Intelligent monitoring method and system for full-process progress of agricultural machinery part production

Abstract

The invention relates to the technical field of agricultural machinery part production full-process progress intelligent monitoring, in particular to an agricultural machinery part production full-process progress intelligent monitoring method and system. Comprising the steps that firstly, real-time production information is collected through a data input layer, and then data preprocessing, clustering and screening are carried out to finally enter a production optimization layer; the key point is that a progress-quality influence matrix is introduced, the matrix evaluates quality and progress errors in each production step based on a historical data set, and the influence of different production parameters on product quality and progress is analyzed. The method comprises the following steps: calculating a parameter error vector, combining the parameter error vector with historical data, constructing an extended quality analysis data set, predicting subsequent production steps after optimizing a current production step, analyzing historical data of a qualified product, constructing a production step incidence matrix, calculating the influence degree between the steps, and generating an influence vector, thereby obtaining the quality of the product. And the production parameters of the subsequent steps are adjusted by using the method, so that the production quality and progress are guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent monitoring of the progress of the entire production process of agricultural machinery parts, and specifically to an intelligent monitoring method and system for the progress of the entire production process of agricultural machinery parts. Background Art

[0002] Agricultural machinery parts refer to various parts used in agricultural machinery, which are mainly used for the repair, maintenance and update of agricultural machinery equipment in the agricultural production process. As an important part of agricultural machinery, agricultural machinery parts directly affect the operating efficiency and service life of agricultural machinery. Different types of agricultural machinery parts have different functions, covering areas including planting, tillage, harvesting, irrigation and transportation and other agricultural production steps.

[0003] As a high-precision device, agricultural machinery parts have extremely high requirements for precision, quality and reliability in their production process. In order to ensure the production of high-precision agricultural machinery parts that meet the standards, strict quality control and precision manufacturing are required in multiple links. At present, the production of agricultural machinery parts has been realized through automation. In this process, the equipment parameters used in each production step must be kept highly accurate; if there is a deviation in any link, the final agricultural machinery parts produced will not be able to be used normally.

[0004] The whole process of agricultural machinery parts production involves multiple links, and each production step has a close interdependence. The output of the previous link not only directly affects the production quality of the next link, but may also have a profound impact on the overall production schedule. If this interrelationship between the production steps is ignored and the parameters and processes of each link are not properly regulated, it may lead to delayed production schedules, quality fluctuations, and even defective products, thus affecting the efficiency and stability of the entire production process. Therefore, the parameters of each production step must be accurately controlled and optimized to ensure the efficient operation of the entire process and the continuous improvement of product quality.

[0005] To this end, the present invention proposes a method and system for intelligently monitoring the progress of the entire production process of agricultural machinery parts. Summary of the invention

[0006] The purpose of the present invention is to provide a method and system for intelligently monitoring the progress of the entire production process of agricultural machinery parts, which is used for intelligently optimizing production parameters in the production process of agricultural machinery parts to ensure that the production progress and production quality meet the expected standards; the present invention mainly starts from the following points: intelligent monitoring of the progress of the entire production process of agricultural machinery parts. During the monitoring process, the production parameters of each monitoring point are optimized by collecting real-time production information of the monitoring point and combining the agricultural machinery parts production monitoring optimization model. The production optimization model includes: a data input layer, a data preprocessing layer, a data clustering layer, a data screening layer, a production optimization layer and a parameter output layer. In the production optimization layer, the parameters of the current production step are optimized by analyzing the historical data of the same production step.

[0007] In addition, a progress-quality impact matrix is ​​introduced into the optimization model to assist in the optimization of production parameters. The calculation of this matrix evaluates the quality errors of agricultural machinery parts in different production steps by extending the historical data set, and further analyzes the impact of various production parameters on the final production quality. The core of the method is to calculate the parameter error vector of agricultural machinery parts and merge it into the historical data to construct an extended quality analysis data set, thereby realizing production progress and qualification analysis.

[0008] After optimizing the current production parameters, the production parameters of the subsequent production steps are predicted. The specific steps include: by analyzing the historical data of qualified products, constructing the production step association matrix, normalizing it, and calculating the influence between each production step. Based on the normalized matrix, the production step influence vector is generated to optimize the current production step and achieve the desired production results. Finally, the production parameters of the subsequent production steps are adjusted according to the desired results and the influence vector.

[0009] To achieve the above object, the present invention provides the following technical solutions: A method for intelligently monitoring the progress of the entire production process of agricultural machinery parts, comprising: Collect current production information of the monitoring point, including: current production steps, current equipment operating parameters, data of materials to be processed, current expected production progress, and current production environment data; Further, a plurality of historical data of agricultural machinery parts are obtained; wherein the historical data include: historical raw material data, historical equipment operation parameters, historical production environment data, historical production output data, historical actual production progress, and historical expected production progress; the historical production output data refers to the production result of agricultural machinery parts when each production step is completed; each production step includes: the historical actual production progress and the historical expected production progress; Furthermore, the historical data is processed to obtain standard historical data; wherein the processing of the historical data includes: performing data cleaning on the historical data to obtain first historical data; performing smoothing on the first historical data to obtain second historical data; performing formatting on the second historical data to obtain third historical data; and performing normalization on the third historical data to obtain the standard historical data.

[0010] Further, a plurality of said standard historical data are annotated and a historical data set is constructed; Further, the historical data set is analyzed to obtain a progress-quality impact matrix for each production step; The calculation process of the progress-quality impact matrix includes: Calculating the error between the agricultural machinery quality inspection data and the standard quality in the historical data set to obtain a quality error vector; Furthermore, the quality error vector is merged into the historical data set according to the historical data sequence to obtain the extended historical data set; the extended historical data set is expressed as: ; in, is the extended historical data set; for Extended quality analysis history data for each production plan; For the A standard historical raw material data vector for a production plan; For the Production steps in a production plan The standard historical equipment operating parameter vector; For the Production steps in a production plan Standard historical production environment data vector; For the Production steps in a production plan The completed standard historical production output data vector; For the Production steps in a production plan The standard historical actual production progress value; For the Production steps in a production plan The standard historical expected production progress value; For the The quality error vector in the production plan; For the The quality evaluation value of agricultural machinery parts in a production plan; the quality evaluation value of agricultural machinery parts is 0 or 1, 0 means qualified quality; 1 means unqualified quality; is the total number of production plans; is the total number of production steps; Furthermore, the extended historical data set is input into the progress-qualification analysis function to obtain the progress-quality impact matrix of each production step; wherein the progress-quality impact matrix is ​​expressed as: ;in, For production steps The progress-quality impact matrix; is the progress-qualification analysis function.

[0011] Further, the progress-quality impact matrix corresponding to the current production step is obtained as a first coefficient matrix; Further, the current production information, the historical data set and the first coefficient matrix are input into an agricultural machinery parts production monitoring optimization model to obtain current production step optimization parameters; The agricultural machinery parts production monitoring optimization model includes: a data input layer, used to receive the current production information, the historical data set and the first coefficient matrix; a data preprocessing layer, used to preprocess the production information to obtain the current production data vector; a data clustering layer, used to cluster the current production data vector with the data in the historical data set to obtain a historical data cluster set; a data screening layer, used to extract the historical data of the same production step as the current production step from the historical data cluster set to obtain a historical data cluster subset; The production optimization layer is used to optimize the production parameters of the current production step; Wherein, the production optimization layer comprises the following steps: Step 1: Obtain the current production data vector and the historical data cluster subset; Step 2: Calculate the error between the current production data vector and each data in the historical data cluster subset, and construct a current production error matrix in combination with the quality annotation content; Step 3: Calculate the forward discreteness and reverse discreteness in the current production error matrix; wherein the forward discreteness refers to the error data in the current production error matrix that is completed according to the expected progress and the quality is marked as qualified; the reverse discreteness refers to the error data in the current production error matrix that is not completed according to the expected progress and the quality is marked as unqualified; The calculation formula of the forward dispersion is: ;in, is the number of data that meets the positive dispersion condition, that is And ; For the The actual progress of historical data; For the Expected progress of historical data; For the The historical data quality is marked, 0 means qualified and 1 means unqualified; The index set of all data that meets the positive condition; is the first cluster in the historical data cluster subset that meets the positive condition. The error of the data vector; The calculation formula of the forward dispersion is: ;in, is the number of data that meets the inverse dispersion condition, that is And ; The index set of all data that meets the reverse condition; is the first cluster in the historical data cluster subset that meets the reverse condition. The error of the data vector; Step 4: Optimize the production parameters in the current production data vector using a first coefficient matrix according to the forward discreteness and the reverse discreteness; Step 5: Using the optimized current production data vector as the current production data vector; Step 6: Repeat step 2 to step 5 until the conditions are met to obtain the optimization parameters of the current production step.

[0012] A parameter output layer, used to output the optimization parameters of the current production step; Furthermore, the current production step optimization parameters are used to produce the material to be processed, and the production parameters of the next production step are generated; the specific process includes: Acquire the data of agricultural machinery parts that have passed the quality inspection in the historical data set, and construct a historical data set of qualified products; Further, a correlation analysis is performed on each production step in the qualified product historical data set to obtain a production step correlation matrix; Furthermore, the values ​​in the production step association matrix are normalized to obtain a normalized association matrix; wherein the normalized association matrix is ​​expressed as: ;in, is the normalized correlation matrix; For production steps Production steps The normalized correlation of For production steps Production steps degree of relevance; Furthermore, the production step influence vector on the next production step is calculated according to the normalized correlation matrix; the calculation formula of the production step influence vector is: ;in, For production steps The production step influence vector; For production steps Optimization parameters of the current production step; Further, the production is performed according to the optimized parameters of the current production step to obtain the desired production result; Furthermore, the production parameters of the next production step are obtained according to the expected production result, the production step impact vector and the next production target.

[0013] An intelligent monitoring system for the progress of the entire production process of agricultural machinery parts, the system comprising: Data collection unit, used to collect various data in the whole process of agricultural machinery parts production; A data storage unit, used for storing data; Data processing unit, used to process data; intelligent monitoring unit, intelligent monitoring of the progress of the entire production process of agricultural machinery parts; Among them, the data processing unit includes: acquiring multiple historical data stored in the data storage unit; performing data cleaning on the historical data to obtain first historical data; performing smoothing processing on the first historical data to obtain second historical data; performing formatting processing on the second historical data to obtain third historical data; and performing normalization processing on the third historical data to obtain standard historical data.

[0014] Intelligent analysis unit, used to analyze the progress and quality of the entire process of agricultural machinery parts production; Wherein, the intelligent analysis unit includes: Calculating the error between the agricultural machinery quality inspection data and the standard quality in the historical data set to obtain a quality error vector; The quality error vector is merged into the historical data set according to the historical data sequence to obtain the extended historical data set; the extended historical data set is expressed as: ; in, is the extended historical data set; for Extended quality analysis history data for each production plan; For the A standard historical raw material data vector for a production plan; For the Production steps in a production plan The standard historical equipment operating parameter vector; For the Production steps in a production plan Standard historical production environment data vector; For the Production steps in a production plan The completed standard historical production output data vector; For the Production steps in a production plan The standard historical actual production progress value; For the Production steps in a production plan The standard historical expected production progress value; For the The quality error vector in the production plan; For the The quality evaluation value of agricultural machinery parts in a production plan; the quality evaluation value of agricultural machinery parts is 0 or 1, 0 means qualified quality; 1 means unqualified quality; is the total number of production plans; is the total number of production steps; The extended historical data set is input into the progress-qualification analysis function to obtain the progress-quality impact matrix of each production step; wherein the progress-quality impact matrix is ​​expressed as: ;in, For production steps The progress-quality impact matrix; is the progress-qualification analysis function.

[0015] Production optimization unit, used to optimize production parameters during the production of agricultural machinery parts; Wherein, during the monitoring process of the intelligent monitoring unit, the production optimization unit realizes real-time optimization by applying the agricultural machinery parts production monitoring optimization model; wherein, the agricultural machinery parts production monitoring optimization model includes: A data input layer, used for receiving current production information, a historical data set and a first coefficient matrix; A data preprocessing layer, used to preprocess the production information to obtain a current production data vector; A data clustering layer, used for clustering the current production data vector with the data in the historical data set to obtain a historical data cluster set; A data screening layer, used to extract historical data of the same production step as the current production step from the historical data cluster set to obtain a historical data cluster subset; The production optimization layer is used to optimize the production parameters of the current production step; The parameter output layer is used to output the optimization parameters of the current production step.

[0016] Feedback unit, used to feedback production status; The production parameter generation unit is used to generate production parameters during the intelligent monitoring of the progress of the entire production process of agricultural machinery parts. The production parameter generation unit includes: Obtain the data of agricultural machinery parts that have passed quality inspection in the historical data set, and build a historical data set of qualified products; Performing correlation analysis on each production step in the qualified product historical data set to obtain a production step correlation matrix; The values ​​in the production step association matrix are normalized to obtain a normalized association matrix; wherein the normalized association matrix is ​​expressed as: ;in, is the normalized correlation matrix; For production steps Production steps The normalized correlation of For production steps Production steps degree of relevance; The production step influence vector on the next production step is calculated according to the normalized correlation matrix; the calculation formula of the production step influence vector is: ;in, For production steps The production step influence vector; For production steps Optimization parameters of the current production step; Optimize the parameters according to the current production step to produce and obtain the expected production results; The production parameters of the next production step are obtained according to the expected production result, the production step impact vector and the next production target.

[0017] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention proposes a production parameter impact analysis method for each production step in the production process of agricultural machinery accessories. The method obtains the parameter error vector of agricultural machinery accessories by calculating the parameter error between the agricultural machinery accessories data of the final production step in the historical data set and the quality inspection standard, and merges it into the extended historical data set. The extended historical data set includes standard data of each production plan, equipment operating parameters, production environment information, and production output data, which fully reflects the actual execution of each production step. By inputting the expanded quality analysis historical data into the progress-qualification analysis function, the progress-quality impact matrix of each production step is generated, thereby providing reliable data support for the subsequent optimization of production parameters and the improvement of production progress and product quality.

[0018] 2. The present invention proposes a production monitoring and optimization model for agricultural machinery parts, which aims to optimize the production parameters in the entire production process of agricultural machinery parts. The model collects and organizes various types of information in the current production process through the data input and preprocessing layer, and uses cluster analysis technology to process and group historical data to screen out historical data similar to the current production steps. By constructing an error matrix and calculating forward and reverse discreteness, the model can accurately identify parameter deviations in production steps and adjust relevant parameters in a timely manner, thereby optimizing the production process, improving production efficiency, and ensuring quality stability. The optimization process adopts an iterative update method to continuously refine parameter adjustments, ultimately achieving more accurate production control, significantly improving production efficiency, and effectively reducing the generation of defective products.

[0019] 3. The present invention proposes a method for generating production parameters in subsequent production steps. The method first obtains the data of agricultural machinery parts that have passed the quality inspection in the historical data set, and constructs a historical data set of qualified products to ensure that the data based on it has high quality and accuracy. Next, the method performs a correlation analysis on the historical data of these qualified products to obtain the correlation matrix between each production step, and converts it into a standardized correlation matrix through normalization. By calculating the normalized correlation matrix, the influence vector of each production step on the next production step can be further derived. Based on this influence vector, combined with the optimization parameters of the current production step, the key parameters in the production process can be effectively adjusted. By optimizing the parameters and processes of each production step, the production quality and progress of agricultural machinery parts are significantly improved.

[0020] 4. The present invention realizes precise control and optimization of the entire production process of agricultural machinery accessories through the agricultural machinery accessories production monitoring optimization model and the subsequent production parameter generation method, thereby ensuring the continuous stability of production progress and product quality. Specifically, the present invention dynamically optimizes the various parameters of the current production step by collecting key monitoring point data in the production process in real time, combining historical production data and quality control information. This real-time feedback mechanism can ensure that each production link operates under optimal conditions, maximizing production efficiency and product quality. In practical applications, when the monitoring point changes, the system can automatically adjust the production parameters according to the latest monitoring data, and generate the optimized parameters for the subsequent production steps accordingly. Through this mechanism, the present invention can effectively deal with various changing factors that may occur in the production process, ensure that the production process is highly adaptable, and can flexibly respond to different production conditions and demand changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 A flow chart of a method for intelligently monitoring the progress of the entire production process of agricultural machinery parts provided by an embodiment of the present invention; Figure 2 A structural diagram of an intelligent monitoring system for the entire production process of agricultural machinery parts provided by an embodiment of the present invention; Figure 3 An optimization process of a production optimization layer provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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.

[0023] Agricultural machinery accessories refer to various parts and components used in agricultural machinery and equipment to ensure the normal operation of agricultural machinery and equipment and improve their operating efficiency, reliability and durability. Agricultural machinery accessories play a vital role in completing agricultural production tasks such as tillage, sowing, harvesting, spraying and irrigation. With the continuous improvement of the mechanization of agricultural production, the demand for agricultural machinery accessories has also gradually increased.

[0024] As a high-precision device, agricultural machinery parts have extremely high requirements for accuracy, quality and reliability in their production process. With the advancement of science and technology and the improvement of modernization needs, the production of agricultural machinery parts has gradually been automated and systematized. This transformation has not only greatly improved production efficiency, but also improved the stability of the production process and product quality. In the automated production process of agricultural machinery parts, intelligent monitoring of the entire process is a very important aspect. Intelligent monitoring can not only obtain production data in real time, but also optimize production parameters and adjust production progress through efficient data analysis and feedback mechanisms to ensure that each link operates in the optimal state. In this monitoring environment, how to achieve full-process control of agricultural machinery parts production is an urgent problem to be solved. There is a close interdependence between the various production steps of agricultural machinery parts. The output of the previous link not only directly affects the production quality of the next link, but may also have a profound impact on the overall production progress. If the interrelationships between the production steps are ignored, and the parameters and processes of each link are not reasonably regulated, it may lead to delays in production progress, quality fluctuations, and even the production of defective products, thereby affecting the smooth progress of the entire production process and product quality. To this end, the present invention proposes a method and system for intelligent monitoring of the progress of the entire process of agricultural machinery parts production, aiming to intelligently optimize production parameters in the production process of agricultural machinery parts. Two embodiments are described in detail below.

[0025] Embodiment 1: In the embodiment of the present application, the agricultural machinery parts production task of the agricultural machinery parts manufacturer A is monitored by the method and system proposed in the present application; it is known that the agricultural machinery parts manufacturer A mainly produces agricultural machinery parts 1, agricultural machinery parts 2 and agricultural machinery parts 3; the embodiment of the present application will perform full-process progress intelligent monitoring of agricultural machinery parts 1; wherein, for the monitoring process of the embodiment of the present application, combined with Figure 1 and Figure 2 Explain the content; Figure 1 The specific process of the method provided by the present invention includes: S10. Collecting current production information of monitoring points; S20. Obtaining multiple historical data of agricultural machinery parts production; S30. Processing the historical data to obtain standard historical data; S40. Marking multiple standard historical data and constructing a historical data set; S50. Analyzing the historical data set to obtain the progress-quality impact matrix of each production step; S60. Obtaining the progress-quality impact matrix corresponding to the current production step as the first coefficient matrix; S70. Inputting the current production information, the historical data set and the first coefficient matrix into the agricultural machinery parts production monitoring optimization model to obtain the optimization parameters of the current production step; S80. Using the optimization parameters of the current production step to produce, and generate the production parameters of the next production step. Figure 2It is a structural diagram of the system of the present invention, comprising: a data acquisition unit, a data storage unit, a data processing unit, an intelligent monitoring unit, an intelligent analysis unit, a production optimization unit, a feedback unit and a production parameter generation unit; according to Figure 1 and Figure 2 The contents are described in detail below; The intelligent monitoring unit of the system is used to monitor the entire process of the agricultural machinery accessory 1. The intelligent monitoring unit can be used to know that the agricultural machinery accessory 1 is in the production step 2. It is known that the production of the agricultural machinery accessory 1 has 6 steps, including: raw material preparation, parts processing, assembly, surface treatment, quality inspection and packaging. Further, the data acquisition unit of the system is used to collect the current production information of the monitoring point of the agricultural machinery accessory 1, that is, the information collected at the production step 2 (parts processing), corresponding to the above-mentioned step S10; wherein the current production information includes: the current production step, the current equipment operation parameters, the data of the material to be processed in the production step 2, the current expected production progress of the production step 2, and the current production environment data at the production step 2; refer to Table 1, which shows the data collected by the data acquisition unit at the monitoring point; Table 1 Data collected from monitoring points

[0026] Further, a plurality of historical data of the production of the agricultural machinery parts 1 is obtained from the data storage unit of the system, corresponding to step S20; wherein, in the embodiment of the present application, the historical data refers to the historical production data of each production step of the agricultural machinery parts 1 in the entire production process; including: historical raw material data, historical equipment operation parameters, historical production environment data, historical production output data (material output results when each production step is completed), and production progress expectations of each production step (including comparison data of historical actual production progress and historical expected production progress); Furthermore, multiple historical data acquired by the data processing unit of the system are processed to obtain multiple standard historical data, corresponding to step S30; wherein the processing of historical data includes: performing data cleaning on the historical data to obtain first historical data; performing smoothing on the first historical data to obtain second historical data; performing formatting on the second historical data to obtain third historical data; and performing normalization on the third historical data to obtain the standard historical data.

[0027] In the embodiment of the present application, a data processing unit is used to process the multiple historical data obtained, including: data cleaning, smoothing, formatting and normalization; the above data processing means can significantly improve the accuracy and effectiveness of the production parameter impact analysis in subsequent production steps.

[0028] Furthermore, the processed historical data is marked and a historical data set is constructed, corresponding to step S40; wherein the marking of the standard historical data is divided into two categories according to the historical quality inspection results: qualified and unqualified.

[0029] Furthermore, the intelligent analysis unit of the system is used to analyze the data in the historical data set, corresponding to step S50; wherein the intelligent analysis unit is used to analyze the impact of each production step on the production quality and production progress; the specific process is as follows: Calculate the error between the quality inspection data of agricultural machinery parts 1 in the historical data set and the standard quality, and obtain the quality error vector of the agricultural machinery parts; Further, the quality error vectors are merged according to corresponding positions in the historical data set to obtain an extended historical data set; Among them, the extended historical data set of agricultural machinery accessories 1 is expressed as: ; in, It is an extended historical data set of agricultural machinery accessories 1; for Extended quality analysis history data for each production plan; For the A standard historical raw material data vector for a production plan; For the Production steps in a production plan The standard historical equipment operating parameter vector; For the Production steps in a production plan Standard historical production environment data vector; For the Production steps in a production plan The completed standard historical production output data vector; For the Production steps in a production plan The standard historical actual production progress value; For the Production steps in a production plan The standard historical expected production progress value; For the The quality error vector in the production plan; For the The quality evaluation value of agricultural machinery parts in a production plan; the quality evaluation value of agricultural machinery parts is 0 or 1, 0 means qualified quality; 1 means unqualified quality; is the total number of production plans; is the total number of production steps; in the present embodiment of the present application, the agricultural machinery accessory 1 has a total of 6 production steps, so The value is 6; Furthermore, the data in the extended historical data set is input into the progress-qualification analysis function to obtain the progress-quality impact matrix of each production step; wherein the progress-quality impact matrix is ​​expressed as: ;in, For production steps The progress-quality impact matrix; It is the progress-qualification analysis function.

[0030] Among them, for the progress-qualification analysis function It includes: quality analysis sub-function and progress analysis sub-function; that is, the progress-qualification analysis function is expressed as: ;in, is the quality analysis sub-function; is a constant term; is the weight vector affecting the quality; is the error correction constant; It is the progress analysis sub-function; is a random forest model; refer to Table 2, which shows the impact of the agricultural machinery accessory 1 in the production step 2 of the embodiment of the present application; Table 2 The impact of production step 2 on the production of agricultural machinery parts 1

[0031] In the embodiment of the present application, the intelligent analysis unit is used to realize the influence of each production step of agricultural machinery parts 1 on the quality and progress of production; wherein, the unit can quantify the relationship between progress and quality in each step of agricultural machinery parts production, and identify the key factors affecting production efficiency and product quality. By combining the progress-qualification analysis function with the extended historical data, the system can provide comprehensive production process optimization suggestions and realize the optimal control of production quality and progress.

[0032] Further, according to the production step 2 corresponding to the monitoring point of the agricultural machinery accessory 1, the progress-quality influence matrix of the step is selected as the first coefficient matrix, corresponding to step S60; Further, the current production information, the historical data set and the first coefficient matrix of the production step 2 are input into the production optimization unit of the system to optimize the production parameters, corresponding to step S70; wherein the optimization unit adopts the agricultural machinery parts production monitoring optimization model, specifically including: The current production information, the historical data set and the first coefficient matrix are input from the input layer; Furthermore, the data preprocessing layer is used to preprocess the current production information to obtain the current production data vector; wherein the data preprocessing layer includes: (1) removing noise and outliers; (2) data standardization; (3) time synchronization and correlation processing; (4) missing value filling and interpolation; (5) data normalization; (6) data encoding and vectorization; (7) generating the current production data vector.

[0033] According to the above preprocessing formula, the preprocessing results in Table 1 are: [5000, 25, 100, 0.05, 30, 0.5, 2, 22, 0.5, 60, 1]; each data item represents the production information after preprocessing, for example: 5000: processing equipment speed (unit: RPM); 25: equipment temperature (unit: ℃); 100: processing equipment pressure (unit: bar); 0.05: raw material size tolerance (unit: mm); 30: material hardness (unit: HRC); 0.5: completed processing volume (50%); 2: expected completion time (unit: hour); 22: production environment temperature (unit: ℃); 0.5: humidity (50%); 60: workshop noise (unit: dB); 1: material type code (such as steel is 1, other materials are 0).

[0034] Furthermore, after data preprocessing, the data is input to the data clustering layer; wherein, the data clustering layer is used to cluster the current production data vector with the data in the historical data set; the specific process is: Step 1: Define K-means clustering as the clustering algorithm; Step 2: Define clustering features. When comparing the current production data vector with the historical data set, it is necessary to define the feature space. For example, the production data items mentioned in Table 1, such as equipment speed, equipment temperature, and processing pressure, can all be used as features for cluster analysis; Step 3: Calculate the similarity by using the cosine similarity calculation method to calculate the similarity between each set of historical data samples and the current production data vector; Step 4: Execute the K-means clustering algorithm to cluster the current production data vector and the historical data set. The clustering result will form several clusters, each of which contains a set of similar historical data. At this time, the current production data vector will be associated with the historical data in a cluster, indicating that the current production status is similar to the historical production status in the cluster.

[0035] Step 5: Obtain a clustering set of historical data; Furthermore, the historical data cluster set is input into the data screening layer, and is screened according to the screening rules to obtain the historical data cluster subset; wherein the screening rule of the data screening layer is to use label matching: if each record of the historical data has a clear production step label, it can be screened by comparing the current production step with the historical data label. In the embodiment of the present application, the current production step is "production step 2", so according to the production step label, the historical data other than "production step 2" in the historical data cluster set should be eliminated; Furthermore, the current production data vector and the historical data cluster subset are input into the production optimization layer; wherein the production optimization layer is used to optimize the production parameters of the current production step; Figure 3 As shown, the specific process includes: Step 1: Obtain the current production data vector and the historical data cluster subset; Step 2: Calculate the error between the current production data vector and each data vector in the historical data cluster subset, and construct the current production error matrix in combination with the quality annotation content; refer to Table 3, which gives an example of the error between the first three data in the historical data cluster subset and the current production data vector; Table 3. Example of error comparison between historical data cluster subset and current production data vector.

[0036] Step 3: Calculate the forward discreteness and reverse discreteness in the current production error matrix; wherein the forward discreteness refers to the error data in the current production error matrix that is completed according to the expected progress and the quality is marked as qualified; the reverse discreteness refers to the error data in the current production error matrix that is not completed according to the expected progress and the quality is marked as unqualified; The calculation formula of the forward dispersion is: ;in, is the number of data that meets the positive dispersion condition, that is and ; For the The actual progress of historical data; For the Expected progress of historical data; For the The quality of historical data is qualified; The index set of all data that meets the positive condition; is the first cluster in the historical data cluster subset that meets the positive condition. The error of the data vector; The calculation formula of the forward dispersion is: Among them, is the number of data that meets the inverse dispersion condition, that is and ; For the The quality of historical data is unqualified; The index set of all data that meets the reverse condition; is the first cluster in the historical data cluster subset that meets the reverse condition. The error of the data vector; Step 4: Optimize the current production data vector using the first coefficient matrix according to the forward discreteness and the reverse discreteness; The process of optimizing the current production data vector includes: The positive dispersion and the reverse dispersion are used as the optimized positive index and the optimized reverse index; The objective function is constructed as, ;in, To optimize the positive indicator balance factor; To optimize the balance factor of the reverse indicator; Substitute the first coefficient matrix into the nonlinear adjustment function to optimize the current production data vector; the nonlinear adjustment function is: ;in, is the optimized current production data vector; is the current production data vector; is the first coefficient matrix; is the gradient of the objective function; Step 5, using the optimized current production data vector as the current production data vector; Step 6: Repeat steps 2 to 5 until the conditions are met to obtain the optimization parameters of the current production step. In this embodiment of the present application, the optimization condition is set to stop iteration when the error between the current production data vector and the historical data cluster subset is less than 0.01; Refer to Table 4, which gives the forward discrete value, reverse discrete value and error value after multiple rounds of iterations; Table 4 Forward discreteness, reverse discreteness and error

[0037] In Table 4, the total error of the forward discreteness is 0.15 in the first round, 0.12 in the second round, 0.10 in the third round, 0.08 in the fourth round, and 0.11 in the fifth round. After four rounds of iteration, the total error of the forward discreteness decreased from 0.15 to 0.08, showing a trend of gradual optimization. However, in the fifth round, the forward discreteness rebounded to 0.11. This may indicate that some optimized parameter adjustments fluctuated during this round of iteration. The total error of the reverse discreteness is 0.30 in the first round, 0.25 in the second round, 0.20 in the third round, 0.18 in the fourth round, and 0.20 in the fifth round. During the iteration process, the total error of the reverse discreteness gradually decreased in the first few rounds, reflecting the improvement of the optimization on the unqualified data. However, in the fifth round, the error rebounded, just like the forward discreteness, indicating that some optimized parameters may be unstable. Error value: The error value of the first round is 0.020, the second round is 0.017, the third round is 0.013, the fourth round is 0.007, and the fifth round is 0.010. The error value gradually decreases from 0.020 in the first round to 0.007 in the fourth round, showing the effectiveness of the optimization process. Although the error value of the fifth round slightly rebounded to 0.010, the overall trend is that the error gradually decreases, indicating that the optimization is gradually becoming stable.

[0038] Refer to Table 5, which gives an explanation of the parameter optimization in the current production step; Table 5 Changes in equipment parameters in the current production step

[0039] In the embodiment of the present application, the production optimization layer is used to optimize the parameters of the current production step; the production optimization layer workflow includes the following core steps: Data acquisition and clustering: First, the current production data vector is obtained and clustered with the historical production data to form a cluster subset of the historical data. Error calculation and error matrix construction: The error between the current production data and the historical data is calculated, and the error matrix is ​​constructed in combination with the quality annotation information. The error matrix reflects the difference between the current production step and the historical production step. Discreteness calculation: The quality of the production data is evaluated by calculating the forward discreteness and reverse discreteness of the current production error matrix. The forward discreteness focuses on the error data that is completed according to the expected schedule and has qualified quality, while the reverse discreteness focuses on the error data that is not completed according to the expected schedule and has unqualified quality. Data optimization: According to the calculation results of the forward and reverse discreteness, the current production data vector is optimized using the first coefficient matrix to improve the accuracy of the production parameters and production efficiency. Iterative optimization: By repeatedly iterating steps 2 to 5, the current production data is continuously optimized until the expected optimization effect is achieved, and finally the optimized current production step parameters are obtained. The production optimization layer ensures accurate tuning of production step parameters through multiple rounds of error calculation, discrete analysis and data optimization, thereby improving production quality and efficiency.

[0040] Further, the parameter output layer is used to output the optimization parameters of the current production step of "production step 2"; Further, the production parameter generation unit is used to generate corresponding production parameters for the next production step according to the optimization parameters of the current production step, corresponding to step S80; according to the embodiment of the present application, the next production step of the agricultural machinery accessory 1 can be "assembly"; the specific process includes: Obtain the data of agricultural machinery parts that have passed the quality inspection in the historical data set of agricultural machinery parts 1, and construct a historical data set of qualified products; The correlation analysis of each production step in the qualified product historical data set is performed to obtain a production step correlation matrix; wherein, the correlation analysis of each production step adopts a covariance calculation method; refer to Table 6, which gives the correlation description of the 6 production steps of the agricultural machinery parts 1 in the implementation of this application; Table 6 Correlation description of production steps of agricultural machinery parts 1

[0041] Furthermore, the values ​​in the production step association matrix are normalized to obtain a normalized association matrix; wherein the normalized association matrix is ​​expressed as: ;in, is the normalized correlation matrix; For production steps Production steps The normalized correlation of For production steps Production steps Refer to Table 7; the normalized values ​​are given in Table 7; Table 7 Normalized correlation matrix of agricultural machinery accessories 1

[0042] Furthermore, the production step influence vector on the next production step is calculated according to the normalized correlation matrix; the calculation formula of the production step influence vector is: ;in, For production steps The production step influence vector; For production steps Optimization parameters of the current production step; Further, the production is performed according to the current production step optimization parameters of "production step 2" to obtain the desired production result; Furthermore, the production parameters of the next production step (assembly) are obtained according to the expected production results, the production step impact vector and the next production target (assembly target).

[0043] In an embodiment of the present application, a method for generating production parameters in subsequent production steps is used to generate parameters for the "assembly" step; the method first obtains the data of agricultural machinery parts that have passed the quality inspection in the historical data set, and constructs a historical data set of qualified products to ensure that the data based on it has high quality and accuracy. Next, the method performs a correlation analysis on the historical data of these qualified products to obtain the correlation matrix between each production step, and converts it into a standardized correlation matrix through normalization. By calculating the normalized correlation matrix, the influence vector of each production step on the next production step can be further derived. Based on this influence vector, combined with the optimization parameters of the current production step (parts processing), the key parameters in the production process can be effectively adjusted.

[0044] Furthermore, the feedback unit of the system is used to provide real-time feedback on the production status of the monitoring process, including progress and production change process.

[0045] In the embodiment of the present application, the intelligent monitoring method and system of the present invention are used to perform full-process progress intelligent monitoring on the production process of agricultural machinery parts 1 of agricultural machinery parts manufacturer A. Taking the "parts processing" step as an example, the monitoring process is completed through various functional modules of the system, including data acquisition, historical data processing, progress-quality impact matrix analysis, production parameter optimization, etc. First, the data acquisition unit collects relevant information of the current production step in real time, such as equipment operating parameters, material data, production progress and environmental data. Then, the system constructs a progress-quality impact matrix by processing, marking and analyzing historical data, and identifies the impact of various production factors on quality and progress. In the analysis stage, the system uses the progress-qualification analysis function combined with historical data to quantify the impact of production steps on quality and progress. The optimization unit further optimizes the production parameters according to the current production information and the progress-quality impact matrix, and adjusts the production parameters through an iterative process to ensure that the progress and quality of each production step are in the best state. Through this system, each production link can be accurately monitored, the production strategy can be adjusted in real time, and the production efficiency and product quality can be optimized, thereby improving the controllability and intelligence level of the production process. Ultimately, this method not only improves production efficiency, but also reduces quality fluctuations and promotes the realization of intelligent manufacturing.

[0046] Implementation 2: In Example 1, the method and system of the present invention are used to realize the intelligent monitoring of the progress of the whole process of the production of agricultural machinery parts 1 of agricultural machinery parts manufacturer A; for further explanation, the present invention has the same applicable effect on other agricultural machinery parts. In the embodiment of the present application, the whole process of the production of agricultural machinery parts 2 of agricultural machinery parts manufacturer A is monitored, and the specific process is as follows: Collect the current production information of the monitoring point during the production of agricultural machinery parts 2; Further, a plurality of historical data of the agricultural machinery parts 2 during production is obtained; Further, the historical data is processed to obtain standard historical data; Further, the standard historical data is labeled for eligibility and a historical data set is constructed; Furthermore, the historical data set of the agricultural machinery parts 2 is analyzed to obtain the progress-quality impact matrix of each production step of the agricultural machinery parts 2; Further, the progress-quality impact matrix corresponding to the production step of the monitoring point of the agricultural machinery accessory 2 is obtained as the first coefficient matrix; Further, the current production information, the historical data set and the first coefficient matrix are input into the agricultural machinery parts production monitoring optimization model to obtain the current production step optimization parameters; Among them, the agricultural machinery parts production monitoring optimization model includes: A data input layer, used for receiving current production information, a historical data set and a first coefficient matrix; The data preprocessing layer is used to preprocess the production information to obtain the current production data vector; The data clustering layer is used to cluster the current production data vector with the data in the historical data set to obtain a historical data cluster set; The data screening layer is used to extract the historical data of the same production step as the current production step from the historical data cluster set to obtain the historical data cluster subset; The production optimization layer is used to optimize the production parameters of the current production step; Wherein, the production optimization layer comprises the following steps: Step 1: Obtain the current production data vector and the historical data cluster subset; Step 2: Calculate the error between the current production data vector and each data vector in the historical data cluster subset, and construct the current production error matrix in combination with the quality annotation content; Step 3, respectively calculating the positive discreteness and the negative discreteness in the current production error matrix; wherein the positive discreteness refers to the error data in the current production error matrix that is completed according to the expected progress and the quality is marked as qualified; the negative discreteness refers to the error data in the current production error matrix that is not completed according to the expected progress and the quality is marked as unqualified; The calculation formula of the forward dispersion is: ;in, is the number of data that meets the positive dispersion condition, that is and ; For the The actual progress of historical data; For the Expected progress of historical data; For the The quality of historical data is qualified; The index set of all data that meets the positive condition; is the first cluster in the historical data cluster subset that meets the positive condition. The error of the data vector; The calculation formula of the forward dispersion is: ;in, is the number of data that meets the inverse dispersion condition, that is and ; For the The quality of historical data is unqualified; The index set of all data that meets the reverse condition; is the first cluster in the historical data cluster subset that meets the reverse condition. The error of the data vector; Step 4: Optimize the production parameters in the current production data vector using a first coefficient matrix according to the forward discreteness and the reverse discreteness; Step 5: Using the optimized current production data vector as the current production data vector; Step 6: Repeat step 2 to step 5 until the conditions are met to obtain the optimization parameters of the current production step.

[0047] The parameter output layer is used to output the optimization parameters of the current production step; Furthermore, the current production step optimization parameters are used to produce the material to be processed, and the production parameters of the next production step are generated; the specific process includes: Acquire the data of agricultural machinery parts that have passed the quality inspection in the historical data set, and construct a historical data set of qualified products; Furthermore, the correlation analysis is performed on each production step in the qualified product historical data set to obtain a production step correlation matrix; Furthermore, the values ​​in the production step association matrix are normalized to obtain a normalized association matrix; wherein the normalized association matrix is ​​expressed as: ;in, is the normalized correlation matrix; For production steps Production steps The normalized correlation of For production steps Production steps degree of relevance; Furthermore, the production step influence vector on the next production step is calculated according to the normalized correlation matrix; the calculation formula of the production step influence vector is: ;in, For production steps The production step influence vector; For production steps Optimization parameters of the current production step; Further, the production is performed according to the optimized parameters of the current production step to obtain the desired production result; Furthermore, according to the expected production result, the production step impact vector and the next production target, the production parameter of the next production step is obtained.

[0048] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent monitoring of the progress of the entire production process of agricultural machinery parts, characterized in that: include: Collect current production information of monitoring points; obtain multiple historical data of agricultural machinery parts production; Processing the historical data to obtain standard historical data; marking a plurality of the standard historical data and constructing a historical data set; analyzing the historical data set to obtain a progress-quality impact matrix for each production step; obtaining the progress-quality impact matrix corresponding to the current production step as a first coefficient matrix; Inputting the current production information, the historical data set and the first coefficient matrix into the agricultural machinery parts production monitoring optimization model to obtain the current production step optimization parameters; The agricultural machinery parts production monitoring optimization model includes: a data input layer, used to receive the current production information, the historical data set and the first coefficient matrix; a data preprocessing layer, used to preprocess the production information to obtain the current production data vector; a data clustering layer, used to cluster the current production data vector with the data in the historical data set to obtain a historical data clustering set; a data screening layer, used to extract the historical data of the same production step as the current production step from the historical data clustering set to obtain a historical data clustering subset; a production optimization layer, used to optimize the production parameters of the current production step; a parameter output layer, used to output the optimization parameters of the current production step; The optimized parameters of the current production step are used for production, and the production parameters of the next production step are generated.

2. The method for intelligently monitoring the progress of the entire production process of agricultural machinery parts according to claim 1 is characterized in that: The processing process of the historical data includes: performing data cleaning on the historical data to obtain first historical data; performing smoothing on the first historical data to obtain second historical data; performing formatting on the second historical data to obtain third historical data; and performing normalization on the third historical data to obtain the standard historical data.

3. The method for intelligently monitoring the progress of the entire production process of agricultural machinery parts according to claim 1 is characterized in that: The calculation process of the progress-quality impact matrix includes: Calculating the error between the agricultural machinery quality inspection data and the standard quality in the historical data set to obtain a quality error vector; The quality error vector is merged into the historical data set according to the historical data sequence to obtain an extended historical data set; the extended historical data set is expressed as: ； in, is the extended historical data set; for Extended quality analysis history data for each production plan; For the A standard historical raw material data vector for a production plan; For the Production steps in a production plan The standard historical equipment operating parameter vector; For the Production steps in a production plan Standard historical production environment data vector; For the Production steps in a production plan The completed standard historical production output data vector; For the Production steps in a production plan The standard historical actual production progress value; For the Production steps in a production plan The standard historical expected production progress value; For the The quality error vector in the production plan; For the Quality assessment value of agricultural machinery parts in a production plan; is the total number of production plans; is the total number of production steps; The extended historical data set is input into the progress-qualification analysis function to obtain the progress-quality impact matrix of each production step; wherein the progress-quality impact matrix is ​​expressed as: ;in, For production steps The progress-quality impact matrix; is the progress-qualification analysis function.

4. The method for intelligently monitoring the progress of the entire production process of agricultural machinery parts according to claim 1 is characterized in that: The production optimization layer includes the following steps: Step 1: Obtain the current production data vector and the historical data cluster subset; Step 2: Calculate the error between the current production data vector and each data in the historical data cluster subset, and construct a current production error matrix in combination with the quality annotation content; Step 3: Calculate the forward discreteness and reverse discreteness in the current production error matrix; wherein the forward discreteness refers to the error data in the current production error matrix that is completed according to the expected progress and the quality is marked as qualified; the reverse discreteness refers to the error data in the current production error matrix that is not completed according to the expected progress and the quality is marked as unqualified; The calculation formula of the forward dispersion is: ;in, is the number of data that meets the positive dispersion condition, that is and ; For the The actual progress of historical data; For the Expected progress of historical data; For the The quality of historical data is qualified; The index set of all data that meets the positive condition; is the first cluster in the historical data cluster subset that meets the positive condition. The error of the data vector; The calculation formula of the forward dispersion is: ;in, is the number of data that meets the inverse dispersion condition, that is and ; For the The quality of historical data is unqualified; The index set of all data that meets the reverse condition; is the first cluster in the historical data cluster subset that meets the reverse condition. The error of the data vector; Step 4: Optimize the production parameters in the current production data vector using a first coefficient matrix according to the forward discreteness and the reverse discreteness; Step 5: Using the optimized current production data vector as the current production data vector; Step 6: Repeat step 2 to step 5 until the conditions are met to obtain the optimization parameters of the current production step.

5. The method for intelligently monitoring the progress of the entire production process of agricultural machinery parts according to claim 1 is characterized in that: The process of using the optimized parameters of the current production step to produce the material to be processed and generating the production parameters of the next production step includes: Acquire the data of agricultural machinery parts that have passed the quality inspection in the historical data set, and construct a historical data set of qualified products; Performing correlation analysis on each production step in the qualified product historical data set to obtain a production step correlation matrix; The values ​​in the production step association matrix are normalized to obtain a normalized association matrix; wherein the normalized association matrix is ​​expressed as: ;in, is the normalized correlation matrix; For production steps Production steps The normalized correlation of For production steps Production steps degree of relevance; The production step influence vector on the next production step is calculated according to the normalized correlation matrix; the calculation formula of the production step influence vector is: ;in, For production steps The production step influence vector; For production steps Optimization parameters of the current production step; Perform production according to the optimized parameters of the current production step to obtain the expected production results; The production parameters of the next production step are obtained according to the expected production result, the production step impact vector and the next production target.

6. An intelligent monitoring system for the whole process of agricultural machinery parts production, characterized in that: The system includes: a data acquisition unit for collecting various data in the whole process of agricultural machinery parts production; a data storage unit for storing data; a data processing unit for processing data; an intelligent monitoring unit for intelligently monitoring the progress of the whole process of agricultural machinery parts production; an intelligent analysis unit for analyzing the progress and quality of the whole process of agricultural machinery parts production; a production optimization unit for optimizing production parameters in the process of agricultural machinery parts production; a feedback unit for feeding back production status; and a production parameter generation unit for generating production parameters in the process of intelligently monitoring the progress of the whole process of agricultural machinery parts production.

7. The intelligent monitoring system for the whole process of agricultural machinery parts production according to claim 6 is characterized in that: The data processing unit comprises: Acquire multiple historical data stored in the data storage unit; perform data cleaning on the historical data to obtain first historical data; perform smoothing on the first historical data to obtain second historical data; perform formatting on the second historical data to obtain third historical data; perform normalization on the third historical data to obtain standard historical data.

8. The intelligent monitoring system for the whole process of agricultural machinery parts production according to claim 6 is characterized in that: The intelligent analysis unit comprises: Calculate the error between the agricultural machinery quality inspection data and the standard quality in the historical data set to obtain the quality error vector; The quality error vector is merged into the historical data set according to the historical data sequence to obtain an extended historical data set; the extended historical data set is expressed as: ； in, is the extended historical data set; for Extended quality analysis history data for each production plan; For the A standard historical raw material data vector for a production plan; For the Production steps in a production plan The standard historical equipment operating parameter vector; For the Production steps in a production plan Standard historical production environment data vector; For the Production steps in a production plan The completed standard historical production output data vector; For the Production steps in a production plan The standard historical actual production progress value; For the Production steps in a production plan The standard historical expected production progress value; For the The quality error vector in the production plan; For the Quality assessment value of agricultural machinery parts in a production plan; is the total number of production plans; is the total number of production steps; The extended historical data set is input into the progress-qualification analysis function to obtain the progress-quality impact matrix of each production step; wherein the progress-quality impact matrix is ​​expressed as: ;in, For production steps The progress-quality impact matrix; is the progress-qualification analysis function.

9. The intelligent monitoring system for the whole process of agricultural machinery parts production according to claim 6 is characterized in that: The production optimization unit uses the agricultural machinery parts production monitoring optimization model to perform real-time optimization during the monitoring process of the intelligent monitoring unit; wherein the agricultural machinery parts production monitoring optimization model includes: A data input layer, used for receiving current production information, a historical data set and a first coefficient matrix; A data preprocessing layer, used to preprocess the production information to obtain a current production data vector; A data clustering layer, used for clustering the current production data vector with the data in the historical data set to obtain a historical data cluster set; A data screening layer, used to extract historical data of the same production step as the current production step from the historical data cluster set to obtain a historical data cluster subset; The production optimization layer is used to optimize the production parameters of the current production step; The parameter output layer is used to output the optimization parameters of the current production step.

10. The intelligent monitoring system for the whole process of agricultural machinery parts production according to claim 6 is characterized in that: The production parameter generating unit comprises: Obtain the data of agricultural machinery parts that have passed quality inspection in the historical data set, and build a historical data set of qualified products; Performing correlation analysis on each production step in the qualified product historical data set to obtain a production step correlation matrix; The values ​​in the production step association matrix are normalized to obtain a normalized association matrix; wherein the normalized association matrix is ​​expressed as: ;in, is the normalized correlation matrix; For production steps Production steps The normalized correlation of For production steps Production steps degree of relevance; The production step influence vector on the next production step is calculated according to the normalized correlation matrix; the calculation formula of the production step influence vector is: ;in, For production steps The production step influence vector; For production steps Optimization parameters of the current production step; Optimize the parameters according to the current production step to produce and obtain the expected production results; The production parameters of the next production step are obtained according to the expected production result, the production step impact vector and the next production target.

Images