A method for determining the target value of decomposition furnace outlet temperature based on dual drive of data and rules

Abstract

The present invention discloses a method for determining a target value of a decomposition furnace outlet temperature based on dual-drive of data and rules, comprising: (1) collecting characteristic data in a historical cement production process to obtain a plurality of analysis samples to form a sample set; (2) using a clustering method to preliminarily screen the sample set based on the target value of the decomposition furnace outlet temperature; (3) using a clustering method to sub-classify each grade of samples based on the characteristic data of each grade of samples; (4) establishing a result classification table based on the results of the classification; (5) obtaining characteristic data of the current working condition and determining a target value of the decomposition furnace outlet temperature based on a determination rule, and adjusting the decomposition furnace outlet temperature based on the determined target value of the decomposition furnace outlet temperature; the present invention can obtain an accurate target value of the decomposition furnace outlet temperature.

Description

Technical Field

[0001] The invention relates to the technical field of intelligent control of cement production, and in particular to a method for determining a target value of a decomposition furnace outlet temperature based on dual-drive of data and rules. Background Art

[0002] The outlet temperature of the calciner is a very important process parameter in cement production, which reflects the decomposition rate of raw materials entering the kiln. If the outlet temperature of the calciner is too low, it will easily lead to a low decomposition rate of raw materials entering the kiln, increasing the burden on the kiln system; if the outlet temperature of the calciner is too high, it will not only consume too much tail coal and cause energy waste, but also easily cause the preheater to be blocked by crusting, thus affecting the operation of the kiln system. Therefore, it is of great significance to give the optimal setting value of the calciner temperature under the current working conditions.

[0003] The existing document "Study on Optimization Control of Decomposition Furnace Link in Cement Production Process" records a decomposition furnace temperature optimization setting model, which consists of a decomposition furnace temperature pre-setting model based on LS-SVM (least squares support vector machine), a temperature setting compensation model based on an expert system, and a temperature setting correction model based on a fuzzy system. The decomposition furnace outlet temperature target value is obtained through this model. However, the decomposition furnace temperature pre-setting based on LS-SVM in this scheme does not have corresponding label data, and the LS-SVM effect depends on the selection of kernel functions, and the parameter adjustment is complex, and the process of obtaining the decomposition furnace outlet temperature target value is relatively complex. In the prior art, there is an algorithm model that uses historical data based on deep learning to predict the decomposition furnace outlet temperature target value, but such methods generally require manual labeling of historical data to determine the decomposition furnace outlet temperature target value, and do not consider the data of real-time working conditions. The process of obtaining the decomposition furnace outlet temperature target value is complex and the result is not accurate. Summary of the invention

[0004] Purpose of the invention: The present invention provides a method for determining the target value of the decomposition furnace outlet temperature based on dual drive of data and rules, which learns from historical characteristic data and sets adjustment rules considering the characteristic data of real-time working conditions to obtain an accurate target value of the decomposition furnace outlet temperature.

[0005] Technical solution: The present invention provides a method for determining the target value of the outlet temperature of a decomposition furnace based on dual drive of data and rules, which specifically includes the following steps:

[0006] (1) Collect characteristic data in the historical cement production process, take the clinker sampling time corresponding to the determination of the free calcium content of the clinker as the benchmark, select characteristic data within a period of time to form an analysis sample; obtain multiple analysis samples to form a sample set, pre-process the characteristic data in the sample set and obtain the characteristic data change rate; the characteristic data include operating condition variables: target value of decomposition furnace outlet temperature, feedback value of decomposition furnace outlet temperature, feed amount into the kiln, three rates of raw material into the kiln, secondary air temperature, tertiary air temperature, kiln head colorimetric pyrometer temperature, kiln tail smoke chamber NOx; control variables: head coal consumption, tail coal consumption; result variable: free calcium content;

[0007] (2) Based on the target value of the decomposition furnace outlet temperature, the k-means clustering method is used to perform a preliminary screening of the sample set, divide the samples into multiple levels, and calculate the median of the target value of the decomposition furnace outlet temperature in each level of the sample;

[0008] (3) Based on the characteristic data of each sample, the k-means clustering method is used to sub-classify each sample to obtain multiple categories of samples, and the median of the outlet temperature feedback value of each category of samples is calculated;

[0009] (4) Establishing a result classification table based on the classification results;

[0010] (5) obtaining characteristic data of the current working condition and the characteristic data change rate and determining the target value of the decomposition furnace outlet temperature according to the determination rule, and adjusting the decomposition furnace outlet temperature according to the determined target value of the decomposition furnace outlet temperature;

[0011] The determination rules include: (5.1) obtaining the median of the decomposition furnace outlet temperature feedback value of the current working condition and calculating the distance between it and the median of all the decomposition furnace outlet temperature target values ​​in the result setting and classification table, and selecting the gear with the smallest distance as the gear of the current working condition; (5.2) calculating the distance between the characteristic data of the current working condition and the characteristic data of each type of sample in the gear of the current working condition, and taking the classification corresponding to the smallest distance as the classification of the current working condition; (5.3) determining the target gear and sub-classification of the expected adjustment according to the free calcium content of the current working condition, the three-rate value of the raw material entering the kiln, the characteristic data change rate, the ratio of the head and tail coal consumption, and the median of the decomposition furnace outlet temperature feedback value, and taking the median of the outlet temperature feedback value corresponding to the target gear and sub-classification of the expected adjustment as the decomposition furnace outlet temperature target value.

[0012] Furthermore, step (5.3) specifically includes the following steps:

[0013] (5.3.1) Determine whether the characteristic data of the current working condition are within the preset range. If they are within the preset range, it is considered to be a normal working condition and go to step (5.3.2). Otherwise, go to step (5.3.3);

[0014] (5.3.2) Set adjustment conditions and rules for the gear and classification for free calcium content, three rates of raw materials entering the kiln, characteristic data change rate, head and tail coal usage ratio, and median of the outlet temperature feedback value of the decomposition furnace; in the result classification table, the samples of each gear are arranged from small to large according to the median of the outlet temperature target value, and the samples of each type are arranged from small to large according to the median of the outlet temperature feedback value;

[0015] The judgment is based on the obtained free calcium content of the current working condition, the three rates of raw materials entering the kiln, the characteristic data change rate, the ratio of head and tail coal consumption, and the median of the decomposition furnace outlet temperature feedback value. If there is a variable in each variable that meets the adjustment condition and the adjustment rule is to find the subcategory closest to the characteristic data of the current working condition in all categories of the previous category and the previous gear of the category where the current working condition is located, the count is increased by 1; if there is a variable in each variable that meets the adjustment condition and the adjustment rule is to find the subcategory closest to the characteristic data of the current working condition in all categories of the subsequent subcategory and the next gear of the category where the current working condition is located, the count is reduced by 1; if the free calcium content meets the adjustment condition and the adjustment rule is not to make adjustments, the count is 0; the front category refers to the category that is in the same gear as the category where the current working condition is located and the median of the outlet temperature feedback value is less than the median of the outlet temperature feedback value corresponding to the category; the back category refers to the category that is in the same gear as the category where the current working condition is located and the median of the outlet temperature feedback value is greater than the median of the outlet temperature feedback value corresponding to the category;

[0016] According to the counting accumulation result, if the accumulated count value is greater than or equal to 0, the final adjustment rule is determined to be to find the subcategory closest to the characteristic data of the current working condition in the previous category of the current working condition and all categories of the previous gear as the target gear and subcategory for expected adjustment; if the accumulated count value is less than 0, the final adjustment rule is determined to be to find the subcategory closest to the characteristic data of the current working condition in the subsequent category of the current working condition and all categories of the next gear as the target gear and subcategory for expected adjustment;

[0017] If the free calcium content and the three rates of raw materials entering the kiln of the current working condition are not obtained, and the characteristic data change rate, the head and tail coal usage ratio, and the decomposition furnace outlet temperature feedback value do not meet the adjustment conditions, then the final adjustment rule is to find the sub-category closest to the characteristic data of the current working condition in all the categories of the current working condition and the previous gear as the target gear and sub-category for the expected adjustment;

[0018] (5.3.3) The final adjustment rule is to find the sub-category closest to the characteristic data of the current working condition in all the categories preceding the category of the current working condition and the upper N gears or lower N gears as the target gear and sub-category for the expected adjustment.

[0019] Further, the adjustment conditions and adjustment rules for each variable gear and classification in step (5.3.2) are specifically as follows:

[0020] The adjustment conditions and adjustment rules for the free calcium content gear and classification are as follows: when A < fCao < B, no adjustment is made; when fCao ≤ A, search for the subclass with the closest characteristic data to the current working condition among the classifications before the current working condition's classification and all classifications in the previous gear; when fCao ≥ B, search for the subclass with the closest characteristic data to the current characteristic among the subclasses after the current working condition's classification and all classes in the next gear; fCao represents the average free calcium content of the current working condition, and both A and B are preset thresholds;

[0021] The adjustment conditions and adjustment rules for the three-ratio value gear and classification of the raw meal entering the kiln are as follows: when the KH and SM of the current working condition increase, search for the subclass with the closest characteristic data to the current working condition among the subclasses after the current working condition's classification and all classes in the next gear; when the IM of the current working condition increases, search for the subclass with the closest characteristic data to the current working condition among the subclasses before the current working condition's classification and all classes in the previous gear;

[0022] The adjustment conditions and adjustment rules for the proportion gear of the head and tail coal consumption are as follows: when the proportion of the head and tail coal consumption of the current working condition decreases, search for the subclass with the closest characteristic data to the current working condition among the subclasses after the current working condition's classification and all classes in the next gear;

[0023] The adjustment conditions and adjustment rules for the characteristic data change rate gear and classification are as follows: when the characteristic data change rate of the current working condition decreases at a certain rate, search for the subclass with the closest characteristic data to the current working condition among the subclasses after the current working condition's classification and all classes in the next gear;

[0024] The adjustment conditions and adjustment rules for the median gear of the decomposition furnace outlet temperature feedback value and classification are as follows: when the median of the decomposition furnace outlet temperature feedback value of the current working condition is lower than the median M℃ of the decomposition furnace outlet temperature feedback value of the current working condition's classification, search for the subclass with the closest characteristic data to the current working condition among the subclasses after the current working condition's classification and all classes in the next gear.

[0025] Further, the adjustment timing for adjusting the decomposition furnace outlet temperature according to the determined decomposition furnace outlet temperature target value in step (5) includes fixed-cycle triggering or event triggering; the fixed-cycle triggering means setting a fixed cycle, determining the decomposition furnace outlet temperature target value at the fixed cycle point and adjusting the decomposition furnace outlet temperature according to the decomposition furnace outlet temperature target value; the event triggering is to judge whether the characteristic data change rate exceeds a preset value within a period of time. If it exceeds the preset value, it is considered that an event has occurred, and the decomposition furnace outlet temperature target value is determined according to the working condition characteristic data of this period of time and the decomposition furnace outlet temperature is adjusted according to the decomposition furnace outlet temperature target value.

[0026] Furthermore, the pretreatment in step (1) includes:

[0027] (1.1) Retain the sample data with free calcium content in the range of A% to B%;

[0028] (1.2) Use the 3σ rule to eliminate abnormal sample data in the sample set;

[0029] (1.3) Perform moving average processing, median calculation, slope calculation and normalization processing on the sample data after removing outliers;

[0030] (1.4) Multiply the processed sample data by the weights respectively.

[0031] Furthermore, the data in the result classification table in step (4) is automatically iterated or manually adjusted during the cement production operation process.

[0032] In addition, the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program. A computer-readable storage medium stores a computer program, wherein the computer program implements the steps of the above method when executed by the processor.

[0033] Beneficial effect: The present invention provides a method for determining the target value of the outlet temperature of a decomposition furnace based on dual drive of data and rules. Compared with the prior art, its significant advantage is: a method combining clustering and classification is used to analyze the characteristic data of historical samples, that is, firstly a preliminary screening and setting of the gears is performed based on the temperature target values ​​of historical samples, and then the samples in each gear are further subdivided and classified according to the historical characteristic data, and a table is established, and the characteristic data of the current working conditions are considered in combination with the table to obtain the corresponding gear and classification; then the gear and classification are further adjusted in combination with the characteristic data of the current working conditions and the adjustment rules, and a more accurate outlet temperature feedback value is obtained according to the final gear and classification for temperature adjustment, thereby ensuring better production quality and saving energy at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 Shown is a flow chart of the method described in the present invention. DETAILED DESCRIPTION

[0035] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0036] like Figure 1 As shown, the method for determining the target value of the outlet temperature of a decomposition furnace based on dual-drive of data and rules of the present invention firstly analyzes the historical data by machine learning method to provide a candidate set of target temperature values, and then determines the specific target temperature value based on the rules. Specifically, it includes the following steps:

[0037] Step 1: Collect characteristic data from the historical cement production process to obtain a sample data set. Specifically include:

[0038] (1) Selection of characteristic data. Specifically, it includes operating variables: target value of decomposition furnace outlet temperature, feedback value of decomposition furnace outlet temperature, feed amount into kiln, three rates of raw material into kiln, secondary air temperature, tertiary air temperature, kiln head colorimetric pyrometer temperature, kiln tail smoke chamber NOx; control variables: head coal consumption, tail coal consumption; result variables: free calcium content, etc.

[0039] (2) Sample set acquisition. Based on the clinker sampling time corresponding to the determination of clinker free calcium content, select the sample set t forward. specimen All variable values ​​within a time period constitute a sample, and multiple samples are obtained to form a sample set; the original feature data in each sample is a multidimensional time series, recorded as:

[0040]

[0041] In the formula, n represents the number of feature data; t specimen Generally, 30 to 60 minutes are used.

[0042] (3) Preprocess the data in the sample data set, specifically:

[0043] (3.1) Only retain the sample data with free calcium in the range of A% to B%;

[0044] (3.2) Use the 3σ rule to eliminate abnormal sample data in the sample set;

[0045] (3.3) Perform moving average processing on the sample data after removing outliers, and the averaging period can be set to 1 minute; calculate the median and its change rate of the feature data x after averaging processing, which is recorded as:

[0046]

[0047] In the formula, Corresponding to The rate of change of, that is, the rate of change of characteristic data.

[0048] All sample data are normalized element by element and the normalized data are multiplied by the weight, which is recorded as:

[0049]

[0050] In the formula, ω i ≥0, and

[0051] Step 2: Preliminary screening and setting of the target values ​​of the outlet temperature of the decomposition furnace is performed by using the k-means clustering method, and the number of clusters in each level can be determined by the elbow method.

[0052] Assume that the result is C = {C1, C2, ..., C k}, where C1 represents the first gear, C k Indicates the kth level; T i is the average target value of the outlet temperature of the sample decomposition furnace of the i-th gear, and assuming that T1 <T2<…<T k Under normal working conditions, the adjustment range of the decomposition furnace outlet temperature is generally not more than 15°C. Therefore, the number of grades k can also be selected based on experience, for example, k is 4.

[0053] Step 3: Subdividing and classifying. i The sample data is further clustered by k-means to form sub-categories. The number of sub-categories in each level can be determined by the elbow method.

[0054] Assume that the result is The median of the outlet temperature feedback value and the median of free calcium of the sample of the i-th class and the l-th class are counted and recorded as T il and fCao il , and assuming Considering that the average temperature range of each level is 4 to 5°C, the number of categories in each level is considered to be set to 3 to 4.

[0055] Step 4: Create a result classification table. According to the results of step 2 and step 3, create a classification table, as shown in Table 1 below.

[0056] Table 1

[0057]

[0058] Step 5: Obtain the characteristic data of the current working condition and determine the target value of the decomposition furnace outlet temperature according to the determination rule, and adjust the decomposition furnace outlet temperature according to the determined target value of the decomposition furnace outlet temperature. Specifically including:

[0059] (1) Obtain the median of the decomposition furnace outlet temperature feedback value of the current working condition and calculate its distance from the median of all decomposition furnace outlet temperature target values ​​in the result setting classification table. Select the gear with the smallest distance as the gear of the current working condition, assuming it is C i kind;

[0060] (2) Calculate the distance between the characteristic data of the current working condition and the characteristic data of each type of sample in the gear where the current working condition is located, and take the classification corresponding to the minimum distance as the classification of the current working condition, assuming it is C il kind;

[0061] (3) To ensure the smooth operation of the decomposition furnace and kiln system, the target value of the decomposition furnace outlet temperature under normal conditions is adjusted gear by gear according to Table 2 below to determine the target gear and sub-classification of the expected adjustment. When abnormal conditions occur, cross-gear switching is allowed, or manual intervention is performed directly. The basis for judging whether it is a normal operating condition is whether the current operating condition characteristic data are all within the preset range. If they are within the preset range, it is considered a normal operating condition, otherwise it is an abnormal situation. For example: when free calcium changes from qualified to unqualified and the deviation is greater than 0.5, the slope change rate of the characteristic data change rate is greater than the preset threshold.

[0062] Table 2

[0063]

[0064] In addition, the adjustment conditions and operation rules for the three raw material rate values, the head and tail coal ratio, and the characteristic data change rate data in Table 2 are also valid, and can be combined to form a variety of adjustment rules. For example: KH and SM are reduced, and the corresponding adjustment rule is: find the sub-category closest to the current feature in the sub-category in front of the current gear and all the categories in the previous gear, as the target sub-category. The threshold range of fCao can be adjusted according to production needs. In this embodiment, A is set to 0.8, B is set to 1.2, and M is set to 2°C.

[0065] Specifically, the target gear and sub-classification of expected adjustment are determined according to the free calcium content, three rates of raw materials entering the kiln, characteristic data change rate, head and tail coal usage ratio, and median of the decomposition furnace outlet temperature feedback value under current operating conditions, and the median of the outlet temperature feedback value corresponding to the target gear and sub-classification of expected adjustment is used as the target value of the decomposition furnace outlet temperature.

[0066] (3.1) Under normal working conditions. The free calcium content, three rates of raw materials entering the kiln, characteristic data change rate, head and tail coal usage ratio, and median of the decomposition furnace outlet temperature feedback value of the current working condition are used as the basis for judgment. If there is a variable in each variable that meets the adjustment conditions and the adjustment rule is to find the subcategory closest to the characteristic data of the current working condition in the previous category of the current working condition and all categories of the previous gear, the count is increased by 1; if there is a variable in each variable that meets the adjustment conditions and the adjustment rule is to find the subcategory closest to the characteristic data of the current working condition in the subcategory behind the current working condition and all categories of the next gear, the count is reduced by 1; if the free calcium content meets the adjustment conditions and the adjustment rule is to make no adjustment, the count is 0.

[0067] According to the counting accumulation result, if the count value is greater than 0 or equal to 0, the final adjustment rule is determined to be to find the subcategory closest to the characteristic data of the current working condition in all categories of the previous gear position and the previous gear position as the target gear position and subcategory for expected adjustment; if the count value is less than 0, the final adjustment rule is determined to be to find the subcategory closest to the characteristic data of the current working condition in all categories of the next gear position and the current gear position as the target gear position and subcategory for expected adjustment

[0068] If the free calcium content, the three rates of raw materials entering the kiln of the current operating conditions are not obtained and the characteristic data change rate, the head and tail coal consumption ratio, and the decomposition furnace outlet temperature feedback value do not meet the adjustment conditions, the final adjustment rule is to find the subcategory closest to the characteristic data of the current operating condition among all the categories of the current operating condition and the previous gear as the target gear and subcategory for the expected adjustment.

[0069] (3.2) Under abnormal working conditions. The final adjustment rule is to find the sub-category closest to the characteristic data of the current working condition in all the categories before the current working condition and the upper N gears or lower N gears as the target gear and sub-category for the expected adjustment. Or directly perform manual intervention to determine the final adjustment rule.

[0070] (4) The timing for adjusting the decomposition furnace outlet temperature according to the determined decomposition furnace outlet temperature target value includes fixed period triggering or event triggering.

[0071] (4.1) Fixed period trigger. Set the fixed period t period , determine the target value of the decomposition furnace outlet temperature at a fixed periodic point and adjust the decomposition furnace outlet temperature according to the target value of the decomposition furnace outlet temperature. period If it is too large, the adjustment may not be timely and cannot quickly respond to changes in kiln working conditions; if t period If it is too small, the tail coal adjustment effect will not be reflected in time, which may lead to over-adjustment and affect the stability of the kiln system. period Set it to about 10 minutes.

[0072] (4.2) Event triggering. For example: judging t observe The rate of change of characteristic data within a certain period of time k I Whether the absolute value of exceeds the preset value k th If it exceeds the preset value, it is deemed that an event has occurred. The decomposition furnace outlet temperature target value is determined according to the characteristic data in the current working conditions, and the decomposition furnace outlet temperature is adjusted according to the decomposition furnace outlet temperature target value.

[0073] In addition, the data of the historical samples collected above are continuously optimized, including automatic iterative optimization methods or manual intervention methods.

[0074] Automatic iteration: As new sample data is continuously generated during the operation of the decomposition furnace and kiln system, the older historical data is discarded and cluster samples are reorganized. For example, only the most recent 1,000 samples are used, and steps 2 and 3 are repeated to automatically update the cluster result table in step 4. Update cycle t cluster Can be set to 3 to 5 days.

[0075] Manual intervention: After the system recommends the target value of the decomposition furnace outlet temperature, if the operator thinks it is not appropriate, he can directly adjust it online to change the characteristic variables of the kiln system. Due to the automatic iteration function, the operator's manual intervention knowledge will be implicitly encoded into the clustering results, thereby guiding the subsequent automatic recommendation of the target value of the decomposition furnace outlet temperature.

[0076] Taking the production data of a factory from March 20 to June 19, 2021 as an example, the outlet temperature of the decomposition furnace is divided into 12 levels; by directly clustering the characteristic variables and adopting the present invention to first set 4 levels and divide each level into 3 categories, the comparison results of the target values ​​of the outlet temperature of the decomposition furnace (unit: ℃) are shown in Table 3 below.

[0077] Table 3

[0078] Serial number Direct Clustering Method Direct clustering method difference Method of the present invention The method of the present invention 1 879.15 — 876.74 — 2 879.18 0.03 877.07 0.32 3 879.50 0.31 877.18 0.11 4 879.52 0.02 877.67 0.49 5 879.60 0.08 877.80 0.13 6 879.96 0.36 879.45 1.64 7 880.24 0.28 879.65 0.20 8 880.40 0.17 879.96 0.31 9 881.06 0.65 880.13 0.17 10 882.29 1.23 882.14 2.01 11 882.37 0.09 882.30 0.16 12 882.56 0.19 882.96 0.67

[0079] According to the results in Table 3, the maximum and minimum temperature target values ​​obtained by the direct clustering method in the prior art differ by 3.41°C, and multiple differences are less than 0.1°C. The maximum and minimum temperature target values ​​obtained by the method of the present invention differ by 6.22°C, and the minimum difference is 0.11°C. In actual production, increasing the target temperature at the outlet of the decomposition furnace by only 0.1°C has almost no effect, and the maximum adjustment range of 3.41°C has limited adaptability to different working conditions. The maximum adjustment range of 6.22°C in the present application can better obtain a more accurate target temperature at the outlet of the decomposition furnace.

Claims

1. A method for determining the target value of the outlet temperature of a decomposition furnace based on dual drive of data and rules, characterized in that: The following steps are involved: (1) Collect characteristic data from the historical cement production process, take the clinker sampling time corresponding to the determination of the free calcium content of the clinker as a benchmark, select characteristic data within a period of time to form an analysis sample, obtain multiple analysis samples to form a sample set; pre-process the characteristic data in the sample set and obtain the characteristic data change rate; The characteristic data include working condition variables: target value of decomposition furnace outlet temperature, feedback value of decomposition furnace outlet temperature, feed amount into kiln, three rates of raw material into kiln, secondary air temperature, tertiary air temperature, kiln head colorimetric pyrometer temperature, kiln tail smoke chamber NOx; control variables: head coal consumption, tail coal consumption; Outcome variables: free calcium content; The pretreatment comprises: (1.1) Retain the sample data with free calcium content in the range of A% to B%; (1.2) Use the 3σ rule to eliminate abnormal sample data in the sample set; (1.3) Perform moving average processing, median calculation, slope calculation and normalization processing on the sample data after removing outliers; (1.4) Multiply the processed sample data by the weights respectively; (2) Based on the target value of the decomposition furnace outlet temperature, the k-means clustering method is used to perform a preliminary screening of the sample set, divide the samples into multiple levels, and calculate the median of the target value of the decomposition furnace outlet temperature in each level of the sample; (3) Based on the characteristic data of each sample, the k-means clustering method is used to sub-classify each sample to obtain multiple categories of samples, and the median of the outlet temperature feedback value of each category of samples is calculated; (4) Establishing a result classification table based on the classification results; (5) obtaining characteristic data of the current working condition and the characteristic data change rate and determining the target value of the decomposition furnace outlet temperature according to the determination rule, and adjusting the decomposition furnace outlet temperature according to the determined target value of the decomposition furnace outlet temperature; The determination rules include: (5.1) obtaining the median of the decomposition furnace outlet temperature feedback value of the current working condition and calculating the distance between it and the median of all the decomposition furnace outlet temperature target values ​​in the result setting and classification table, and selecting the gear with the smallest distance as the gear of the current working condition; (5.2) calculating the distance between the characteristic data of the current working condition and the characteristic data of each type of sample in the gear of the current working condition, and taking the classification corresponding to the smallest distance as the classification of the current working condition; (5.3) determining the target gear and sub-classification of the expected adjustment according to the free calcium content of the current working condition, the three-rate value of the raw material entering the kiln, the characteristic data change rate, the ratio of the head and tail coal consumption, and the median of the decomposition furnace outlet temperature feedback value, and taking the median of the outlet temperature feedback value corresponding to the target gear and sub-classification of the expected adjustment as the decomposition furnace outlet temperature target value.

2. The method for determining the target value of the decomposition furnace outlet temperature based on data and rule dual drive according to claim 1 is characterized in that: Step (5.3) specifically includes the following steps: (5.3.1) Determine whether the characteristic data of the current working condition are within the preset range. If they are within the preset range, it is considered to be a normal working condition and go to step (5.3.2). Otherwise, go to step (5.3.3); (5.3.2) Set adjustment conditions and rules for the gear and classification for free calcium content, three rates of raw materials entering the kiln, characteristic data change rate, head and tail coal usage ratio, and median of the outlet temperature feedback value of the decomposition furnace; in the result classification table, the samples of each gear are arranged from small to large according to the median of the outlet temperature target value, and the samples of each type are arranged from small to large according to the median of the outlet temperature feedback value; The obtained free calcium content of the current working condition, the three rates of raw materials entering the kiln, the characteristic data change rate, the ratio of head and tail coal consumption, and the median of the decomposition furnace outlet temperature feedback value are used as the judgment basis. If there is a variable in each variable that meets the adjustment condition and the adjustment rule is to find the subcategory closest to the characteristic data of the current working condition in all categories of the previous category and the previous gear of the category where the current working condition is located, the count is increased by 1; if there is a variable in each variable that meets the adjustment condition and the adjustment rule is to find the subcategory closest to the characteristic data of the current working condition in all categories of the subsequent subcategory and the next gear of the category where the current working condition is located, the count is reduced by 1; if the free calcium content meets the adjustment condition and the adjustment rule is not to make adjustments, the count is 0; the previous category refers to the category that is in the same gear as the category where the current working condition is located and the median of the outlet temperature feedback value is less than the median of the outlet temperature feedback value corresponding to the category; the subsequent subcategory refers to the category that is in the same gear as the category where the current working condition is located and the median of the outlet temperature feedback value is greater than the median of the outlet temperature feedback value corresponding to the category; According to the counting accumulation result, if the accumulated count value is greater than or equal to 0, the final adjustment rule is determined to be to find the subcategory closest to the characteristic data of the current working condition in the previous category of the current working condition and all categories of the previous gear as the target gear and subcategory for expected adjustment; if the accumulated count value is less than 0, the final adjustment rule is determined to be to find the subcategory closest to the characteristic data of the current working condition in the subsequent category of the current working condition and all categories of the next gear as the target gear and subcategory for expected adjustment; If the free calcium content and the three rates of raw materials entering the kiln of the current working condition are not obtained, and the characteristic data change rate, the head and tail coal usage ratio, and the decomposition furnace outlet temperature feedback value do not meet the adjustment conditions, then the final adjustment rule is to find the sub-category closest to the characteristic data of the current working condition in all the categories of the current working condition and the previous gear as the target gear and sub-category for the expected adjustment; (5.3.3) The final adjustment rule is to find the sub-category closest to the characteristic data of the current working condition in all the categories preceding the category of the current working condition and the upper N gears or lower N gears as the target gear and sub-category for the expected adjustment.

3. The method for determining the target value of the decomposition furnace outlet temperature based on dual drive of data and rules according to claim 1 is characterized in that: The adjustment conditions and rules for the variable positions and categories in step (5.3.2) are as follows: The adjustment conditions and rules for the free calcium content grade and classification are as follows: when A < fCao < B, no adjustment is made; when fCao ≤ A, search for the subcategory with the closest characteristic data to the current working condition among all the subcategories in the classification before the current working condition's classification and the previous grade; when fCao ≥ B, search for the subcategory with the closest characteristic data to the current characteristic among all the subcategories in the classification after the current working condition's classification and the next grade; fCao represents the average free calcium content of the current working condition, and both A and B are preset thresholds. The adjustment conditions and rules for the three modulus values grade and classification of the raw meal entering the kiln are as follows: when the KH and SM of the current working condition increase, search for the subcategory with the closest characteristic data to the current working condition among all the subcategories in the classification after the current working condition's classification and the next grade; when the IM of the current working condition increases, search for the subcategory with the closest characteristic data to the current working condition among all the subcategories in the classification before the current working condition's classification and the previous grade. The adjustment conditions and rules for the ratio grade and classification of the head and tail coal consumption are as follows: when the ratio of the head and tail coal consumption of the current working condition decreases, search for the subcategory with the closest characteristic data to the current working condition among all the subcategories in the classification after the current working condition's classification and the next grade. The adjustment conditions and rules for the characteristic data change rate grade and classification are as follows: when the characteristic data change rate of the current working condition decreases at a certain rate, search for the subcategory with the closest characteristic data to the current working condition among all the subcategories in the classification after the current working condition's classification and the next grade. The adjustment conditions and rules for the median grade and classification of the decomposition furnace outlet temperature feedback value are as follows: when the median of the decomposition furnace outlet temperature feedback value of the current working condition is lower than the median M℃ of the decomposition furnace outlet temperature feedback value of the current working condition's classification, search for the subcategory with the closest characteristic data to the current working condition among all the subcategories in the classification after the current working condition's classification and the next grade.

4. The method for determining the target value of the decomposition furnace outlet temperature based on data and rule dual drive according to claim 1 is characterized in that: In step (4), the data in the result grading and classification table are automatically iterated or manually adjusted during the cement production operation process.

5. An embedded system readable storage medium, characterized in that: The embedded system-readable storage medium includes a stored computer program, wherein, when the embedded system program runs, it controls the device where the embedded system-readable storage medium is located to execute the steps of the method described in any one of claims 1 to 4.

6. An embedded system device, characterized in that: A memory, a processor, and a program stored and executable on the memory, the program, when executed by the processor, implements the steps of the method described in any one of claims 1 to 4.

Images