Gas-fired boiler intelligent control system and method based on DCS

Abstract

The invention relates to the technical field of gas-fired boiler control, and discloses a DCS-based gas-fired boiler intelligent control system and method, and the system comprises a setting module which is used for presetting a plurality of monitoring indexes, obtaining the feature monitoring data of each monitoring index, and generating a monitoring evaluation value according to the feature monitoring data; the construction module is used for analyzing the historical control log of each monitoring index, determining historical association control data and historical association characteristics, and constructing a corresponding association influence model; the judgment module is used for judging whether an adjustment instruction is generated or not according to the current monitoring evaluation value, and if yes, associated features are determined; and the adjustment module is used for generating strategies to be subjected to stator adjustment according to the associated features, performing conflict analysis and optimization on the multiple strategies to be subjected to stator adjustment to obtain a final adjustment strategy, and generating a current adjustment instruction according to the final adjustment strategy. The control efficiency of the gas boiler control system is improved, and normal and stable operation of the gas boiler is ensured.

Description

Technical Field

[0001] This application relates to the technical field of gas boiler control, and particularly to an intelligent control system and method for gas boilers based on DCS. Background Art

[0002] The DCS intelligent control system can achieve centralized management and decentralized control during the operation of gas boilers. There are various monitoring parameters involved in the operation of gas boilers. The existing DCS intelligent control system cannot comprehensively monitor and evaluate the operation of gas boilers, resulting in the inability to automatically adjust the control strategy according to the actual situation, reducing the control efficiency of gas boilers and unable to ensure the normal and stable operation of gas boilers. Summary of the Invention

[0003] To solve the above technical problems, this application provides an intelligent control system and method for gas boilers based on DCS. By setting multiple monitoring indicators and determining the characteristic monitoring data of the monitoring indicators, generating a monitoring evaluation value according to the characteristic monitoring data and judging whether to adjust the associated control data of the monitoring indicators. If necessary, generating multiple to-be-determined adjustment strategies, performing conflict analysis and optimization on the multiple to-be-determined adjustment strategies, determining the final adjustment strategy, comprehensively monitoring and accurately evaluating the operation of the gas boiler, and adjusting the corresponding control data according to the actual operation situation, improving the control efficiency of the gas boiler control system and ensuring the normal and stable operation of the gas boiler.

[0004] In some embodiments of this application, an intelligent control system for gas boilers based on DCS is provided, including: A setting module, configured to preset multiple monitoring indicators in advance, obtain the characteristic monitoring data of each monitoring indicator, and generate a monitoring evaluation value for the corresponding monitoring indicator according to the characteristic monitoring data; A construction module, configured to analyze the historical control logs of each monitoring indicator, determine the historical associated control data and the corresponding historical associated characteristics of each monitoring indicator, and construct a corresponding associated influence model; A judgment module, configured to judge whether to generate an adjustment instruction for the current associated control data according to the current monitoring evaluation value. If so, generate an associated characteristic based on the associated influence model and the monitoring evaluation value; An adjustment module, configured to generate a to-be-determined adjustment strategy for the corresponding associated control data according to the associated characteristic, perform conflict analysis and optimization on the multiple to-be-determined adjustment strategies, obtain the final adjustment strategy, and generate the current adjustment instruction according to the final adjustment strategy.

[0005] In some embodiments of this application, the judgment module further includes: If not, obtain the monitoring evaluation value of the corresponding monitoring indicator in the current monitoring period, and construct a change curve of the monitoring evaluation value in the current period; Obtain the change characteristics of the monitoring evaluation value curve, and extrapolate the monitoring evaluation value curve according to the change characteristics to obtain multiple predicted monitoring evaluation values for future time periods; Compare the predicted monitoring evaluation values with the preset monitoring evaluation value thresholds of the corresponding monitoring indicators to obtain multiple differences in predicted monitoring evaluation values; Obtain the predicted change characteristics of multiple predicted monitoring evaluation values for future time periods, and generate a predicted trend coefficient for the corresponding monitoring indicator according to the predicted change characteristics and the differences in predicted monitoring evaluation values; Preset a predicted trend coefficient threshold in advance; If the predicted trend coefficient is less than the predicted trend coefficient threshold, generate a warning instruction for the corresponding monitoring indicator, and predict the predicted failure time point of the corresponding monitoring indicator; Input multiple predicted monitoring evaluation values before the predicted failure time point into the corresponding associated influence model to obtain multiple predicted associated characteristics, and obtain multiple predicted stator adjustment strategies to be determined for the corresponding monitoring indicator according to the multiple predicted associated characteristics; Conduct conflict analysis and optimization on several predicted stator adjustment strategies for future time periods, determine the predicted adjustment strategy for future time periods, and generate an adjustment instruction for future time periods according to the predicted adjustment strategy.

[0006] In some embodiments of the present application, obtain the characteristic monitoring data of each monitoring indicator, including: Obtain the historical monitoring log of each monitoring indicator, extract the historical monitoring evaluation value at each historical monitoring time node in the historical monitoring log, and construct a historical monitoring evaluation value change curve corresponding to the historical monitoring log; Collect the historical monitoring data of the corresponding monitoring indicator in the historical monitoring log according to the historical monitoring time node, and map it into the corresponding historical monitoring evaluation value change curve to obtain a historical monitoring evaluation value - monitoring data change relationship diagram; Obtain the first fluctuation characteristic of the historical monitoring evaluation value in the historical monitoring evaluation value change curve, where the first fluctuation characteristic includes the first fluctuation degree and the first fluctuation magnitude; Screen out the historical monitoring time nodes with the first fluctuation degree greater than the preset fluctuation degree threshold, and set them as the concerned monitoring time nodes, and intercept the influence periods of each concerned monitoring time node in the same historical monitoring evaluation value - monitoring data change relationship diagram; Obtain the second fluctuation characteristic of each historical monitoring data in each influence period, where the second fluctuation characteristic includes the second fluctuation degree and the second fluctuation magnitude; Filter out historical monitoring data with a second fluctuation degree greater than a preset fluctuation degree threshold, and construct a second fluctuation degree sequence and a second fluctuation magnitude sequence respectively according to the second fluctuation degree and the second fluctuation magnitude of the corresponding historical monitoring data in multiple influence periods; Construct a first fluctuation degree sequence and a first fluctuation magnitude sequence respectively according to the first fluctuation degree and the first fluctuation magnitude of the first fluctuation characteristics of the historical monitoring evaluation value at multiple concerned monitoring time nodes; Compare the second fluctuation degree sequence and the second fluctuation magnitude sequence of each historical monitoring data with the first fluctuation degree sequence and the first fluctuation magnitude sequence of the historical monitoring evaluation value respectively to obtain a first sequence similarity and a second sequence similarity; Generate a correlation coefficient between the corresponding historical monitoring data and the historical monitoring evaluation value according to the first sequence similarity and the second sequence similarity; The calculation formula of the correlation coefficient is: ; Where G is the correlation coefficient, g0 is the correlation conversion coefficient, L1 is the first sequence similarity, a1 is the weight coefficient of the first sequence similarity, a2 is the weight coefficient of the second sequence similarity, and L2 is the second sequence similarity; Preset a correlation coefficient threshold in advance; If the correlation coefficient is greater than the correlation coefficient threshold, set the corresponding historical monitoring data as characteristic monitoring data, and set the weight coefficient of the corresponding characteristic monitoring data according to the correlation coefficient difference where the correlation coefficient is greater than the correlation coefficient threshold.

[0007] In some embodiments of the present application, generating a monitoring evaluation value of a corresponding monitoring index according to the characteristic monitoring data includes: Preset several preset sub-data intervals of the characteristic monitoring data of each monitoring index in advance, and each preset sub-data interval is mapped with a corresponding preset sub-monitoring evaluation value; Compare each characteristic monitoring data with the corresponding several preset sub-data intervals to obtain the preset sub-data interval where the corresponding characteristic monitoring data is located, and set the preset sub-monitoring evaluation value of the preset sub-data interval where it is located as the sub-monitoring evaluation value of the corresponding characteristic monitoring data; Generate a monitoring evaluation value of the corresponding monitoring index according to the sub-monitoring evaluation values of all the characteristic monitoring data of the same monitoring index and the weight coefficients of the corresponding characteristic monitoring data.

[0008] In some embodiments of the present application, determining the historical correlation control data and the corresponding historical correlation characteristics of each monitoring index includes: Obtain the historical control logs of each monitoring indicator, extract the historical monitoring evaluation values and historical control data of the corresponding monitoring indicators in each historical control log according to the historical monitoring time nodes, and construct a historical monitoring evaluation value-control data change relationship graph for each historical control log. Among them, the historical monitoring evaluation value-control data change relationship graph includes a historical monitoring evaluation value change curve and several historical control data change curves. Preset a monitoring evaluation value threshold in advance. Screen out the first historical monitoring time node in the same historical monitoring evaluation value-control data change relationship graph where the historical monitoring evaluation value is less than the monitoring evaluation value threshold, and set it as the initial monitoring time node. Obtain the first historical monitoring time node after the initial monitoring time node where the historical monitoring evaluation value is greater than the monitoring evaluation value threshold, and set it as the end monitoring time node. Intercept the historical monitoring evaluation value change curve segment and the historical control data change curve segment between the initial monitoring time node and the end monitoring time node in the same historical monitoring evaluation value-control data change relationship graph. Analyze the intercepted historical monitoring evaluation value change curve segment to determine the historical monitoring evaluation value difference and the positive trend points in the historical monitoring evaluation value change curve segment. Obtain the historical first adjustment value of the historical control data of each historical control data change curve segment between the positive trend point and the initial monitoring time node. Set the historical control data with the historical first adjustment value greater than the preset adjustment value threshold as the historical associated control data corresponding to the historical monitoring evaluation value difference, and calculate the historical delay duration between the associated control data and the historical monitoring evaluation value. Construct a historical associated control data subset corresponding to the historical monitoring evaluation value difference based on several historical associated control data. Obtain the historical second adjustment value of the historical associated control data between the positive trend point and the end monitoring time node, and generate the historical comprehensive adjustment value of the corresponding historical associated control data according to the first historical adjustment value and the second historical adjustment value. Set the historical delay duration, the historical comprehensive adjustment value, the historical adjustment cost required for the historical comprehensive adjustment value, and the historical adjustment duration as the historical associated characteristics of the corresponding historical associated control data and the corresponding historical monitoring evaluation value difference. Classify the historical associated control data subsets and the corresponding historical associated characteristics of multiple historical monitoring evaluation value differences of the same monitoring indicator to obtain historical associated control data sets for multiple historical monitoring evaluation value difference intervals, where the historical associated control data sets include several historical associated control data subsets.

[0009] In some embodiments of the present application, constructing a corresponding association impact model includes: Using the historical monitoring evaluation value differences in the historical monitoring evaluation value difference intervals for each monitoring indicator as the training input data, and using several historical associated control data and the corresponding historical associated features in the historical associated control data subsets for each historical monitoring evaluation value difference as the training output data; Based on the training input data and the training output data, perform neural network training to obtain an association influence model between the historical monitoring evaluation value differences and the historical associated control data in the corresponding monitoring indicators.

[0010] In some embodiments of the present application, determine whether to generate an adjustment instruction for the associated control data according to the current monitoring evaluation value. If so, generate associated features based on the association influence model and the monitoring evaluation value, including: If the current monitoring evaluation value is less than the monitoring evaluation value threshold, generate an adjustment instruction for the associated control data of the corresponding monitoring indicator; Calculate the monitoring evaluation value difference between the monitoring evaluation value of the corresponding monitoring indicator and the monitoring evaluation value threshold, and input the monitoring evaluation value difference into the corresponding association influence model to obtain multiple associated control data subsets; Wherein, each associated control data subset includes several associated control data, and each associated control data is mapped with a corresponding associated feature, and the associated feature includes the delay duration, the comprehensive adjustment amount value, the adjustment cost required for the comprehensive adjustment amount value, and the adjustment duration of the corresponding associated control data; Generate the adjustment application degree of the corresponding associated control data subset according to the associated features of the associated control data in each associated control data subset; Eliminate the associated control data subsets with an adjustment application degree less than the preset adjustment application degree, and generate multiple to-be-determined sub-adjustment strategies for the corresponding monitoring indicator according to the associated control data and the corresponding associated features in the remaining associated control data subsets.

[0011] In some embodiments of the present application, the calculation formula of the adjustment application degree is: ; Wherein, Y is the adjustment application degree of the associated control data subset, y1 is the first adjustment application conversion coefficient, n is the total number of the associated control data in the corresponding associated control data subset, t1i is the delay duration of the i-th associated control data, is the preset extended duration threshold, bi is the weight coefficient of the i-th associated control data, y2 is the second adjustment application conversion coefficient, is the comprehensive adjustment amount value of the i-th associated control data, y3 is the third adjustment application conversion coefficient, is the adjustment duration corresponding to the comprehensive adjustment amount value of the i-th associated control data, is the preset adjustment duration threshold, y4 is the fourth adjustment application conversion coefficient, and mi is the adjustment cost corresponding to the comprehensive adjustment amount value of the i-th associated control data. is the preset adjustment cost threshold.

[0012] In some embodiments of the present application, conflict analysis and optimization are performed on multiple stator adjustment strategies to be determined to obtain the final adjustment strategy, including: Set the arrangement order of the stator adjustment strategies to be determined for the corresponding monitoring indicators according to the adjustment application degrees of the multiple stator adjustment strategies for the same monitoring indicator; Obtain the stator adjustment strategies to be determined for all current monitoring indicators and the corresponding arrangement order, and sequentially set the stator adjustment strategies to be determined for each monitoring indicator as the target stator adjustment strategies according to the arrangement order; Perform conflict analysis on the target stator adjustment strategies for different monitoring indicators. If the analysis result shows that there is a conflict, replace the target stator adjustment strategies for the corresponding monitoring indicators until there is no conflict; If there is a conflict in replacing all the target stator adjustment strategies, optimize the target stator adjustment strategies for the corresponding monitoring indicators until there is no conflict; Generate the final adjustment strategy according to the target stator adjustment strategies for all monitoring indicators without conflict.

[0013] In some embodiments of the present application, it further includes an intelligent control method for gas boilers based on DCS: Preset multiple monitoring indicators, obtain the characteristic monitoring data of each monitoring indicator, and generate the monitoring evaluation value for the corresponding monitoring indicator according to the characteristic monitoring data; Analyze the historical control logs of each monitoring indicator, determine the historical associated control data and the corresponding historical associated characteristics of each monitoring indicator, and construct the corresponding associated influence model; Judge whether to generate an adjustment instruction for the associated control data according to the current monitoring evaluation value. If so, generate the associated characteristics based on the associated influence model and the monitoring evaluation value; Generate the stator adjustment strategies to be determined for the corresponding associated control data according to the associated characteristics, perform conflict analysis and optimization on the multiple stator adjustment strategies to be determined, obtain the final adjustment strategy, and generate the current adjustment instruction according to the final adjustment strategy.

[0014] The intelligent control system and method for gas boilers based on DCS in the embodiments of the present application, compared with the prior art, have the beneficial effects that: By setting multiple monitoring indicators and determining the characteristic monitoring data of the monitoring indicators, generating a monitoring evaluation value according to the characteristic monitoring data and judging whether it is necessary to adjust the associated control data of the monitoring indicators. If necessary, generating multiple pending stator adjustment strategies, conducting conflict analysis and optimization on the multiple pending stator adjustment strategies, determining the final adjustment strategy, comprehensively monitoring and accurately evaluating the operation status of the gas boiler, and adjusting the corresponding control data according to the actual operation status, improving the control efficiency of the gas boiler control system, and ensuring the normal and stable operation of the gas boiler. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a schematic diagram of an intelligent control system for a gas boiler based on DCS in an embodiment of the present application; Figure 2 is a schematic flowchart of an intelligent control method for a gas boiler based on DCS in an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] The following further describes in detail the specific embodiments of the present application in conjunction with the drawings and embodiments. The following embodiments are used to illustrate the present application, but are not used to limit the scope of the present application.

[0017] In the description of the present application, it should be understood that the orientation or positional relationship indicated by the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying 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 construed as limiting the present application.

[0018] The terms "first" and "second" are only used for descriptive purposes, and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present application, unless otherwise specified, the meaning of "plurality" is two or more.

[0019] In the description of the present application, it should be noted that unless otherwise clearly defined and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected, or indirectly connected through an intermediate medium, and it may be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific situations.

[0020] Such asFigure 1 As shown in Figure 1 , the intelligent control system for gas boilers based on DCS in the embodiment of the present application includes: A setting module, configured to preset a plurality of monitoring indicators in advance, obtain characteristic monitoring data for each monitoring indicator, and generate a monitoring evaluation value for the corresponding monitoring indicator according to the characteristic monitoring data; A construction module, configured to analyze the historical control logs of each monitoring indicator, determine the historical associated control data and corresponding historical associated characteristics of each monitoring indicator, and construct a corresponding associated influence model; A judgment module, configured to judge whether to generate an adjustment instruction for the current associated control data according to the current monitoring evaluation value. If so, generate an associated characteristic based on the associated influence model and the monitoring evaluation value; An adjustment module, configured to generate a stator adjustment strategy to be determined for the corresponding associated control data according to the associated characteristic, perform conflict analysis and optimization on a plurality of stator adjustment strategies to be determined, obtain a final adjustment strategy, and generate a current adjustment instruction according to the final adjustment strategy.

[0021] In this embodiment, the preset monitoring indicators include boiler operation state monitoring indicators, feed water flow monitoring indicators, combustion efficiency monitoring indicators, flue gas emission monitoring indicators, etc. By monitoring and evaluating the characteristic monitoring data of a plurality of monitoring indicators in real time, the comprehensive monitoring and monitoring accuracy of gas boilers are improved, and the safe and stable operation of gas boilers is ensured.

[0022] In this embodiment, the characteristic monitoring data refers to the data that has a great influence on the evaluation value of the monitoring indicator, and the monitoring evaluation value refers to the value for evaluating the state of the corresponding monitoring indicator. The larger the monitoring evaluation value, the better the operation state of the corresponding monitoring indicator, and vice versa.

[0023] In this embodiment, the historical change amount value and historical change trend of the historical monitoring evaluation value before and after control are extracted from the historical control log, and the historical control data with the historical adjustment amount greater than the preset adjustment amount threshold within the preset time period before the historical change time node of the historical monitoring evaluation value is extracted. The historical adjustment strategy includes the historical adjustment amount, historical adjustment cost, and historical adjustment time of the associated control data, and the corresponding historical control data is set as the associated control data.

[0024] In this embodiment, the historical correlation feature refers to the correlation relationship between the historical change feature of the historical monitoring evaluation value of the monitoring index and the historical adjustment strategy of the corresponding associated control data, that is, the corresponding relationship between the historical adjustment amount of the corresponding associated control data and the historical change amount value and historical change trend of the historical monitoring evaluation value, and is also associated with the historical adjustment time and historical adjustment cost in the corresponding historical adjustment strategy, laying a foundation for subsequent selection of the final adjustment strategy, improving the control efficiency of the intelligent control system and reducing the control cost, and ensuring the safe and stable operation of the gas boiler.

[0025] In some embodiments of the present application, the judgment module further includes: If not, obtain the monitoring evaluation value of the corresponding monitoring index in the current monitoring period, and construct the monitoring evaluation value change curve of the current period; Obtain the change feature of the monitoring evaluation value change curve, and perform curve extrapolation on the monitoring evaluation value change curve according to the change feature to obtain multiple predicted monitoring evaluation values in the future period; Compare the predicted monitoring evaluation values with the preset monitoring evaluation value threshold of the corresponding monitoring index to obtain multiple predicted monitoring evaluation value differences; Obtain the predicted change feature of multiple predicted monitoring evaluation values in the future period, and generate the predicted trend coefficient of the corresponding monitoring index according to the predicted change feature and the predicted monitoring evaluation value difference; Preset the predicted trend coefficient threshold in advance; If the predicted trend coefficient is less than the predicted trend coefficient threshold, generate a warning instruction for the corresponding monitoring index, and predict the predicted failure time point of the corresponding monitoring index; Input multiple predicted monitoring evaluation values before the predicted failure time point into the corresponding associated influence model to obtain multiple predicted correlation features, and obtain multiple predicted sub-adjustment strategies for the corresponding monitoring index according to the multiple predicted correlation features; Perform conflict analysis and optimization on several predicted sub-adjustment strategies in the future period to determine the predicted adjustment strategy in the future period, and generate an adjustment instruction for the future period according to the predicted adjustment strategy.

[0026] In some embodiments of the present application, obtaining the characteristic monitoring data of each monitoring index includes: Obtain the historical monitoring log of each monitoring index, extract the historical monitoring evaluation value at each historical monitoring time node in the historical monitoring log, and construct the historical monitoring evaluation value change curve of the corresponding historical monitoring log; Collect the historical monitoring data of the corresponding monitoring index in the historical monitoring log according to the historical monitoring time node, and map it into the corresponding historical monitoring evaluation value change curve to obtain the historical monitoring evaluation value - monitoring data change relationship diagram; Obtain the first fluctuation characteristics of the historical monitoring evaluation values in the historical monitoring evaluation value change curve, where the first fluctuation characteristics include the first fluctuation degree and the first fluctuation magnitude; Filter out the historical monitoring time nodes whose first fluctuation degree is greater than the preset fluctuation degree threshold, and set them as the concerned monitoring time nodes, and intercept the influence periods of each concerned monitoring time node in the same historical monitoring evaluation value - monitoring data change relationship graph; Obtain the second fluctuation characteristics of each historical monitoring data in each influence period, where the second fluctuation characteristics include the second fluctuation degree and the second fluctuation magnitude; Filter out the historical monitoring data whose second fluctuation degree is greater than the preset fluctuation degree threshold, and respectively construct a second fluctuation degree sequence and a second fluctuation magnitude sequence according to the second fluctuation degree and the second fluctuation magnitude of the corresponding historical monitoring data in multiple influence periods; Respectively construct a first fluctuation degree sequence and a first fluctuation magnitude sequence according to the first fluctuation degree and the first fluctuation magnitude of the first fluctuation characteristics of the historical monitoring evaluation value at multiple concerned monitoring time nodes; Compare the second fluctuation degree sequence and the second fluctuation magnitude sequence of each historical monitoring data with the first fluctuation degree sequence and the first fluctuation magnitude sequence of the historical monitoring evaluation value respectively to obtain the first sequence similarity and the second sequence similarity; Generate the correlation coefficient between the corresponding historical monitoring data and the historical monitoring evaluation value according to the first sequence similarity and the second sequence similarity; The calculation formula of the correlation coefficient is: ; Where G is the correlation coefficient, g0 is the correlation conversion coefficient, L1 is the first sequence similarity, a1 is the weight coefficient of the first sequence similarity, a2 is the weight coefficient of the second sequence similarity, and L2 is the second sequence similarity; Preset the correlation coefficient threshold in advance; If the correlation coefficient is greater than the correlation coefficient threshold, set the corresponding historical monitoring data as the characteristic monitoring data, and set the weight coefficient of the corresponding characteristic monitoring data according to the difference of the correlation coefficients greater than the correlation coefficient threshold.

[0027] In this embodiment, the historical monitoring evaluation value is obtained by evaluating the pre-selected representative historical monitoring data. Other historical monitoring data with an associated degree with the historical monitoring evaluation value are obtained according to the historical monitoring evaluation value - monitoring data change relationship graph and set as the characteristic monitoring data. The monitoring evaluation value is generated according to the characteristic monitoring data and the corresponding weight coefficient, laying the calculation basis for the monitoring evaluation value, improving the accuracy of the monitoring evaluation value, laying the foundation for subsequent judgment on whether to generate an adjustment instruction, and improving the control efficiency of the gas boiler.

[0028] In this embodiment, the influence period refers to 15 minutes between the concerned monitoring time nodes. By analyzing the second fluctuation characteristics of the historical monitoring data in the influence period, the historical monitoring data that affects the historical monitoring evaluation value is screened out, that is, the characteristic monitoring data.

[0029] In this embodiment, the first sequence similarity refers to the similarity of the change of the first fluctuation degree in the first fluctuation degree sequence with the corresponding second fluctuation degree in the second fluctuation degree sequence. When the first fluctuation degree changes with the change of the corresponding second fluctuation degree and the number of similar changes is more, the corresponding first sequence similarity is greater, and vice versa.

[0030] In this embodiment, the second sequence similarity refers to the similarity of the change of the first fluctuation magnitude in the first fluctuation magnitude sequence with the corresponding second fluctuation magnitude in the second fluctuation magnitude sequence. When the first fluctuation magnitude changes with the change of the corresponding second fluctuation magnitude and the number of similar changes is more, the corresponding second sequence similarity is greater, and vice versa.

[0031] In this embodiment, the correlation conversion coefficient refers to converting the first sequence similarity and the second sequence similarity into values with the same dimension as the correlation coefficient. When the first sequence similarity is greater and the second sequence similarity is greater, the corresponding correlation coefficient is greater, and vice versa.

[0032] In this embodiment, by calculating the correlation coefficient between each historical monitoring data and the historical monitoring evaluation value, the corresponding characteristic monitoring data is determined, providing a calculation basis for the subsequent monitoring evaluation value and improving the calculation accuracy, so as to accurately evaluate whether an adjustment instruction needs to be generated for the control data, improve the control efficiency of the gas boiler, and ensure the normal and stable operation of the gas boiler.

[0033] In some embodiments of the present application, generating a monitoring evaluation value of a corresponding monitoring index according to the characteristic monitoring data includes: Presetting a number of preset sub-data intervals for the characteristic monitoring data of each monitoring index, and each preset sub-data interval is mapped with a corresponding preset sub-monitoring evaluation value; Comparing each characteristic monitoring data with the corresponding number of preset sub-data intervals to obtain the preset sub-data interval where the corresponding characteristic monitoring data is located, and setting the preset sub-monitoring evaluation value of the preset sub-data interval where it is located as the sub-monitoring evaluation value of the corresponding characteristic monitoring data; Generating a monitoring evaluation value of the corresponding monitoring index according to the sub-monitoring evaluation values of all the characteristic monitoring data of the same monitoring index and the weight coefficients of the corresponding characteristic monitoring data.

[0034] In this embodiment, the preset sub-data interval is set in advance, including sub-data intervals corresponding to the characteristic monitoring data in multiple states such as good, normal, suspected abnormal, abnormal, and faulty, and corresponding preset sub-monitoring evaluation values are configured. When the state is better, the corresponding preset sub-monitoring evaluation value is larger, and vice versa.

[0035] In some embodiments of the present application, determining the historical associated control data and corresponding historical associated features of each monitoring index includes: Obtain the historical control logs of each monitoring index, extract the historical monitoring evaluation values and historical control data of the corresponding monitoring index in each historical control log according to the historical monitoring time nodes, and construct a historical monitoring evaluation value-control data change relationship graph for each historical control log. Among them, the historical monitoring evaluation value-control data change relationship graph includes a historical monitoring evaluation value change curve and several historical control data change curves; Preset a monitoring evaluation value threshold in advance; Screen out the first historical monitoring time node in the same historical monitoring evaluation value-control data change relationship graph where the historical monitoring evaluation value is less than the monitoring evaluation value threshold, and set it as the initial monitoring time node. Obtain the first historical monitoring time node after the initial monitoring time node where the historical monitoring evaluation value is greater than the monitoring evaluation value threshold, and set it as the end monitoring time node; Intercept the historical monitoring evaluation value change curve segment and the historical control data change curve segment between the initial monitoring time node and the end monitoring time node in the same historical monitoring evaluation value-control data change relationship graph; Analyze the intercepted historical monitoring evaluation value change curve segment to determine the historical monitoring evaluation value difference and the positive trend points in the historical monitoring evaluation value change curve segment; Obtain the historical first adjustment value of the historical control data of each historical control data change curve segment between the positive trend point and the initial monitoring time node. Set the historical control data with the historical first adjustment value greater than the preset adjustment value threshold as the historical associated control data corresponding to the historical monitoring evaluation value difference, and calculate the historical delay duration between the associated control data and the historical monitoring evaluation value; Based on several historical associated control data, construct a subset of historical associated control data corresponding to the historical monitoring evaluation value difference; Obtain the historical second adjustment value of the historical associated control data between the positive trend point and the end monitoring time node, and generate the historical comprehensive adjustment value of the corresponding historical associated control data according to the first historical adjustment value and the second historical adjustment value; Set the historical delay duration, the historical comprehensive adjustment value, the historical adjustment cost required for the historical comprehensive adjustment value, and the historical adjustment duration as the historical association features corresponding to the difference between the corresponding historical association control data and the corresponding historical monitoring evaluation value; Classify the historical association control data subsets corresponding to the differences in multiple historical monitoring evaluation values of the same monitoring indicator and the corresponding historical association features to obtain a set of historical association control data for multiple historical monitoring evaluation value difference intervals, where the set of historical association control data includes several historical association control data subsets.

[0036] In this embodiment, the positive trend point is the initial monitoring time node with an upward trend in the change curve segment of the historical monitoring evaluation value. The historical delay duration is the time difference between the adjustment time node of the historical association control data between the positive trend point and the initial monitoring time node and the positive trend point.

[0037] In this embodiment, there are several historical association control data corresponding to each historical monitoring evaluation value difference, and the number of the several historical association control data is at least one. The historical monitoring evaluation value difference interval means that the difference between historical monitoring evaluation value differences is less than a preset difference threshold.

[0038] In this embodiment, by determining the historical association control data subsets corresponding to multiple historical monitoring evaluation value differences and the corresponding historical association features, it lays a foundation for constructing the association influence model of each monitoring indicator in the follow-up, and quickly determines the adjustment strategy of the control data through the association influence model, improving the control efficiency and control effect, and ensuring the safe and stable operation of the gas boiler.

[0039] In some embodiments of the present application, constructing the corresponding association influence model includes: Based on the historical monitoring evaluation value differences in the historical monitoring evaluation value difference intervals of each monitoring indicator as the training input data, and based on several historical association control data and the corresponding historical association features in the historical association control data subsets corresponding to each historical monitoring evaluation value difference as the training output data; Perform neural network training based on the training input data and the training output data to obtain the association influence model between the historical monitoring evaluation value difference and the historical association control data in the corresponding monitoring indicator.

[0040] In this embodiment, by constructing the association influence model to determine the association features between different monitoring evaluation value differences and the association control data of each monitoring indicator, it lays a foundation for determining the adjustment strategy of the association control data of the current monitoring evaluation value difference of each monitoring indicator in the follow-up, improves the control efficiency, and ensures the safe and stable operation of the gas boiler.

[0041] In some embodiments of the present application, it is determined whether to generate an adjustment instruction for associated control data according to the current monitoring evaluation value. If so, associated features are generated based on the associated influence model and the monitoring evaluation value, including: If the current monitoring evaluation value is less than the monitoring evaluation value threshold, an adjustment instruction for the associated control data of the corresponding monitoring index is generated; Calculate the monitoring evaluation value difference between the monitoring evaluation value of the corresponding monitoring index and the monitoring evaluation value threshold, and input the monitoring evaluation value difference into the corresponding associated influence model to obtain multiple subsets of associated control data; Among them, each subset of associated control data includes several associated control data, and each associated control data is mapped with a corresponding associated feature. The associated feature includes the delay duration, comprehensive adjustment amount value, adjustment cost required for the comprehensive adjustment amount value, and adjustment duration of the corresponding associated control data; Generate the adjustment application degree of the corresponding subset of associated control data according to the associated features of the associated control data in each subset of associated control data; Eliminate the subsets of associated control data with an adjustment application degree less than the preset adjustment application degree, and generate multiple pending sub-adjustment strategies for the corresponding monitoring index according to the associated control data and the corresponding associated features in the remaining subsets of associated control data.

[0042] In some embodiments of the present application, the calculation formula of the adjustment application degree is: ; Among them, Y is the adjustment application degree of the subset of associated control data, y1 is the first adjustment application conversion coefficient, n is the total number of associated control data in the corresponding subset of associated control data, t1i is the delay duration of the i-th associated control data, is the preset extended duration threshold, bi is the weight coefficient of the i-th associated control data, y2 is the second adjustment application conversion coefficient, is the comprehensive adjustment amount value of the i-th associated control data, y3 is the third adjustment application conversion coefficient, is the adjustment duration corresponding to the comprehensive adjustment amount value of the i-th associated control data, is the preset adjustment duration threshold, y4 is the fourth adjustment application conversion coefficient, mi is the adjustment cost corresponding to the comprehensive adjustment amount value of the i-th associated control data, is the preset adjustment cost threshold.

[0043] In this embodiment, the first adjustment application conversion coefficient, the second adjustment application conversion coefficient, the third adjustment application conversion coefficient, and the fourth adjustment application conversion coefficient respectively refer to converting the delay duration, the comprehensive adjustment magnitude, the adjustment duration, and the adjustment cost into values with the same dimension as the adjustment application degree. When the delay duration is shorter, the comprehensive adjustment magnitude is smaller, the adjustment duration is shorter, and the adjustment cost is shorter, the corresponding adjustment application degree is larger, and vice versa.

[0044] In this embodiment, by calculating the adjustment application degree, the pending stator adjustment strategies for the associated control data of each monitoring index are screened out, that is, the adjustment strategies with short adjustment duration, fast influence speed on the corresponding monitoring evaluation value difference, and low adjustment cost, so as to improve the control efficiency of the gas boiler control system and reduce the cost, and ensure the normal and stable operation of the gas boiler.

[0045] In some embodiments of the present application, conflict analysis and optimization are performed on multiple pending stator adjustment strategies to obtain the final adjustment strategy, including: Set the arrangement order of the pending stator adjustment strategies of the corresponding monitoring index according to the adjustment application degree of the multiple pending stator adjustment strategies of the same monitoring index; Obtain the pending stator adjustment strategies of all current monitoring indexes and the corresponding arrangement order, and set the pending stator adjustment strategies of each monitoring index as the target pending stator adjustment strategies in turn according to the arrangement order; Perform conflict analysis on the target pending stator adjustment strategies of different monitoring indexes. If the analysis result shows that there is a conflict, replace the target pending stator adjustment strategies of the corresponding monitoring indexes until there is no conflict; If conflicts exist in all replaced target pending stator adjustment strategies, optimize the target pending stator adjustment strategies of the corresponding monitoring indexes until there is no conflict; Generate the final adjustment strategy according to the target pending stator adjustment strategies of all monitoring indexes without conflict.

[0046] In this embodiment, conflict analysis means that multiple pending self-adjustment strategies of different monitoring indexes conflict with each other at the current monitoring time node. When there is a conflict, the target pending stator adjustment strategies of the monitoring indexes with smaller weight coefficients are replaced in turn until there is no conflict. If conflicts exist in all the pending stator adjustment strategies of the corresponding monitoring indexes, optimize the pending stator adjustment strategies to make the control adjustment effect the best and the cost control the lowest, that is, the final adjustment strategy, so as to improve the control efficiency and control effect of the gas boiler.

[0047] In some embodiments of the present application, as Figure 2 shown, it further includes a gas boiler intelligent control method based on DCS: Step S201: Preset multiple monitoring indicators in advance, obtain the characteristic monitoring data of each monitoring indicator, and generate a monitoring evaluation value for the corresponding monitoring indicator according to the characteristic monitoring data; Step S202: Analyze the historical control logs of each monitoring indicator, determine the historical associated control data and corresponding historical associated characteristics of each monitoring indicator, and construct a corresponding associated influence model; Step S203: Judge whether to generate an adjustment instruction for the associated control data according to the current monitoring evaluation value. If so, generate an associated characteristic based on the associated influence model and the monitoring evaluation value; Step S204: Generate a to-be-determined adjustment strategy for the corresponding associated control data according to the associated characteristic, perform conflict analysis and optimization on multiple to-be-determined adjustment strategies to obtain the final adjustment strategy, and generate the current adjustment instruction according to the final adjustment strategy.

[0048] The above are only the preferred embodiments of the present application. It should be noted that for those of ordinary skill in the art in the technical field of the present application, without departing from the technical principle of the present application, several improvements and replacements can still be made, and these improvements and replacements should also be regarded as the protection scope of the present application.

Claims

1. The intelligent control system of gas boiler based on DCS is characterized by: include: A setting module is used to pre-set multiple monitoring indicators, obtain characteristic monitoring data of each monitoring indicator, and generate a monitoring evaluation value of the corresponding monitoring indicator according to the characteristic monitoring data; A construction module is used to analyze the historical control log of each monitoring indicator, determine the historical correlation control data and corresponding historical correlation features of each monitoring indicator, and construct a corresponding correlation impact model; A judgment module, used to judge whether to generate an adjustment instruction for the current associated control data according to the current monitoring evaluation value, and if so, generate an associated feature based on the associated impact model and the monitoring evaluation value; The adjustment module is used to generate a pending sub-adjustment strategy corresponding to the associated control data according to the associated features, perform conflict analysis and optimization on multiple pending sub-adjustment strategies, obtain a final adjustment strategy, and generate a current adjustment instruction according to the final adjustment strategy.

2. The DCS-based intelligent control system for gas boilers according to claim 1, characterized in that: The judging module further includes: If not, obtain the monitoring evaluation value of the corresponding monitoring indicator in the current monitoring period, and construct a monitoring evaluation value change curve for the current period; Obtaining the change characteristics of the monitoring and evaluation value change curve, and performing curve extrapolation on the monitoring and evaluation value change curve according to the change characteristics to obtain multiple predicted monitoring and evaluation values ​​for future time periods; Comparing the predicted monitoring evaluation value with the preset monitoring evaluation value threshold of the corresponding monitoring indicator to obtain a plurality of predicted monitoring evaluation value differences; Obtain the predicted change characteristics of multiple predicted monitoring and evaluation values ​​in the future period, and generate the predicted trend coefficient of the corresponding monitoring indicator according to the predicted change characteristics and the difference between the predicted monitoring and evaluation values; Pre-set the prediction trend coefficient threshold; If the predicted trend coefficient is less than the predicted trend coefficient threshold, an early warning instruction for the corresponding monitoring indicator is generated, and the predicted failure time point of the corresponding monitoring indicator is predicted; Inputting multiple predicted monitoring evaluation values ​​before the predicted fault time point into the corresponding correlation impact model to obtain multiple predicted correlation features, and obtaining multiple predicted pending sub-adjustment strategies of the corresponding monitoring indicators according to the multiple predicted correlation features; Conflict analysis and optimization are performed on several predicted undetermined sub-adjustment strategies for future time periods, the predicted adjustment strategy for future time periods is determined, and adjustment instructions for future time periods are generated according to the predicted adjustment strategy.

3. The DCS-based intelligent control system for gas boilers according to claim 2 is characterized in that: Obtain characteristic monitoring data for each monitoring indicator, including: Obtain the historical monitoring log of each monitoring indicator, extract the historical monitoring evaluation value at each historical monitoring time node in the historical monitoring log, and construct a historical monitoring evaluation value change curve corresponding to the historical monitoring log; Collect historical monitoring data of corresponding monitoring indicators in historical monitoring logs according to historical monitoring time nodes, and map them to corresponding historical monitoring evaluation value change curves to obtain a historical monitoring evaluation value-monitoring data change relationship diagram; Acquire a first fluctuation feature of a historical monitoring and evaluation value in a historical monitoring and evaluation value change curve, wherein the first fluctuation feature includes a first fluctuation degree and a first fluctuation magnitude; Filter out the historical monitoring time nodes whose first fluctuation degree is greater than the preset fluctuation degree threshold, and set them as the monitoring time nodes of interest, and intercept the influence period of each monitoring time node of interest in the same historical monitoring evaluation value-monitoring data change relationship diagram; Acquire a second fluctuation feature of each historical monitoring data in each impact period, wherein the second fluctuation feature includes a second fluctuation degree and a second fluctuation value; Filter out historical monitoring data whose second fluctuation degree is greater than a preset fluctuation degree threshold, and construct a second fluctuation degree sequence and a second fluctuation value sequence according to the second fluctuation degrees and second fluctuation values ​​of the corresponding historical monitoring data in multiple impact time periods; According to the first fluctuation degree and the first fluctuation magnitude of the first fluctuation feature of the historical monitoring evaluation value at a plurality of monitoring time nodes of interest, respectively construct a first fluctuation degree sequence and a first fluctuation magnitude sequence; Compare the second fluctuation degree sequence and the second fluctuation magnitude sequence of each historical monitoring data with the first fluctuation degree sequence and the first fluctuation magnitude sequence of the historical monitoring evaluation value, respectively, to obtain the first sequence similarity and the second sequence similarity; Generate a correlation coefficient between the corresponding historical monitoring data and the historical monitoring evaluation value according to the first sequence similarity and the second sequence similarity; The calculation formula of the correlation coefficient is: ； Where G is the correlation coefficient, g0 is the correlation conversion coefficient, L1 is the first sequence similarity, a1 is the weight coefficient of the first sequence similarity, a2 is the weight coefficient of the second sequence similarity, and L2 is the second sequence similarity; Pre-set correlation coefficient threshold; If the correlation coefficient is greater than the correlation coefficient threshold, the corresponding historical monitoring data is set as the characteristic monitoring data, and the weight coefficient of the corresponding characteristic monitoring data is set according to the correlation coefficient difference value that is greater than the correlation coefficient threshold.

4. The DCS-based intelligent control system for gas boilers according to claim 3 is characterized in that: Generate monitoring evaluation values ​​of corresponding monitoring indicators based on characteristic monitoring data, including: A number of preset sub-data intervals of characteristic monitoring data of each monitoring indicator are preset, and each preset sub-data interval is mapped with a corresponding preset sub-monitoring evaluation value; Compare each characteristic monitoring data with a corresponding number of preset sub-data intervals to obtain the preset sub-data interval where the corresponding characteristic monitoring data is located, and set the preset sub-monitoring evaluation value of the preset sub-data interval as the sub-monitoring evaluation value of the corresponding characteristic monitoring data; The monitoring evaluation value of the corresponding monitoring indicator is generated according to the sub-monitoring evaluation values ​​of all characteristic monitoring data of the same monitoring indicator and the weight coefficient of the corresponding characteristic monitoring data.

5. The DCS-based intelligent control system for gas boilers according to claim 4 is characterized in that: Determine the historical correlation control data and corresponding historical correlation characteristics of each monitoring indicator, including: Obtain the historical control log of each monitoring indicator, extract the historical monitoring evaluation value and historical control data of the corresponding monitoring indicator in each historical control log according to the historical monitoring time node, and construct the historical monitoring evaluation value-control data change relationship diagram of each historical control log. Wherein, the historical monitoring evaluation value-control data change relationship diagram includes a historical monitoring evaluation value change curve and a number of historical control data change curves; Pre-set monitoring and evaluation value thresholds; Filter out the first historical monitoring time node whose historical monitoring evaluation value is less than the monitoring evaluation value threshold value in the same historical monitoring evaluation value-control data change relationship diagram, and set it as the initial monitoring time node; obtain the first historical monitoring time node whose historical monitoring evaluation value after the initial monitoring time node is greater than the monitoring evaluation value threshold value, and set it as the terminal monitoring time node; Intercept the historical monitoring evaluation value change curve segment and the historical control data change curve segment between the initial monitoring time node and the terminal monitoring time node of the same historical monitoring evaluation value-control data change relationship diagram; Analyze the intercepted historical monitoring and evaluation value change curve segments to determine the historical monitoring and evaluation value difference and the positive trend point in the historical monitoring and evaluation value change curve segments; Obtain the historical first adjustment value of the historical control data between the positive trend point and the initial monitoring time node of each historical control data change curve segment, set the historical control data whose historical first adjustment value is greater than the preset adjustment value threshold as the historical associated control data corresponding to the historical monitoring evaluation value difference, and calculate the historical delay time between the associated control data and the historical monitoring evaluation value; Constructing a subset of historical associated control data corresponding to the difference of historical monitoring evaluation values ​​based on a number of historical associated control data; Obtain a historical second adjustment value of the historical correlation control data between the positive trend point and the terminal monitoring time node, and generate a historical comprehensive adjustment value of the corresponding historical correlation control data according to the first historical adjustment value and the second historical adjustment value; The historical delay duration, the historical comprehensive adjustment value, the historical adjustment cost required for the historical comprehensive adjustment value, and the historical adjustment duration are set as the historical correlation characteristics of the difference between the corresponding historical correlation control data and the corresponding historical monitoring evaluation value; The historical correlation control data subsets and corresponding historical correlation features of multiple historical monitoring evaluation value differences of the same monitoring indicator are classified to obtain a historical correlation control data set of multiple historical monitoring evaluation value difference intervals, wherein the historical correlation control data set includes several historical correlation control data subsets.

6. The DCS-based intelligent control system for gas boilers according to claim 5, characterized in that: Construct the corresponding correlation impact model, including: The historical monitoring evaluation value difference in the historical monitoring evaluation value difference interval of each monitoring indicator is used as training input data, and the historical correlation control data and the corresponding historical correlation features in the historical correlation control data subset based on each historical monitoring evaluation value difference are used as training output data; The neural network is trained based on the training input data and the training output data to obtain the correlation influence model between the historical monitoring evaluation value difference and the historical correlation control data in the corresponding monitoring index.

7. The DCS-based intelligent control system for gas boilers according to claim 6, characterized in that: According to the current monitoring and evaluation value, it is determined whether to generate an adjustment instruction for the associated control data. If so, an associated feature is generated based on the associated impact model and the monitoring and evaluation value, including: If the current monitoring evaluation value is less than the monitoring evaluation value threshold, an adjustment instruction for the associated control data of the corresponding monitoring indicator is generated; Calculate the difference between the monitoring evaluation value of the corresponding monitoring indicator and the monitoring evaluation value threshold, input the monitoring evaluation value difference into the corresponding association influence model, and obtain multiple association control data subsets; The associated control data subset includes a plurality of associated control data, and each associated control data is mapped with a corresponding associated feature, and the associated feature includes a delay duration, a comprehensive adjustment value, an adjustment cost required for the comprehensive adjustment value, and an adjustment duration of the corresponding associated control data; Generate an adjustment application degree of the corresponding association control data subset according to the association characteristics of the association control data in each association control data subset; The associated control data subsets whose adjustment application degree is less than the preset adjustment application degree are eliminated, and a plurality of pending sub-adjustment strategies corresponding to the monitoring indicators are generated according to the associated control data in the remaining associated control data subsets and the corresponding associated features.

8. The DCS-based intelligent control system for gas boilers according to claim 7, characterized in that: The calculation formula for the adjustment application degree is: ； Wherein, Y is the adjustment application degree of the associated control data subset, y1 is the first adjustment application conversion coefficient, n is the total number of associated control data in the corresponding associated control data subset, t1i is the delay time of the i-th associated control data, is the preset extension time threshold, bi is the weight coefficient of the i-th associated control data, y2 is the second adjustment application conversion coefficient, is the comprehensive adjustment value of the i-th associated control data, y3 is the third adjustment application conversion coefficient, is the adjustment duration corresponding to the comprehensive adjustment value of the i-th associated control data, is the preset adjustment duration threshold, y4 is the fourth adjustment application conversion coefficient, mi is the adjustment cost corresponding to the comprehensive adjustment value of the i-th associated control data, Adjust the cost threshold for the preset.

9. The DCS-based intelligent control system for gas boilers according to claim 8, characterized in that: Conflict analysis and optimization are performed on multiple pending sub-adjustment strategies to obtain the final adjustment strategy, including: According to the adjustment application degree of multiple pending sub-adjustment strategies of the same monitoring indicator, the arrangement order of the pending sub-adjustment strategies of the corresponding monitoring indicator is set; Obtain the pending sub-adjustment strategies of all current monitoring indicators and their corresponding arrangement order, and set the pending sub-adjustment strategy of each monitoring indicator as the target pending sub-adjustment strategy in turn according to the arrangement order; Conduct conflict analysis on the target adjustment strategies for different monitoring indicators. If the analysis result shows that there is a conflict, replace the target adjustment strategy for the corresponding monitoring indicator until there is no conflict. If there are conflicts when replacing all the target pending adjustment strategies, the target pending adjustment strategies of the corresponding monitoring indicators are optimized until there are no conflicts; The final adjustment strategy is generated according to the target pending adjustment strategies of all monitoring indicators without conflicts.

10. The intelligent control method of gas boiler based on DCS is characterized by: include: Preset multiple monitoring indicators, obtain characteristic monitoring data of each monitoring indicator, and generate monitoring evaluation values ​​of corresponding monitoring indicators according to the characteristic monitoring data; Analyze the historical control logs of each monitoring indicator, determine the historical correlation control data and corresponding historical correlation features of each monitoring indicator, and build the corresponding correlation impact model; Determine whether to generate an adjustment instruction for the associated control data according to the current monitoring and evaluation value, and if so, generate an associated feature based on the associated impact model and the monitoring and evaluation value; According to the associated features, a pending sub-adjustment strategy corresponding to the associated control data is generated, and conflict analysis and optimization are performed on multiple pending sub-adjustment strategies to obtain a final adjustment strategy, and the current adjustment instruction is generated according to the final adjustment strategy.

Images