ = b3×GL + b4×BL + b5×QL;

[0081] In the formula: CQ is the reference value; b3, b4, and b5 are all proportionality coefficients, and the value range is 0 < b3 ≤ 1, 0 < b4 ≤ 1, 0 < b5 ≤ 1; GL is the change value of the engineering quantity; BL is the change value of the cost; QL is the change value of the construction period.

[0082] The above formulas are all calculated by removing the dimension and taking their numerical values. The formula is obtained by collecting a large amount of data for software simulation to obtain a formula closest to the actual situation. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through a large amount of data simulation.

[0083] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. Engineering planning and design data processing and management system based on multidimensional data, characterized by: Includes data source management module, requirement module and design management module; The data source management module is used to determine each target source corresponding to the engineering planning and design; The demand module is used to collect data according to the design requirements to obtain the user's design requirements; collect data in each target source according to the design requirements to obtain corresponding collected data, and process the collected data to obtain target auxiliary data corresponding to the design requirements; The design management module is used to manage the user's design completion data, obtain the user's design completion data, determine the completion target according to the design completion data, collect reference data from various target sources according to the completion target; evaluate the design completion data through various reference data to obtain corresponding reference values; Integrate each parameter data and reference value into calibration recommended data; Recommend verification recommendation data to users; Methods for determining target sources include: Determine the design goal according to the engineering planning and design, and collect the corresponding engineering description data based on the design goal; set the engineering planning process according to the engineering description data; determine the corresponding demand data according to the engineering description data and the engineering planning process; Determine each candidate data source according to the demand data, screen each candidate data source, and obtain each target source; establish a data channel with each target source; Methods for obtaining demand data include: Obtain historical planning data, identify the engineering planning process corresponding to each historical planning data, and set the historical demand data corresponding to each engineering planning process according to the historical planning data; classify each engineering planning process according to each historical demand data, and obtain each process classification and the historical demand data corresponding to each process classification; Establish a demand matching library based on each process classification and historical demand data; Match the corresponding demand data from the demand matching library according to the engineering planning process and engineering description data; Methods for categorizing the various engineering planning processes include: Step SA1: Identify the same engineering planning processes and integrate the same engineering planning processes into a first category; identify the historical demand data corresponding to the engineering planning processes in the first category and integrate the engineering planning processes with the same historical demand data into a second category; Step SA2: determine the reasons for the differences in historical demand data between the second categories, mark them as difference reasons, perform an equivalent analysis on the difference reasons, and obtain corresponding equivalent analysis results. When the equivalent analysis result is a difference in reasons, no corresponding operation is performed; when the equivalent analysis result is an equivalent reason, merge the corresponding two second categories to obtain a new second category, and merge the historical demands corresponding to the two second categories; Step SA3: Loop step SA2 until the second classifications cannot be merged; mark each second classification as a process classification.

2. The engineering planning and design data processing and management system based on multidimensional data according to claim 1 is characterized in that: Methods for screening candidate data sources include: Evaluate each candidate data source to obtain the corresponding true value and timeliness value; mark the obtained true value and timeliness value as SA and HA respectively; Calculate the corresponding screening value according to the formula PA=b1×SA+b2×HA; Where: PA is the screening value; b1 and b2 are both proportionality coefficients, and the value ranges are 0 < b1 ≤ 1, 0 < b2 ≤ 1; Mark each candidate data source with a screening value greater than the threshold X1 as the target source.

3. The engineering planning and design data processing and management system based on multidimensional data according to claim 1, characterized in that: The method for processing the collected data includes: Deduplicate the collected data to obtain the initial data; perform feature recognition on each initial data to obtain the corresponding feature item data of each initial data, and integrate the feature item data into the feature group data; Perform equivalent classification on each feature group data to obtain several initial classifications, set the representative data corresponding to each initial classification; perform conflict recognition based on each representative data to obtain each conflict classification data; check each conflict analysis data to obtain the real data corresponding to the conflict analysis data, and integrate each real data with the remaining representative data into the target auxiliary data.

4. The engineering planning and design data processing and management system based on multidimensional data according to claim 3 is characterized in that: The method for checking each conflict analysis data includes: Obtain the check points corresponding to each conflict analysis data and the weight coefficients corresponding to each check point, and establish a corresponding check model based on each check point; The expression of the calibration model is ; Where: i represents the corresponding check point, i = 1, 2,..., n, n is a positive integer; si represents the input data corresponding to the corresponding check point; the output data is the check point value 1, 0 or -1; Analyze each pair of representative data in the conflict analysis data based on the check model to determine the real data.

5. The engineering planning and design data processing and management system based on multidimensional data according to claim 4 is characterized in that: According to the formula Calculate the corresponding verification evaluation value; Where: RP is the check evaluation value; δi represents the weight coefficient of the corresponding check point; HP(si) is the check model; Eliminate the representative data with a check evaluation value lower than the threshold X2, and compare the remaining representative data pairwise again; and so on until the real data is obtained.

6. The engineering planning and design data processing and management system based on multidimensional data according to claim 1, characterized in that: The calculation method of the reference value includes: Compare and analyze the designed completed data with each reference data to obtain the engineering quantity change value, cost change value, and construction period change value corresponding to each reference data; Calculate the corresponding reference value according to the formula CQ = b3 × GL + b4 × BL + b5 × QL; Where: CQ is the reference value; b3, b4, and b5 are all proportionality coefficients, and the value ranges are 0 < b3 ≤ 1, 0 < b4 ≤ 1, 0 < b5 ≤ 1; GL is the engineering quantity change value; BL is the cost change value; QL is the construction period change value.

Images