Urea dissolution on-line control system based on particle size monitoring

Abstract

The invention discloses a urea dissolution online control system based on particle size monitoring, and the system comprises a dividing module which is used for dividing a historical monitoring log of each monitoring point, and obtaining the historical monitoring logs of the monitoring points in the same preset ammonia production demand interval; the extraction module is used for extracting historical granularity monitoring data and historical control data in the historical monitoring log, and determining feature control data based on an association relationship between the historical granularity monitoring data and the historical control data; the evaluation module is used for obtaining a historical particle size evaluation value of the historical particle size monitoring data, performing screening according to the historical particle size evaluation value characteristic control data, obtaining preferable characteristic control data and constructing a urea dissolution control model; the generation module is used for generating a real-time granularity evaluation value according to the real-time granularity monitoring data; the control module is used for selecting a urea dissolution control model and obtaining control data, the urea dissolution efficiency is improved, and the denitration emission requirement of the power plant is guaranteed.

Description

Technical Field

[0001] The present application relates to the technical field of urea hydrolysis to produce ammonia, and in particular to an online control system for urea dissolution based on particle size monitoring. Background Art

[0002] In order to completely eliminate the potential leakage risk caused by the large amount of liquid ammonia storage, the traditional liquid ammonia storage method is transformed into a safer urea hydrolysis process to ensure the personal safety of power plant employees and the public safety of surrounding communities. Urea hydrolysis technology has become one of the ideal alternatives to liquid ammonia due to its lower risk, especially in large coal-fired power plants. Its application can greatly reduce the existence of major hazardous sources. Therefore, there is an urgent need for an online control system for urea dissolution based on particle size monitoring to optimize the urea dissolution process and improve the dissolution efficiency. Summary of the invention

[0003] To solve the above technical problems, the present application provides an online control system for urea dissolution based on particle size monitoring, which divides a plurality of preset ammonia production demand intervals and analyzes the correlation between historical particle size monitoring data and historical control data of the same preset ammonia production demand interval, obtains characteristic control data of each preset ammonia production demand interval, and calculates the historical particle size evaluation value of each characteristic control data corresponding to the historical particle size monitoring data, selects the preferred characteristic control data according to the historical particle size evaluation value, constructs a urea dissolution control model corresponding to the preset ammonia production demand interval, selects the corresponding urea dissolution control model according to the real-time ammonia production demand, obtains control data of the real-time particle size evaluation value based on the urea dissolution control model, improves urea dissolution efficiency, and ensures the denitrification emission requirements of the power plant.

[0004] In some embodiments of the present application, a urea dissolution online control system based on particle size monitoring is provided, comprising:

[0005] A partitioning module is used to pre-set multiple monitoring points, obtain the historical monitoring log of each monitoring point, and partition the historical monitoring log to obtain the historical monitoring log of the monitoring points in the same preset ammonia production demand interval;

[0006] An extraction module is used to extract historical granularity monitoring data and historical control data from the historical monitoring log of the same preset ammonia production demand interval, and determine characteristic control data of each preset ammonia production demand interval based on the correlation between the historical granularity monitoring data and the historical control data;

[0007] An evaluation module is used to evaluate the historical particle size monitoring data of each preset ammonia production demand interval to obtain multiple historical particle size evaluation values, screen the corresponding characteristic control data according to the multiple historical particle size evaluation values ​​to obtain a number of preferred characteristic control data, and construct a urea dissolution control model corresponding to the preset ammonia production demand interval according to the several preferred characteristic control data;

[0008] A generation module is used to obtain the real-time ammonia production demand and the real-time particle size monitoring data of each monitoring point, and generate a real-time particle size evaluation value according to the real-time particle size monitoring data of each monitoring point;

[0009] The control module is used to select a urea dissolution control model based on the real-time ammonia production demand, input the real-time particle size evaluation value into the selected urea dissolution control model, obtain control data, and generate control instructions according to the control data.

[0010] In some embodiments of the present application, extracting historical granularity monitoring data and historical control data from historical monitoring logs in the same preset ammonia production demand interval includes:

[0011] Obtain historical urea hydrolysis ammonia production instructions, and extract historical ammonia production requirements of each historical urea hydrolysis ammonia production instruction;

[0012] Based on the correspondence between the preset ammonia production demand interval and the historical ammonia production demand, the historical urea hydrolysis ammonia production instructions in the same preset ammonia production demand interval and the historical monitoring logs of the monitoring points corresponding to the historical urea hydrolysis ammonia production instructions are obtained;

[0013] The same preset monitoring period is set for the historical monitoring logs of the monitoring points in the same preset ammonia production demand interval, the preset monitoring period is used as the time reference line, and the data collection nodes corresponding to the preset monitoring period are set according to the preset time interval;

[0014] According to several data collection nodes of each time reference line, historical granular monitoring data and historical control data in multiple historical monitoring logs corresponding to the preset ammonia production demand interval are collected, and mapped to the corresponding time reference line to obtain a correlation relationship diagram between the historical granular monitoring data and the historical control data of the historical monitoring logs of the monitoring points corresponding to the preset ammonia production demand interval;

[0015] The correlation between the historical granularity monitoring data and the historical control data corresponding to the preset ammonia production demand interval is determined according to the correlation diagram.

[0016] In some embodiments of the present application, based on the correlation between the historical granularity monitoring data and the historical control data, the characteristic control data of each preset ammonia production demand interval is determined, including:

[0017] Obtain the dependency relationship between the historical granularity monitoring data and the historical control data at each data collection node according to the correlation relationship diagram of each preset ammonia production demand interval, and calculate the dependency degree between the historical granularity monitoring data and the historical control data at the corresponding data collection node;

[0018] The number of data collection nodes whose dependence degree is greater than a preset dependence degree threshold and the dependence degree difference value whose dependence degree is greater than the preset dependence degree threshold are used to generate the correlation evaluation value of the corresponding historical granular monitoring data and the historical control data;

[0019] Compare the correlation evaluation value between the historical granular monitoring data and the historical control data in the historical monitoring log of each monitoring point with the preset correlation evaluation value threshold; if the correlation evaluation value is greater than the preset correlation evaluation value threshold, set the corresponding historical control data as the focus control data of the corresponding historical granular monitoring data in the current historical monitoring log of the monitoring point;

[0020] Compare the concerned control data of the same historical granularity monitoring data in all historical monitoring logs of the same preset ammonia production demand interval to obtain the appearance ratio of the concerned control data of the same historical granularity monitoring data in all historical monitoring logs of the current preset ammonia production demand interval;

[0021] If the occurrence ratio is less than the preset occurrence ratio threshold, the corresponding attention control data will be eliminated. If the occurrence ratio is greater than the preset occurrence ratio threshold, the corresponding attention control data will be set as the characteristic control data of the corresponding historical granularity monitoring data.

[0022] In some embodiments of the present application, the calculation formula of the association evaluation value is:

[0023]

[0024] Among them, G is the correlation evaluation value, z is the correlation evaluation conversion coefficient, n is the number of data collection nodes whose degree of dependence between historical granular monitoring data and historical control data is greater than the preset dependence threshold, m is the total number of data collection nodes, △yi is the difference in the degree of dependence between historical granular monitoring data and historical control data at the i-th data collection node that is greater than the preset dependence threshold.

[0025] In some embodiments of the present application, the historical particle size monitoring data of each preset ammonia production demand interval is evaluated to obtain multiple historical particle size evaluation values, including:

[0026] Multiple particle size evaluation indicators are pre-set, and each particle size rating indicator includes multiple standard particle size monitoring data;

[0027] Compare the historical particle size monitoring data of each monitoring point with the corresponding standard particle size monitoring data of the particle size evaluation index to obtain the data difference between the historical particle size monitoring data and the corresponding standard particle size monitoring data;

[0028] Generate a historical granularity evaluation value of the corresponding data collection node according to the data difference between all historical granularity monitoring data of the same data collection node and the corresponding standard granularity monitoring data of the granularity evaluation index, and the weight coefficient of the corresponding granularity evaluation index;

[0029] The calculation formula of the historical granularity evaluation value is:

[0030]

[0031] Among them, L is the historical granularity evaluation value of the corresponding data collection node, l0 is the granularity evaluation conversion coefficient, w1 is the total number of historical granularity monitoring data of the first granularity evaluation indicator of the corresponding data collection node, △Js is the data difference between the sth historical granularity monitoring data of the first granularity evaluation indicator of the corresponding data collection node and the corresponding standard granularity monitoring data, a1 is the weight coefficient of the first granularity evaluation indicator, w2 is the total number of historical granularity monitoring data of the second granularity evaluation indicator of the corresponding data collection node, △Jf is the data difference between the fth historical granularity monitoring data of the second granularity evaluation indicator of the corresponding data collection node and the corresponding standard granularity monitoring data, a2 is the weight coefficient of the second granularity evaluation indicator, wg is the total number of historical granularity monitoring data of the gth granularity evaluation indicator of the corresponding data collection node, △Jd is the data difference between the dth historical granularity monitoring data of the gth granularity evaluation indicator of the corresponding data collection node and the corresponding standard granularity monitoring data, ag is the weight coefficient of the gth granularity evaluation indicator, and g is the total number of granularity evaluation indicators.

[0032] In some embodiments of the present application, the corresponding feature control data is screened according to a plurality of historical granularity evaluation values ​​to obtain several preferred feature control data, including:

[0033] All historical monitoring logs contained in the same preset ammonia production demand interval are divided again to obtain historical monitoring logs of all monitoring points of the same historical urea hydrolysis ammonia production instruction in the same preset ammonia production demand interval;

[0034] The historical granularity evaluation value of the historical monitoring logs of all monitoring points of the same historical urea hydrolysis ammonia production instruction in the same preset ammonia production demand interval at each data collection node is mapped onto a time reference line to obtain a historical granularity evaluation value change curve of the historical granularity monitoring data of all monitoring points under each historical urea hydrolysis ammonia production instruction in the current preset ammonia production demand interval;

[0035] Analyze the historical particle size evaluation values ​​of all data collection nodes in the historical particle size evaluation value change curve of each historical urea hydrolysis ammonia production instruction to determine the change characteristics of the historical particle size evaluation value of each monitoring point, wherein the change characteristics include the change trend, the change rate and the change time;

[0036] If the change trend of the historical granularity evaluation value of the current monitoring point is in an upward trend, the change rate is greater than the preset change rate threshold, and the change time is greater than the preset change time threshold, then the change feature of the historical granularity evaluation value of the current monitoring point is set as the preferred change feature, and the change time corresponding to the preferred change feature of the current monitoring point is marked as the focus period;

[0037] Determine whether the change characteristics of the historical granularity evaluation values ​​of the remaining monitoring points are preferred change characteristics and calculate the number of monitoring points whose change characteristics are preferred change characteristics. If the number of monitoring points is greater than the preset number of monitoring points, set the feature control data of the current focus period as the preferred feature control data;

[0038] A urea dissolution control model corresponding to the preset ammonia production demand interval is constructed based on the preferred characteristic control data determined by the historical particle size evaluation value change curve of all historical urea hydrolysis ammonia production instructions in each preset ammonia production demand interval.

[0039] In some embodiments of the present application, a urea dissolution control model corresponding to a preset ammonia production demand interval is constructed according to a number of preferred characteristic control data, including:

[0040] Based on the same time node principle, the preferred characteristic control data of the current preset ammonia production demand interval, the historical granularity monitoring data of the data collection node where the preferred characteristic control data is located, and the historical granularity evaluation values ​​of all monitoring points of the data collection node are obtained, and the historical granularity evaluation values ​​of all monitoring points of the data collection node are averaged to obtain the average of the historical granularity evaluation values ​​of the data collection node where the preferred characteristic control data is located;

[0041] Presetting a plurality of preset particle size evaluation values ​​for each preset ammonia production demand interval;

[0042] Compare the mean of the historical granularity evaluation values ​​of the data collection node where the preferred characteristic control data of each preset ammonia production demand interval is located with the corresponding preset granularity evaluation value to determine whether the mean of all historical granularity evaluation values ​​includes all preset granularity evaluation values;

[0043] If so, the mean of the historical particle size evaluation values ​​of the data collection node where the preferred feature control data is located is used as the training input data, and the corresponding preferred feature control data is used as the training output data to perform neural network training to obtain a urea dissolution control model corresponding to the preset ammonia production demand range.

[0044] In some embodiments of the present application, a urea dissolution control model corresponding to a preset ammonia production demand interval is constructed according to a number of preferred characteristic control numbers, further comprising:

[0045] If not, the mean of the historical granularity evaluation values ​​of the data collection node where the preferred feature control data is located is used as the first training input data set, and the corresponding preferred feature control data is used as the training output label set;

[0046] Screen out the preset particle size evaluation values ​​that are not included in the average of all historical particle size evaluation values, perform similarity analysis based on the preset particle size evaluation values ​​that are not included and the average of historical particle size evaluation values ​​in the current preset ammonia production demand interval, use the average of historical particle size evaluation values ​​with the greatest similarity as the second training input parameter set, and use the corresponding feature control data as the training output pseudo label set;

[0047] Based on the first training input parameter set, the training output label set and the second training input parameter set, the training output pseudo label set, semi-supervised learning training is performed to generate a urea dissolution control model corresponding to a preset ammonia production demand range.

[0048] In some embodiments of the present application, a urea dissolution control model is selected based on the real-time ammonia production demand, and the real-time particle size evaluation value is input into the selected urea dissolution control model to obtain control data, including:

[0049] Obtain the current urea hydrolysis ammonia production instruction, extract the real-time ammonia production demand of the current urea hydrolysis ammonia production instruction, determine the preset ammonia production demand interval in which the real-time ammonia production demand is located, and select a urea dissolution control model corresponding to the preset ammonia production demand interval;

[0050] Acquire real-time particle size monitoring data of each monitoring point, compare the real-time particle size monitoring data with corresponding standard particle size monitoring data of the particle size evaluation index, and obtain a real-time particle size evaluation value of the real-time particle size monitoring data of each monitoring point according to the comparison result;

[0051] A real-time particle size evaluation value mean is generated according to the real-time particle size evaluation value of each monitoring point at the same monitoring time node, the real-time particle size evaluation value mean is input into the selected urea dissolution control model to obtain control data, and a control instruction is generated according to the control data.

[0052] Compared with the prior art, the urea dissolution online control system based on particle size monitoring in the embodiment of the present application has the following beneficial effects:

[0053] By dividing a plurality of preset ammonia production demand intervals and analyzing the correlation between the historical particle size monitoring data and the historical control data of the same preset ammonia production demand interval, the characteristic control data of each preset ammonia production demand interval is obtained, and the historical particle size evaluation value of the historical particle size monitoring data corresponding to each characteristic control data is calculated. The preferred characteristic control data is screened out according to the historical particle size evaluation value, and a urea dissolution control model corresponding to the preset ammonia production demand interval is constructed. The corresponding urea dissolution control model is selected according to the real-time ammonia production demand, and the control data of the real-time particle size evaluation value is obtained based on the urea dissolution control model, so as to improve the urea dissolution efficiency and ensure the denitrification emission demand of the power plant. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a schematic diagram of an online control system for urea dissolution based on particle size monitoring in a preferred embodiment of the present application. DETAILED DESCRIPTION

[0055] The specific implementation methods of the present application are further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present application but are not intended to limit the scope of the present application.

[0056] In the description of the present application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0057] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0058] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0059] like Figure 1As shown, an online control system for urea dissolution based on particle size monitoring according to a preferred embodiment of the present application includes:

[0060] A partitioning module is used to pre-set multiple monitoring points, obtain the historical monitoring log of each monitoring point, and partition the historical monitoring log to obtain the historical monitoring log of the monitoring points in the same preset ammonia production demand interval;

[0061] An extraction module is used to extract historical granularity monitoring data and historical control data from the historical monitoring log of the same preset ammonia production demand interval, and determine characteristic control data of each preset ammonia production demand interval based on the correlation between the historical granularity monitoring data and the historical control data;

[0062] An evaluation module is used to evaluate the historical particle size monitoring data of each preset ammonia production demand interval to obtain multiple historical particle size evaluation values, screen the corresponding characteristic control data according to the multiple historical particle size evaluation values ​​to obtain a number of preferred characteristic control data, and construct a urea dissolution control model corresponding to the preset ammonia production demand interval according to the several preferred characteristic control data;

[0063] A generation module is used to obtain the real-time ammonia production demand and the real-time particle size monitoring data of each monitoring point, and generate a real-time particle size evaluation value according to the real-time particle size monitoring data of each monitoring point;

[0064] The control module is used to select a urea dissolution control model based on the real-time ammonia production demand, input the real-time particle size evaluation value into the selected urea dissolution control model, obtain control data, and generate control instructions according to the control data.

[0065] In this embodiment, the historical particle size monitoring data includes historical particle size, historical particle size distribution, etc., and the historical control data includes dissolution time, stirring time, dissolution temperature, urea concentration, stirring intensity, etc.

[0066] In some embodiments of the present application, extracting historical granularity monitoring data and historical control data from historical monitoring logs in the same preset ammonia production demand interval includes:

[0067] Obtain historical urea hydrolysis ammonia production instructions, and extract historical ammonia production requirements of each historical urea hydrolysis ammonia production instruction;

[0068] Based on the correspondence between the preset ammonia production demand interval and the historical ammonia production demand, the historical urea hydrolysis ammonia production instructions in the same preset ammonia production demand interval and the historical monitoring logs of the monitoring points corresponding to the historical urea hydrolysis ammonia production instructions are obtained;

[0069] The same preset monitoring period is set for the historical monitoring logs of the monitoring points in the same preset ammonia production demand interval, the preset monitoring period is used as the time reference line, and the data collection nodes corresponding to the preset monitoring period are set according to the preset time interval;

[0070] According to several data collection nodes of each time reference line, historical granular monitoring data and historical control data in multiple historical monitoring logs corresponding to the preset ammonia production demand interval are collected, and mapped to the corresponding time reference line to obtain a correlation relationship diagram between the historical granular monitoring data and the historical control data of the historical monitoring logs of the monitoring points corresponding to the preset ammonia production demand interval;

[0071] The correlation between the historical granularity monitoring data and the historical control data corresponding to the preset ammonia production demand interval is determined according to the correlation diagram.

[0072] In this embodiment, the preset monitoring period refers to a standard monitoring period that satisfies both the corresponding preset ammonia production demand interval and the standard urea solution efficiency. By setting the preset monitoring period, it is convenient to subsequently analyze the historical particle size monitoring data and historical control data of the same preset ammonia production demand interval, thereby preventing the different time lengths of the data from causing errors or influences on the analysis results.

[0073] In some embodiments of the present application, based on the correlation between the historical granularity monitoring data and the historical control data, the characteristic control data of each preset ammonia production demand interval is determined, including:

[0074] Obtain the dependency relationship between the historical granularity monitoring data and the historical control data at each data collection node according to the correlation relationship diagram of each preset ammonia production demand interval, and calculate the dependency degree between the historical granularity monitoring data and the historical control data at the corresponding data collection node;

[0075] The number of data collection nodes whose dependence degree is greater than a preset dependence degree threshold and the dependence degree difference value whose dependence degree is greater than the preset dependence degree threshold are used to generate the correlation evaluation value of the corresponding historical granular monitoring data and the historical control data;

[0076] Compare the correlation evaluation value between the historical granular monitoring data and the historical control data in the historical monitoring log of each monitoring point with the preset correlation evaluation value threshold; if the correlation evaluation value is greater than the preset correlation evaluation value threshold, set the corresponding historical control data as the focus control data of the corresponding historical granular monitoring data in the current historical monitoring log of the monitoring point;

[0077] Compare the concerned control data of the same historical granularity monitoring data in all historical monitoring logs of the same preset ammonia production demand interval to obtain the appearance ratio of the concerned control data of the same historical granularity monitoring data in all historical monitoring logs of the current preset ammonia production demand interval;

[0078] If the occurrence ratio is less than the preset occurrence ratio threshold, the corresponding attention control data will be eliminated. If the occurrence ratio is greater than the preset occurrence ratio threshold, the corresponding attention control data will be set as the characteristic control data of the corresponding historical granularity monitoring data.

[0079] In this embodiment, the occurrence ratio refers to the number of attention control data of the same historical granularity monitoring data appearing in the historical monitoring log / the number of all historical monitoring logs in the current preset ammonia production demand interval. The credibility of the attention monitoring data corresponding to the historical granularity monitoring data can be judged by the occurrence ratio. The preset occurrence ratio threshold = 2 / 3. When the occurrence ratio is greater than 2 / 3, it means that the credibility of the attention monitoring data of the current historical granularity monitoring data is relatively large, and it can be set as the characteristic control data of the corresponding historical granularity monitoring data.

[0080] In this embodiment, the characteristic control data of each preset ammonia production demand interval is screened out through the correlation between the historical particle size monitoring data and the historical control data, laying the foundation for further screening out the preferred characteristic control data of each preset ammonia production demand interval, reducing the amount of data analysis, and improving the control efficiency in the urea dissolution process, thereby improving the urea dissolution efficiency.

[0081] In some embodiments of the present application, the calculation formula of the association evaluation value is:

[0082]

[0083] Among them, G is the correlation evaluation value, z is the correlation evaluation conversion coefficient, n is the number of data collection nodes whose degree of dependence between historical granular monitoring data and historical control data is greater than the preset dependence threshold, m is the total number of data collection nodes, △yi is the difference in the degree of dependence between historical granular monitoring data and historical control data at the i-th data collection node that is greater than the preset dependence threshold.

[0084] In some embodiments of the present application, the historical particle size monitoring data of each preset ammonia production demand interval is evaluated to obtain multiple historical particle size evaluation values, including:

[0085] Multiple particle size evaluation indicators are pre-set, and each particle size rating indicator includes multiple standard particle size monitoring data;

[0086] Compare the historical particle size monitoring data of each monitoring point with the corresponding standard particle size monitoring data of the particle size evaluation index to obtain the data difference between the historical particle size monitoring data and the corresponding standard particle size monitoring data;

[0087] Generate a historical granularity evaluation value of the corresponding data collection node according to the data difference between all historical granularity monitoring data of the same data collection node and the corresponding standard granularity monitoring data of the granularity evaluation index, and the weight coefficient of the corresponding granularity evaluation index;

[0088] The calculation formula of the historical granularity evaluation value is:

[0089]

[0090] Among them, L is the historical granularity evaluation value of the corresponding data collection node, l0 is the granularity evaluation conversion coefficient, w1 is the total number of historical granularity monitoring data of the first granularity evaluation indicator of the corresponding data collection node, △Js is the data difference between the sth historical granularity monitoring data of the first granularity evaluation indicator of the corresponding data collection node and the corresponding standard granularity monitoring data, a1 is the weight coefficient of the first granularity evaluation indicator, w2 is the total number of historical granularity monitoring data of the second granularity evaluation indicator of the corresponding data collection node, △Jf is the data difference between the fth historical granularity monitoring data of the second granularity evaluation indicator of the corresponding data collection node and the corresponding standard granularity monitoring data, a2 is the weight coefficient of the second granularity evaluation indicator, wg is the total number of historical granularity monitoring data of the gth granularity evaluation indicator of the corresponding data collection node, △Jd is the data difference between the dth historical granularity monitoring data of the gth granularity evaluation indicator of the corresponding data collection node and the corresponding standard granularity monitoring data, ag is the weight coefficient of the gth granularity evaluation indicator, and g is the total number of granularity evaluation indicators.

[0091] In this embodiment, the particle size evaluation index includes particle size uniformity, particle size distribution, particle size, etc. The particle size evaluation conversion coefficient refers to converting the data difference between the relevant particle size monitoring data of all particle size evaluation indicators and the corresponding standard particle size monitoring data into a numerical value of the same dimension as the particle size evaluation value. When the data difference is smaller, the larger the historical particle size evaluation value of the corresponding data collection node is, which means that the urea particle size at the corresponding data collection node is closer to the standard particle size monitoring data set by the corresponding particle size evaluation index, and the urea dissolution efficiency is higher.

[0092] In some embodiments of the present application, the corresponding feature control data is screened according to a plurality of historical granularity evaluation values ​​to obtain several preferred feature control data, including:

[0093] All historical monitoring logs contained in the same preset ammonia production demand interval are divided again to obtain historical monitoring logs of all monitoring points of the same historical urea hydrolysis ammonia production instruction in the same preset ammonia production demand interval;

[0094] The historical granularity evaluation value of the historical monitoring logs of all monitoring points of the same historical urea hydrolysis ammonia production instruction in the same preset ammonia production demand interval at each data collection node is mapped onto a time reference line to obtain a historical granularity evaluation value change curve of the historical granularity monitoring data of all monitoring points under each historical urea hydrolysis ammonia production instruction in the current preset ammonia production demand interval;

[0095] Determine whether the historical particle size evaluation value at the last data collection node in all historical particle size evaluation value change curves in the current preset ammonia production demand interval is greater than the preset particle size evaluation value threshold and whether the historical ammonia production amount meets the historical ammonia production demand. If not, remove the corresponding historical urea hydrolysis ammonia production instruction and the corresponding historical particle size evaluation value change curve;

[0096] If yes, analyze the historical particle size evaluation values ​​of all data collection nodes in the historical particle size evaluation value change curve of the remaining historical urea hydrolysis ammonia production instructions to determine the change characteristics of the historical particle size evaluation value of each monitoring point, wherein the change characteristics include the change trend, change rate and change time;

[0097] If the change trend of the historical granularity evaluation value of the current monitoring point is in an upward trend, the change rate is greater than the preset change rate threshold, and the change time is greater than the preset change time threshold, then the change feature of the historical granularity evaluation value of the current monitoring point is set as the preferred change feature, and the change time corresponding to the preferred change feature of the current monitoring point is marked as the focus period;

[0098] Determine the number of preferred change features of all monitoring points in the same period of interest, and if the number is greater than a preset number threshold, set the feature control data of the current period of interest as the preferred feature control data;

[0099] A urea dissolution control model corresponding to the preset ammonia production demand interval is constructed based on the preferred characteristic control data determined by the historical particle size evaluation value change curve of all historical urea hydrolysis ammonia production instructions in each preset ammonia production demand interval.

[0100] In this embodiment, the preset number threshold refers to 4 / 5 of the number of all monitoring points.

[0101] In this embodiment, the historical particle size evaluation value and the historical ammonia production at the last data collection node are first determined, so as to eliminate the historical urea hydrolysis ammonia production instructions and the corresponding characteristic control data with low urea dissolution efficiency or not meeting the ammonia production demand, thereby improving the analysis efficiency and the setting efficiency of the control data.

[0102] In some embodiments of the present application, a urea dissolution control model corresponding to a preset ammonia production demand interval is constructed according to a number of preferred characteristic control data, including:

[0103] Based on the same time node principle, the preferred characteristic control data of the current preset ammonia production demand interval, the historical granularity monitoring data of the data collection node where the preferred characteristic control data is located, and the historical granularity evaluation values ​​of all monitoring points of the data collection node are obtained, and the historical granularity evaluation values ​​of all monitoring points of the data collection node are averaged to obtain the average of the historical granularity evaluation values ​​of the data collection node where the preferred characteristic control data is located;

[0104] Presetting a plurality of preset particle size evaluation values ​​for each preset ammonia production demand interval;

[0105] Compare the mean of the historical granularity evaluation values ​​of the data collection node where the preferred characteristic control data of each preset ammonia production demand interval is located with the corresponding preset granularity evaluation value to determine whether the mean of all historical granularity evaluation values ​​includes all preset granularity evaluation values;

[0106] If so, the mean of the historical particle size evaluation values ​​of the data collection node where the preferred feature control data is located is used as the training input data, and the corresponding preferred feature control data is used as the training output data to perform neural network training to obtain a urea dissolution control model corresponding to the preset ammonia production demand range.

[0107] In this embodiment, the preset particle size evaluation value is set based on the historical particle size evaluation values ​​that appear multiple times in the historical monitoring logs in the preset ammonia production demand interval. The preset particle size evaluation value is used to determine whether the selected preferred feature control data can cope with various particle size conditions in the urea dissolution process in the preset ammonia production demand interval, thereby laying the foundation for providing accurate and effective control data for the subsequent urea dissolution process.

[0108] In some embodiments of the present application, a urea dissolution control model corresponding to a preset ammonia production demand interval is constructed according to a number of preferred characteristic control numbers, further comprising:

[0109] If not, the mean of the historical granularity evaluation values ​​of the data collection node where the preferred feature control data is located is used as the first training input data set, and the corresponding preferred feature control data is used as the training output label set;

[0110] Screen out the preset particle size evaluation values ​​that are not included in the average of all historical particle size evaluation values, perform similarity analysis based on the preset particle size evaluation values ​​that are not included and the average of historical particle size evaluation values ​​in the current preset ammonia production demand interval, use the average of historical particle size evaluation values ​​with the greatest similarity as the second training input parameter set, and use the corresponding feature control data as the training output pseudo label set;

[0111] Based on the first training input parameter set, the training output label set and the second training input parameter set, the training output pseudo label set, semi-supervised learning training is performed to generate a urea dissolution control model corresponding to a preset ammonia production demand range.

[0112] In this embodiment, through semi-supervised learning in the presence of a pseudo label set, the pseudo label set can be used to improve the generalization ability and prediction accuracy of the model, thereby improving the accuracy of the urea dissolution control model corresponding to the ammonia production demand range, and ensuring the control efficiency and control accuracy of subsequent control data.

[0113] In some embodiments of the present application, a urea dissolution control model is selected based on the real-time ammonia production demand, and the real-time particle size evaluation value is input into the selected urea dissolution control model to obtain control data, including:

[0114] Obtain the current urea hydrolysis ammonia production instruction, extract the real-time ammonia production demand of the current urea hydrolysis ammonia production instruction, determine the preset ammonia production demand interval in which the real-time ammonia production demand is located, and select a urea dissolution control model corresponding to the preset ammonia production demand interval;

[0115] Acquire real-time particle size monitoring data of each monitoring point, compare the real-time particle size monitoring data with corresponding standard particle size monitoring data of the particle size evaluation index, and obtain a real-time particle size evaluation value of the real-time particle size monitoring data of each monitoring point according to the comparison result;

[0116] A real-time particle size evaluation value mean is generated according to the real-time particle size evaluation value of each monitoring point at the same monitoring time node, the real-time particle size evaluation value mean is input into the selected urea dissolution control model to obtain control data, and a control instruction is generated according to the control data.

[0117] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and substitutions can be made without departing from the technical principles of the present application. These improvements and substitutions should also be regarded as the scope of protection of the present application.

Claims

1. A urea dissolution online control system based on particle size monitoring, characterized in that: include: A partitioning module is used to pre-set multiple monitoring points, obtain the historical monitoring log of each monitoring point, and partition the historical monitoring log to obtain the historical monitoring log of the monitoring points in the same preset ammonia production demand interval; An extraction module is used to extract historical granularity monitoring data and historical control data from the historical monitoring log of the same preset ammonia production demand interval, and determine characteristic control data of each preset ammonia production demand interval based on the correlation between the historical granularity monitoring data and the historical control data; An evaluation module is used to evaluate the historical particle size monitoring data of each preset ammonia production demand interval to obtain multiple historical particle size evaluation values, screen the corresponding characteristic control data according to the multiple historical particle size evaluation values ​​to obtain a number of preferred characteristic control data, and construct a urea dissolution control model corresponding to the preset ammonia production demand interval according to the several preferred characteristic control data; A generation module is used to obtain the real-time ammonia production demand and the real-time particle size monitoring data of each monitoring point, and generate a real-time particle size evaluation value according to the real-time particle size monitoring data of each monitoring point; The control module is used to select a urea dissolution control model based on the real-time ammonia production demand, input the real-time particle size evaluation value into the selected urea dissolution control model, obtain control data, and generate control instructions according to the control data.

2. The urea dissolution online control system based on particle size monitoring according to claim 1, characterized in that: Extract the historical granular monitoring data and historical control data from the historical monitoring log for the same preset ammonia demand interval, including: Obtain historical urea hydrolysis ammonia production instructions, and extract historical ammonia production requirements of each historical urea hydrolysis ammonia production instruction; Based on the correspondence between the preset ammonia production demand interval and the historical ammonia production demand, the historical urea hydrolysis ammonia production instructions in the same preset ammonia production demand interval and the historical monitoring logs of the monitoring points corresponding to the historical urea hydrolysis ammonia production instructions are obtained; The same preset monitoring period is set for the historical monitoring logs of the monitoring points in the same preset ammonia production demand interval, the preset monitoring period is used as the time reference line, and the data collection nodes corresponding to the preset monitoring period are set according to the preset time interval; According to several data collection nodes of each time reference line, historical granular monitoring data and historical control data in multiple historical monitoring logs corresponding to the preset ammonia production demand interval are collected, and mapped to the corresponding time reference line to obtain a correlation relationship diagram between the historical granular monitoring data and the historical control data of the historical monitoring logs of the monitoring points corresponding to the preset ammonia production demand interval; The correlation between the historical granularity monitoring data and the historical control data corresponding to the preset ammonia production demand interval is determined according to the correlation diagram.

3. The urea dissolution online control system based on particle size monitoring according to claim 2, characterized in that: Based on the correlation between the historical granularity monitoring data and the historical control data, the characteristic control data of each preset ammonia production demand interval is determined, including: Obtain the dependency relationship between the historical granularity monitoring data and the historical control data at each data collection node according to the correlation relationship diagram of each preset ammonia production demand interval, and calculate the dependency degree between the historical granularity monitoring data and the historical control data at the corresponding data collection node; The number of data collection nodes whose dependence degree is greater than a preset dependence degree threshold and the dependence degree difference value whose dependence degree is greater than the preset dependence degree threshold are used to generate the correlation evaluation value of the corresponding historical granular monitoring data and the historical control data; Compare the correlation evaluation value between the historical granular monitoring data and the historical control data in the historical monitoring log of each monitoring point with the preset correlation evaluation value threshold; if the correlation evaluation value is greater than the preset correlation evaluation value threshold, set the corresponding historical control data as the focus control data of the corresponding historical granular monitoring data in the current historical monitoring log of the monitoring point; Compare the concerned control data of the same historical granularity monitoring data in all historical monitoring logs of the same preset ammonia production demand interval to obtain the appearance ratio of the concerned control data of the same historical granularity monitoring data in all historical monitoring logs of the current preset ammonia production demand interval; If the occurrence ratio is less than the preset occurrence ratio threshold, the corresponding attention control data will be eliminated. If the occurrence ratio is greater than the preset occurrence ratio threshold, the corresponding attention control data will be set as the characteristic control data of the corresponding historical granularity monitoring data.

4. The urea dissolution online control system based on particle size monitoring according to claim 3, characterized in that: The calculation formula of the association evaluation value is: Among them, G is the correlation evaluation value, z is the correlation evaluation conversion coefficient, n is the number of data collection nodes whose degree of dependence between historical granular monitoring data and historical control data is greater than the preset dependence threshold, m is the total number of data collection nodes, △yi is the difference in the degree of dependence between historical granular monitoring data and historical control data at the i-th data collection node that is greater than the preset dependence threshold.

5. The urea dissolution online control system based on particle size monitoring according to claim 4, characterized in that: The historical particle size monitoring data of each preset ammonia production demand interval is evaluated to obtain multiple historical particle size evaluation values, including: Multiple particle size evaluation indicators are pre-set, and each particle size rating indicator includes multiple standard particle size monitoring data; Compare the historical particle size monitoring data of each monitoring point with the corresponding standard particle size monitoring data of the particle size evaluation index to obtain the data difference between the historical particle size monitoring data and the corresponding standard particle size monitoring data; Generate a historical granularity evaluation value of the corresponding data collection node according to the data difference between all historical granularity monitoring data of the same data collection node and the corresponding standard granularity monitoring data of the granularity evaluation index, and the weight coefficient of the corresponding granularity evaluation index; The calculation formula of the historical granularity evaluation value is: Among them, L is the historical granularity evaluation value of the corresponding data collection node, l0 is the granularity evaluation conversion coefficient, w1 is the total number of historical granularity monitoring data of the first granularity evaluation indicator of the corresponding data collection node, △Js is the data difference between the sth historical granularity monitoring data of the first granularity evaluation indicator of the corresponding data collection node and the corresponding standard granularity monitoring data, a1 is the weight coefficient of the first granularity evaluation indicator, w2 is the total number of historical granularity monitoring data of the second granularity evaluation indicator of the corresponding data collection node, △Jf is the data difference between the fth historical granularity monitoring data of the second granularity evaluation indicator of the corresponding data collection node and the corresponding standard granularity monitoring data, a2 is the weight coefficient of the second granularity evaluation indicator, wg is the total number of historical granularity monitoring data of the gth granularity evaluation indicator of the corresponding data collection node, △Jd is the data difference between the dth historical granularity monitoring data of the gth granularity evaluation indicator of the corresponding data collection node and the corresponding standard granularity monitoring data, ag is the weight coefficient of the gth granularity evaluation indicator, and g is the total number of granularity evaluation indicators.

6. The urea dissolution online control system based on particle size monitoring according to claim 5, characterized in that: The corresponding feature control data are screened according to multiple historical granularity evaluation values ​​to obtain several preferred feature control data, including: All historical monitoring logs contained in the same preset ammonia production demand interval are divided again to obtain historical monitoring logs of all monitoring points of the same historical urea hydrolysis ammonia production instruction in the same preset ammonia production demand interval; The historical granularity evaluation value of the historical monitoring logs of all monitoring points of the same historical urea hydrolysis ammonia production instruction in the same preset ammonia production demand interval at each data collection node is mapped onto a time reference line to obtain a historical granularity evaluation value change curve of the historical granularity monitoring data of all monitoring points under each historical urea hydrolysis ammonia production instruction in the current preset ammonia production demand interval; Analyze the historical particle size evaluation values ​​of all data collection nodes in the historical particle size evaluation value change curve of each historical urea hydrolysis ammonia production instruction to determine the change characteristics of the historical particle size evaluation value of each monitoring point, wherein the change characteristics include the change trend, the change rate and the change time; If the change trend of the historical granularity evaluation value of the current monitoring point is in an upward trend, the change rate is greater than the preset change rate threshold, and the change time is greater than the preset change time threshold, then the change feature of the historical granularity evaluation value of the current monitoring point is set as the preferred change feature, and the change time corresponding to the preferred change feature of the current monitoring point is marked as the focus period; Determine whether the change characteristics of the historical granularity evaluation values ​​of the remaining monitoring points are preferred change characteristics and calculate the number of monitoring points whose change characteristics are preferred change characteristics. If the number of monitoring points is greater than the preset number of monitoring points, set the feature control data of the current focus period as the preferred feature control data; A urea dissolution control model corresponding to the preset ammonia production demand interval is constructed based on the preferred characteristic control data determined by the historical particle size evaluation value change curve of all historical urea hydrolysis ammonia production instructions in each preset ammonia production demand interval.

7. The urea dissolution online control system based on particle size monitoring according to claim 6, characterized in that: A urea dissolution control model corresponding to a preset ammonia production demand range is constructed according to a number of preferred characteristic control data, including: Based on the same time node principle, the preferred characteristic control data of the current preset ammonia production demand interval, the historical granularity monitoring data of the data collection node where the preferred characteristic control data is located, and the historical granularity evaluation values ​​of all monitoring points of the data collection node are obtained, and the historical granularity evaluation values ​​of all monitoring points of the data collection node are averaged to obtain the average of the historical granularity evaluation values ​​of the data collection node where the preferred characteristic control data is located; Presetting a plurality of preset particle size evaluation values ​​for each preset ammonia production demand interval; Compare the mean of the historical granularity evaluation values ​​of the data collection node where the preferred characteristic control data of each preset ammonia production demand interval is located with the corresponding preset granularity evaluation value to determine whether the mean of all historical granularity evaluation values ​​includes all preset granularity evaluation values; If so, the mean of the historical particle size evaluation values ​​of the data collection node where the preferred feature control data is located is used as the training input data, and the corresponding preferred feature control data is used as the training output data to perform neural network training to obtain a urea dissolution control model corresponding to the preset ammonia production demand range.

8. The urea dissolution online control system based on particle size monitoring according to claim 7, characterized in that: A urea dissolution control model corresponding to a preset ammonia production demand interval is constructed according to a number of preferred characteristic control numbers, and further includes: If not, the mean of the historical granularity evaluation values ​​of the data collection node where the preferred feature control data is located is used as the first training input data set, and the corresponding preferred feature control data is used as the training output label set; Screen out the preset particle size evaluation values ​​that are not included in the average of all historical particle size evaluation values, perform similarity analysis based on the preset particle size evaluation values ​​that are not included and the average of historical particle size evaluation values ​​in the current preset ammonia production demand interval, use the average of historical particle size evaluation values ​​with the greatest similarity as the second training input parameter set, and use the corresponding feature control data as the training output pseudo label set; Based on the first training input parameter set, the training output label set and the second training input parameter set, the training output pseudo label set, semi-supervised learning training is performed to generate a urea dissolution control model corresponding to a preset ammonia production demand range.

9. The urea dissolution online control system based on particle size monitoring according to claim 8, characterized in that: A urea dissolution control model is selected based on the real-time ammonia production demand, and the real-time particle size evaluation value is input into the selected urea dissolution control model to obtain control data, including: Obtain the current urea hydrolysis ammonia production instruction, extract the real-time ammonia production demand of the current urea hydrolysis ammonia production instruction, determine the preset ammonia production demand interval in which the real-time ammonia production demand is located, and select a urea dissolution control model corresponding to the preset ammonia production demand interval; Acquire real-time particle size monitoring data of each monitoring point, compare the real-time particle size monitoring data with corresponding standard particle size monitoring data of the particle size evaluation index, and obtain a real-time particle size evaluation value of the real-time particle size monitoring data of each monitoring point according to the comparison result; A real-time particle size evaluation value mean is generated according to the real-time particle size evaluation value of each monitoring point at the same monitoring time node, the real-time particle size evaluation value mean is input into the selected urea dissolution control model to obtain control data, and a control instruction is generated according to the control data.

Images