Big data analysis method for production management

Abstract

The application relates to the field of data processing, and provides a big data analysis method for production management, which comprises processing a historical processing parameter set in an ammonium sulfate production process to obtain a processing quality corresponding to each processing parameter sequence in the historical processing parameter set; the historical processing parameter set comprises a plurality of processing parameter sequences, and each processing parameter sequence comprises a plurality of processing parameters; the necessity of optimization of each processing parameter in a processing parameter sequence is determined based on the relationship between each processing parameter sequence and the corresponding processing quality, and if the necessity of optimization corresponding to the processing parameter is greater than a preset threshold value, it is determined that the processing parameter needs to be optimized and adjusted. The method considers the relationship between various processing parameters, makes the relationship between the processing parameters and the processing quality clearer, can effectively optimize the processing parameters, and improves the processing quality.

Description

Technical Field

[0001] This application relates to the field of data processing, and in particular to a big data analysis method for production management. Background Technology

[0002] In the production of ammonium sulfate, many factors affect its quality. These include human factors such as the operator's skill level and sense of responsibility, as well as factors related to processing parameters and equipment. The main factors include: excessive fluctuations in mother liquor acidity, leading to abnormal crystallization; loss of control over water and heat balance in the saturator, affecting the stable operation of the saturator and the particle size of ammonium sulfate crystals; improper operation of the centrifuge, resulting in excessive moisture and free acid in the ammonium sulfate; unreasonable control of the crystal ratio, which is not conducive to the formation of large-particle ammonium sulfate; excessive impurities in the mother liquor, affecting the growth of ammonium sulfate crystals; and low industrial air pressure, which is insufficient for the agitation and material lifting in the saturator.

[0003] The aforementioned factors can lead to a decline in product quality and production efficiency during the production process. Therefore, in order to improve production quality and efficiency, the ammonium sulfate production process can be optimized based on existing data to reduce the losses caused by the above problems.

[0004] In existing technologies, when optimizing processing parameters, the influence of a single parameter on the final processing quality is often considered, while the interrelationships between processing parameters are ignored, resulting in insignificant optimization effects. Summary of the Invention

[0005] This invention provides a big data analysis method for production management. This method considers the relationship between various processing parameters, making the relationship between processing parameters and processing quality clearer, and can effectively optimize processing parameters and improve processing quality.

[0006] Firstly, this application provides a big data analysis method for production management, comprising: The historical processing parameter set in the ammonium sulfate production process is processed to obtain the processing quality corresponding to each processing parameter sequence in the historical processing parameter set; the historical processing parameter set includes multiple processing parameter sequences, and each processing parameter sequence includes multiple processing parameters. The necessity of optimizing each processing parameter in the processing parameter sequence is determined based on the relationship between each processing parameter sequence and the corresponding processing quality. If the necessity for optimization corresponding to the processing parameter is greater than a preset threshold, then it is determined that the processing parameter needs to be optimized and adjusted.

[0007] Optionally, the historical processing parameter set in the ammonium sulfate production process is processed to obtain the processing quality corresponding to each processing parameter sequence in the historical processing parameter set, including: Obtain the ammonium sulfate quality description parameters corresponding to each processing parameter sequence. The quality description parameters include: nitrogen content, free acid ratio, and moisture ratio. The nitrogen content, free acid ratio, and moisture ratio are processed using a preset model algorithm to obtain the processing quality corresponding to each processing parameter sequence. The processing parameters in the processing parameter sequence include: saturator inlet acid gas temperature, mother liquor temperature, mixer resistance, mother liquor acidity, stirring air volume, stirring air temperature, feed mother liquid ratio, and ammonium sulfate consumption.

[0008] Optionally, the necessity of optimizing each processing parameter in the processing parameter sequence is determined based on the relationship between each processing parameter sequence and the corresponding processing quality, including: Calculate the importance of each processing parameter in the processing parameter sequence; Calculate the average processing quality corresponding to each processing parameter; The necessity of optimizing each processing parameter is determined based on the importance of each processing parameter and the average processing quality corresponding to each processing parameter.

[0009] Optionally, the importance of each processing parameter in the processing parameter sequence is calculated, including: The degree of influence of each processing parameter on the processing quality is determined based on the degree of change in the processing quality corresponding to each processing parameter. Calculate the importance of the influence of each processing parameter in the processing parameter sequence on the corresponding processing quality; Based on the degree of influence of each processing parameter in the processing parameter sequence on the processing quality, and the importance of the degree of influence of each processing parameter in the processing parameter sequence on the corresponding processing quality, the importance of each processing parameter in the processing parameter sequence is calculated.

[0010] Optionally, the degree of influence of each processing parameter on the processing quality is determined based on the degree of change in processing quality corresponding to each processing parameter, including: Determine the importance of the current processing parameters across their respective value ranges; The degree of influence of the current processing parameters on the processing quality is determined based on the importance of the current processing parameter values ​​within each range.

[0011] Optionally, the importance of the current processing parameter in each value range is determined, including: The degree of influence of the current processing parameter value a on the processing quality is determined based on the difference between the processing quality when the current processing parameter is value a and the average processing quality corresponding to value a. A coordinate system is established with the current processing parameter values ​​as the horizontal axis and the degree of influence of each value on the processing quality as the vertical axis. The data points on the coordinate system are clustered to obtain multiple first-class clusters. Each first-class cluster represents the degree of influence of each range of current processing parameter values ​​on the processing quality. The processing quality is clustered to obtain multiple second-class clusters, each of which represents processing parameters for different quality ranges; The importance of each value range of the current processing parameter is calculated based on the average value of the influence of the current processing parameter on the processing quality within the value range, the average processing quality corresponding to the current processing parameter within the value range, the number of processing parameter sequences in the second category with the highest quality range, and the number of processing parameter sequences in the second category with the highest quality range corresponding to each value range of the current processing parameter. The degree of influence of the current processing parameters on processing quality is determined based on the importance of the current processing parameter values ​​within each value range, including: The average value of the importance of the current processing parameter across its various value ranges is determined as the degree of influence of the current processing parameter on the processing quality.

[0012] Optionally, the importance of the influence of each processing parameter in the processing parameter sequence on the corresponding processing quality is calculated, including: Calculate the degree of influence of two processing parameters on processing quality; The importance of each processing parameter's influence on the corresponding processing quality is determined based on the degree of influence of the two processing parameters on the processing quality and the degree of influence of one of the two processing parameters on the processing quality.

[0013] Optionally, the degree of influence of the two processing parameters on the processing quality is calculated, including: The influence of two machining parameters on machining quality can be calculated using the following formula: in, This represents the number of times that the i-th value range of the e-th processing parameter and the j-th value range of the f-th processing parameter in the processing parameter sequence simultaneously produce high-quality products. This indicates the number of processing parameter sequences in the second type of cluster, which has the highest quality range. This represents the average processing quality corresponding to the i-th value range of the e-th processing parameter and the j-th value range of the f-th processing parameter. This represents the number of identical processing parameter sequences when the value range of the e-th processing parameter is the same as the value range of the i-th processing parameter, and the value range of the f-th processing parameter is the same as the value range of the j-th processing parameter. This represents the average processing quality corresponding to both the e-th and f-th processing parameters within their respective value ranges of i and j. This represents the processing quality where the value of the e-th processing parameter in the t-th processing parameter sequence is in the ith value range, and the value of the f-th processing parameter is in the j-th value range. This indicates the degree of influence of the i-th value range of the e-th processing parameter and the j-th value range of the f-th processing parameter on the processing quality.

[0014] Optionally, the importance of each processing parameter's influence on the corresponding processing quality is determined based on the degree of influence of the two processing parameters on the processing quality and the degree of influence of one of the two processing parameters on the processing quality, including: The importance of each processing parameter to the processing quality can be calculated using the following formula: Where c represents the number of processing parameters in a processing parameter sequence. This represents the number of values ​​in the range of the e-th processing parameter in the processing parameter sequence. This represents the number of possible values ​​for the f-th processing parameter. This indicates the degree of influence of the i-th value range of the e-th processing parameter and the j-th value range of the f-th processing parameter on the processing quality. This indicates the degree of influence of the f-th processing parameter on the processing quality. This indicates the importance of the e-th processing parameter in terms of its impact on processing quality.

[0015] Optionally, the necessity for optimizing each processing parameter is determined based on the importance of each processing parameter and the average processing quality corresponding to each processing parameter, including: Calculate the product of the importance of each processing parameter, the range of values ​​for each processing parameter, and the number of ranges of values ​​for each processing parameter; Calculate the product and the average processing quality of the first-class cluster with the most data points. The ratio between them determines the necessity of optimizing each processing parameter.

[0016] The beneficial effects of this application, distinct from existing technologies, include a big data analysis method for production management. This method involves processing a historical set of processing parameters during ammonium sulfate production to obtain the processing quality corresponding to each processing parameter sequence within the historical set. The historical set of processing parameters includes multiple processing parameter sequences, and each sequence includes multiple processing parameters. Based on the relationship between each processing parameter sequence and its corresponding processing quality, the optimization necessity of each processing parameter within the sequence is determined. If the optimization necessity of a processing parameter exceeds a preset threshold, then the processing parameter is determined to require optimization. This method considers the relationships between various processing parameters, making the relationship between processing parameters and processing quality clearer, and can effectively optimize processing parameters and improve processing quality. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating an embodiment of a big data analysis method for production management according to the present invention. Figure 2 for Figure 1 A flowchart illustrating an embodiment of step S12; Figure 3 for Figure 2 A schematic flowchart of one embodiment of step S21. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0019] The specific scenario addressed by this invention is: extracting acidic gas containing NH3, H2S, CO2, etc. from coal gas, treating the acidic gas with sulfuric acid to obtain ammonium sulfate crystals. Further analysis and processing yields ammonium sulfate products. The following detailed description of this application is provided in conjunction with the accompanying drawings and embodiments.

[0020] Please see Figure 1 , Figure 1 This is a flowchart illustrating a first embodiment of a big data analysis method for production management according to the present invention, specifically including: Step S11: Process the historical processing parameter set in the ammonium sulfate production process to obtain the processing quality corresponding to each processing parameter sequence in the historical processing parameter set; the historical processing parameter set includes multiple processing parameter sequences, and each processing parameter sequence includes multiple processing parameters.

[0021] Specifically, an IoT big data system is used to obtain historical processing parameter sets during the production of ammonium sulfate. These sets include multiple processing parameter sequences, each containing several parameters: saturator inlet acid gas temperature, mother liquor temperature, mixer resistance, mother liquor acidity, stirring air volume, stirring air temperature, feed-to-mother liquor ratio, and ammonium sulfate consumption. These processing parameter sequences are denoted as... ,...,..., ,in, This indicates the acid gas temperature at the saturator inlet. Indicates the temperature of the mother liquor. The value of c represents the amount of ammonium sulfate consumed. In this invention, c=8 represents the number of processing parameters in the processing parameter sequence.

[0022] Obtain the ammonium sulfate quality description parameters corresponding to each processing parameter sequence. These parameters include nitrogen content, free acid ratio, and moisture ratio. Use a preset model algorithm to process the nitrogen content, free acid ratio, and moisture ratio to obtain the processing quality corresponding to each processing parameter sequence.

[0023] Specifically, the ammonium sulfate quality description parameters for each processing parameter sequence—namely, nitrogen content, free acid ratio, and moisture ratio—are input into the network to obtain the corresponding ammonium sulfate quality evaluation score M (M∈[0,1]). The quality evaluation score characterizes the processing quality corresponding to the processing parameter sequence; that is, the larger the calculated M, the better the ammonium sulfate quality. The network input consists of the ammonium sulfate quality description parameters corresponding to each historical processing parameter sequence, and the network output is the ammonium sulfate quality evaluation score M. This network is trained using a cross-entropy loss function through manual annotation.

[0024] Step S12: Determine the necessity of optimizing each processing parameter in the processing parameter sequence based on the relationship between each processing parameter sequence and the corresponding processing quality.

[0025] Since the production of ammonium sulfate involves multiple processing and extraction processes, each of which affects the final quality, this invention analyzes the degree of influence of processing parameters on the quality and the corresponding quality of processing parameters to determine the necessity of improving processing parameters, and then optimizes the corresponding processing parameters.

[0026] For specific details, please refer to... Figure 2 Step S12 specifically includes: Step S21: Calculate the importance of each processing parameter in the processing parameter sequence.

[0027] In one embodiment, the importance of each processing parameter in the processing parameter sequence is calculated based on the degree of influence of each processing parameter in the processing parameter sequence on the processing quality and the importance of the degree of influence of each processing parameter in the processing parameter sequence on the corresponding processing quality.

[0028] Furthermore, please combine Figure 3 Step S21 specifically includes: Step S31: Determine the degree of influence of each processing parameter on the processing quality based on the degree of change in processing quality corresponding to each processing parameter.

[0029] Specifically, determine the importance of the current processing parameter within each value range. Based on the importance of the current processing parameter value within each value range, determine the degree of influence of the current processing parameter on the processing quality.

[0030] In one specific embodiment, the degree of influence of the current processing parameter value 'a' on the processing quality is determined based on the difference between the processing quality when the current processing parameter is value 'a' and the average processing quality corresponding to value 'a'. Assume the current processing parameter is the stirring air temperature, with values ​​a, a1, a2, a1, a2, a, and a, and processing qualities am1, a1m1, a2m1, a1m2, a2m2, and am2, respectively. Subtracting the average processing quality (am1 + am2) / 2 corresponding to value 'a' from the processing quality am1 when the stirring air temperature is value 'a' yields the difference between the processing quality when the current processing parameter is value 'a' and the average processing quality corresponding to value 'a'. The degree of influence of the current processing parameter value 'a' on the processing quality is then determined based on this calculated difference.

[0031] Specifically, the relationship between a single processing parameter and processing quality is obtained by analyzing a single processing parameter and then, based on the correlation of the product quality corresponding to the same parameter among the numerous data obtained, the greater the correlation, the greater the impact of the current processing parameter on the quality at the time of analysis, and the better the processing quality of the corresponding processing parameter.

[0032] Since the final product quality is influenced by numerous factors, when analyzing a single processing parameter, we can analyze its impact on the product's processing quality by considering the degree of change in the processing quality while keeping the single parameter constant. A greater degree of impact corresponds to a smaller change in the processing quality. Therefore, we can calculate the degree of impact of processing parameters on processing quality based on this principle: in, This indicates the amount of analyzable data obtained when the processing parameter value is 'a'. This indicates that when the value of the b-th processing parameter is a, the s-th processing quality is obtained from the analyzable data. This represents the average machining quality when the value of the b-th machining parameter is a. That is, when the desired machining parameter is... The smaller the difference, that is, the smaller the difference in processing quality when the value of the b-th processing parameter is a, the greater the influence of the value of the b-th processing parameter being a on the processing quality.

[0033] In one embodiment, the relationship between the value of the processing parameter and its influence on processing quality may change with the value of the processing parameter, and abrupt changes may occur. Therefore, a coordinate system is established with the current value of the processing parameter as the horizontal axis and the influence of each value on processing quality as the vertical axis. The data points on the coordinate system are clustered to obtain multiple first clusters. Each first cluster represents the influence of each value range of the current processing parameter on processing quality. Specifically, the DBSCAN algorithm is used to cluster the image curve coordinates to obtain the influence of each value range of the current processing parameter on processing quality. That is, based on the clustering results, the difference between the current value of the processing parameter and the change in the influence of the current value on processing quality is obtained. When the number of first clusters in the clustering results is 1, it indicates that under the current analysis, the relationship between the value of the processing parameter and the influence of the processing quality remains stable throughout the entire value range.

[0034] Obtain the average value of the impact of the processing parameter value range on processing quality for each first cluster. Simultaneously, obtain the importance of each value range for the corresponding processing parameter; that is, the more important each value range of the processing parameter is to the current processing parameter, the more the current processing parameter needs to move closer to its value range.

[0035] Furthermore, the processing quality is clustered to obtain multiple second-class clusters, each representing processing parameters within different quality ranges. The importance of each value range of the current processing parameter is calculated based on the average influence of the current processing parameter on the processing quality within its value range, the average processing quality corresponding to the current processing parameter within its value range, the number of processing parameter sequences in the second-class cluster with the highest quality range, and the number of processing parameter sequences in the second-class cluster with the highest quality range corresponding to each value range of the current processing parameter. The method for calculating the importance of each range of processing parameters is as follows: Let the current processing parameter be the b-th processing parameter. This represents the average value of the influence of the b-th processing parameter within the x-th value range on the processing quality. This represents the average processing quality corresponding to the b-th processing parameter within the x-th value range. This indicates the number of processing parameter sequences in the second type of cluster, which has the highest quality range. This represents the number of processing parameter sequences in the second cluster with the highest quality range that correspond to each value range of the current processing parameter; This indicates the importance of the b-th processing parameter within the x-th value range. In other words, the better and more stable the desired processing quality, and the larger the ratio of the number of times the corresponding processing parameter's quality is optimized to the total number of times the processing quality is optimized, the higher the importance of the current processing parameter's corresponding category range.

[0036] It should be noted that if there are 100 processing parameters and the processing quality distribution is 0-1, clustering the processing quality yields 4 classes, i.e., 4 second-class clusters: [0-0.3], [0.3-0.55], [0.55-0.78], and [0.78-1], corresponding to occurrence frequencies of 20, 30, 30, and 20 respectively. The highest frequency among these is 20. Therefore, in the above formula, If the b-th processing parameter, such as temperature, has a value range of 20-25℃, and the number of processing cycles is counted, for example, 60, then the corresponding processing quality is obtained. The cycle that best represents the processing quality is the one within the range of [0.78-1]. This number is the [number of cycles]. .

[0037] The importance of each value range of the current processing parameter is calculated using the above method. The average value of the importance of the current processing parameter in each value range is determined as the degree of influence of the current processing parameter on the processing quality.

[0038] Specifically, the calculation method for the degree of influence of current processing parameters on processing quality is as follows: Where n represents the number of the first clusters after clustering the b-th processing parameter. This indicates the importance of the b-th processing parameter being within the x-th value range. This indicates the degree of influence of the b-th processing parameter on the processing quality. That is, the greater the importance of each value range of the b-th processing parameter to the final processing quality, and the better the corresponding processing quality, the greater the proportion of the value range, indicating that the current processing parameter has a greater influence on the final processing quality.

[0039] Step S32: Calculate the importance of the influence of each processing parameter in the processing parameter sequence on the corresponding processing quality.

[0040] Specifically, the influence of two processing parameters on processing quality is calculated; based on the influence of the two processing parameters on processing quality and the influence of one of the two processing parameters on processing quality, the importance of the influence of each processing parameter on the corresponding processing quality is determined.

[0041] In one embodiment, the influence of two processing parameters on processing quality is calculated using the following formula: in, This represents the number of times that the i-th value range of the e-th processing parameter and the j-th value range of the f-th processing parameter in the processing parameter sequence simultaneously produce high-quality products. This indicates the number of processing parameter sequences in the second type of cluster, which has the highest quality range. This represents the average processing quality corresponding to the i-th value range of the e-th processing parameter and the j-th value range of the f-th processing parameter. This represents the number of identical processing parameter sequences when the value range of the e-th processing parameter is the same as the value range of the i-th processing parameter, and the value range of the f-th processing parameter is the same as the value range of the j-th processing parameter. This represents the average processing quality corresponding to both the e-th and f-th processing parameters within their respective value ranges of i and j. This represents the processing quality where the value of the e-th processing parameter in the t-th processing parameter sequence is in the ith value range, and the value of the f-th processing parameter is in the j-th value range. This indicates the degree of influence of the i-th value range of the e-th processing parameter and the j-th value range of the f-th processing parameter on the processing quality.

[0042] In other words, the smaller the difference between the processing quality and its corresponding average value under the corresponding range of each processing parameter, and the larger the ratio of the number of times the corresponding processing parameter quality is optimized to the total number of times the processing quality is optimized, the better the corresponding processing quality is. This indicates that the two processing parameters have a higher degree of influence on the product processing quality.

[0043] The importance of each processing parameter's influence on the corresponding processing quality is determined based on the degree of influence of two processing parameters on the processing quality and the degree of influence of one of the two processing parameters on the processing quality, including: The importance of each processing parameter to the processing quality is calculated using the following formula. : Where c represents the number of processing parameters in a processing parameter sequence. This represents the number of values ​​in the range of the e-th processing parameter in the processing parameter sequence. This represents the number of possible values ​​for the f-th processing parameter. This indicates the degree of influence of the i-th value range of the e-th processing parameter and the j-th value range of the f-th processing parameter on the processing quality. This indicates the degree of influence of the f-th processing parameter on the processing quality. This indicates the importance of the e-th processing parameter in terms of its impact on processing quality.

[0044] The greater the influence of the i-th value range of the e-th processing parameter and the j-th value range of the f-th processing parameter on the processing quality, the better the desired result. The larger the value, the lower the influence of the f-th processing parameter on the processing quality, i.e., the desired result. The smaller the value, the greater the importance of the current processing parameters in influencing the processing quality.

[0045] Step S33: Based on the degree of influence of each processing parameter in the processing parameter sequence on the processing quality, and the importance of the degree of influence of each processing parameter in the processing parameter sequence on the corresponding processing quality, calculate the importance of each processing parameter in the processing parameter sequence.

[0046] Specifically, the method for calculating the importance of processing parameters is as follows: ;in, This indicates the importance of the e-th processing parameter. This indicates the importance of the e-th processing parameter in terms of its impact on processing quality. This indicates the degree of influence of the e-th processing parameter on the processing quality.

[0047] Step S22: Calculate the average value of the processing quality corresponding to each processing parameter.

[0048] Specifically, the average processing quality of the first cluster with the most data points in the first cluster is determined as the average processing quality corresponding to the current processing parameters.

[0049] Step S23: Determine the necessity of optimizing each processing parameter based on the importance of each processing parameter and the average processing quality corresponding to each processing parameter.

[0050] Specifically, calculate the product of the importance of each processing parameter, the range of values ​​for each processing parameter, and the number of ranges for each processing parameter; then calculate the product and the average processing quality of the first cluster with the most data points in the first cluster. The ratio between them determines the necessity of optimizing each processing parameter.

[0051] In one embodiment, the calculation method for optimization necessity is as follows: in, This indicates the necessity of optimizing the e-th processing parameter. This indicates the importance of the e-th processing parameter. This indicates the range of values ​​for the e-th processing parameter. This represents the number of possible values ​​for the e-th processing parameter. This represents the average processing quality of the first cluster with the most data points obtained after clustering the e-th processing parameter. In other words, the greater the importance of the corresponding processing parameter, the higher the processing quality. The larger the value, the wider the range of values ​​for the corresponding parameter, the more ranges of values ​​for the corresponding processing parameter, and the smaller the average processing quality of the first cluster with the most data points obtained after clustering. This indicates that the current processing parameter corresponds to a process that needs improvement. This indicates the stability of the processing parameters; the larger the value, the less stable the processing.

[0052] Step S13: If the necessity of optimization corresponding to the processing parameter is greater than a preset threshold, then it is determined that the processing parameter needs to be optimized and adjusted.

[0053] Set a preset threshold ψ=0.65, and normalize the necessity of optimizing the e-th processing parameter obtained above. If the normalized result... If the value exceeds a preset threshold, it can be considered that the current processing parameter, for example, the processing technology corresponding to the e-th processing parameter, needs to be optimized.

[0054] This invention analyzes processing parameters to determine the impact of different parameter values ​​on processing quality, thereby establishing the relationship between processing parameters and final processing quality. Furthermore, it analyzes the interrelationships between various process flows during processing to understand the relationships between individual processing parameters and processing quality. This allows for a more detailed analysis of the relationship between individual processing parameters and processing quality. By considering the relationships between processing parameters, this invention provides a more detailed and clear understanding of the connections between processing parameters and processing quality. This significantly increases the likelihood of selecting key processing techniques during optimization analysis, greatly enhancing the improvement in processing quality resulting from process optimization.

[0055] The above are merely embodiments of this application and do not limit the scope of this application. Any equivalent structural or procedural transformations made based on the description and drawings of this application, or direct or indirect applications in other related technical fields, are similarly included within the protection scope of this application.

Claims

1. A big data analysis method for production management, characterized in that, include: The historical processing parameter set in the ammonium sulfate production process is processed to obtain the processing quality corresponding to each processing parameter sequence in the historical processing parameter set; the historical processing parameter set includes multiple processing parameter sequences, and each processing parameter sequence includes multiple processing parameters. The necessity of optimizing each processing parameter in the processing parameter sequence is determined based on the relationship between each processing parameter sequence and the corresponding processing quality. If the necessity for optimization corresponding to the processing parameter is greater than a preset threshold, then it is determined that the processing parameter needs to be optimized and adjusted.

2. The big data analysis method for production management according to claim 1, characterized in that, The historical processing parameter set in the ammonium sulfate production process is processed to obtain the processing quality corresponding to each processing parameter sequence in the historical processing parameter set, including: Obtain the ammonium sulfate quality description parameters corresponding to each processing parameter sequence. The quality description parameters include: nitrogen content, free acid ratio, and moisture ratio. The nitrogen content, free acid ratio, and moisture ratio are processed using a preset model algorithm to obtain the processing quality corresponding to each processing parameter sequence. The processing parameters in the processing parameter sequence include: saturator inlet acid gas temperature, mother liquor temperature, mixer resistance, mother liquor acidity, stirring air volume, stirring air temperature, feed mother liquid ratio, and ammonium sulfate consumption.

3. The big data analysis method for production management according to claim 1, characterized in that, The necessity of optimizing each processing parameter in the processing parameter sequence is determined based on the relationship between each processing parameter sequence and the corresponding processing quality, including: Calculate the importance of each processing parameter in the processing parameter sequence; Calculate the average processing quality corresponding to each processing parameter; The necessity of optimizing each processing parameter is determined based on the importance of each processing parameter and the average processing quality corresponding to each processing parameter.

4. The big data analysis method for production management according to claim 3, characterized in that, Calculate the importance of each machining parameter in the machining parameter sequence, including: The degree of influence of each processing parameter on the processing quality is determined based on the degree of change in the processing quality corresponding to each processing parameter. Calculate the importance of the influence of each processing parameter in the processing parameter sequence on the corresponding processing quality; Based on the degree of influence of each processing parameter in the processing parameter sequence on the processing quality, and the importance of the degree of influence of each processing parameter in the processing parameter sequence on the corresponding processing quality, the importance of each processing parameter in the processing parameter sequence is calculated.

5. The big data analysis method for production management according to claim 4, characterized in that, The degree of influence of each processing parameter on the processing quality is determined based on the degree of change in processing quality corresponding to each processing parameter, including: Determine the importance of the current processing parameters across their respective value ranges; The degree of influence of the current processing parameters on the processing quality is determined based on the importance of the current processing parameter values ​​within each range.

6. The big data analysis method for production management according to claim 5, characterized in that, Determine the importance of the current processing parameters across different value ranges, including: The degree of influence of the current processing parameter value a on the processing quality is determined based on the difference between the processing quality when the current processing parameter is value a and the average processing quality corresponding to value a. A coordinate system is established with the current processing parameter values ​​as the horizontal axis and the degree of influence of each value on the processing quality as the vertical axis. The data points on the coordinate system are clustered to obtain multiple first-class clusters. Each first-class cluster represents the degree of influence of each range of current processing parameter values ​​on the processing quality. The processing quality is clustered to obtain multiple second-class clusters, each of which represents processing parameters for different quality ranges; The importance of each value range of the current processing parameter is calculated based on the average value of the influence of the current processing parameter on the processing quality within the value range, the average processing quality corresponding to the current processing parameter within the value range, the number of processing parameter sequences in the second category with the highest quality range, and the number of processing parameter sequences in the second category with the highest quality range corresponding to each value range of the current processing parameter. The degree of influence of the current processing parameters on processing quality is determined based on the importance of the current processing parameter values ​​within each value range, including: The average value of the importance of the current processing parameter across its various value ranges is determined as the degree of influence of the current processing parameter on the processing quality.

7. The big data analysis method for production management according to claim 4, characterized in that, Calculate the importance of each processing parameter in the processing parameter sequence on the corresponding processing quality, including: Calculate the degree of influence of two processing parameters on processing quality; The importance of each processing parameter's influence on the corresponding processing quality is determined based on the degree of influence of the two processing parameters on the processing quality and the degree of influence of one of the two processing parameters on the processing quality.

8. The big data analysis method for production management according to claim 7, characterized in that, Calculate the degree of influence of two machining parameters on machining quality, including: The influence of two machining parameters on machining quality can be calculated using the following formula: in, This represents the number of times that the i-th value range of the e-th processing parameter and the j-th value range of the f-th processing parameter in the processing parameter sequence simultaneously produce high-quality products. This indicates the number of processing parameter sequences in the second type of cluster with the highest quality range. This represents the average processing quality corresponding to the i-th value range of the e-th processing parameter and the j-th value range of the f-th processing parameter. This represents the number of identical processing parameter sequences when the value range of the e-th processing parameter is the same as the value range of the i-th processing parameter, and the value range of the f-th processing parameter is the same as the value range of the j-th processing parameter. This represents the average processing quality corresponding to the e-th and f-th processing parameters within their respective i-th and j-th value ranges. This represents the processing quality where the value of the e-th processing parameter in the t-th processing parameter sequence is in the ith value range, and the value of the f-th processing parameter is in the j-th value range. This indicates the degree of influence of the i-th value range of the e-th processing parameter and the j-th value range of the f-th processing parameter on the processing quality.

9. A big data analysis method for production management according to claim 7, characterized in that, The importance of each processing parameter's influence on the corresponding processing quality is determined based on the degree of influence of two processing parameters on the processing quality and the degree of influence of one of the two processing parameters on the processing quality, including: The importance of each processing parameter to the processing quality can be calculated using the following formula: Where c represents the number of processing parameters in a processing parameter sequence. This represents the number of values ​​in the range of the e-th processing parameter in the processing parameter sequence. This represents the number of possible values ​​for the f-th processing parameter. This indicates the degree of influence of the i-th value range of the e-th processing parameter and the j-th value range of the f-th processing parameter on the processing quality. This indicates the degree of influence of the f-th processing parameter on the processing quality. This indicates the importance of the e-th processing parameter in terms of its impact on processing quality.

10. A big data analysis method for production management according to claim 3, characterized in that, The necessity of optimizing each processing parameter is determined based on its importance and the average processing quality corresponding to that parameter, including: Calculate the product of the importance of each processing parameter, the range of values ​​for each processing parameter, and the number of ranges of values ​​for each processing parameter; Calculate the product and the average processing quality of the first-class cluster with the most data points. The ratio between them determines the necessity of optimizing each processing parameter.

Images