Multi-target PID automatic control method and system based on flue gas waste heat recovery

Abstract

The invention provides a multi-target PID automatic control method and system based on flue gas waste heat recovery. The method comprises the steps that 1, temperature information and solution information generated when flue gas waste heat recovery work is executed on a work site are collected, and deep processing is conducted on the temperature information and the solution information through a deep learning model, the method comprises the steps of obtaining a plurality of information parameter values of a work site, calculating the information parameter values by utilizing a PID controller, deducing the flue gas waste heat recovery rate of the work site based on a calculation result, and when the flue gas waste heat recovery rate is unqualified, adjusting function parameters of a specified unit of the work site based on the calculation result. And the automatic adjustment data of the work site are collected, and the intelligent control information of the work site is constructed and displayed, so that the intelligent degree of the flue gas waste heat recovery control process is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of waste heat recovery, and particularly to a multi-objective PID automatic control method and system based on flue gas waste heat recovery. Background Art

[0002] Generally, a flue gas system includes a flue gas-water plate heat exchanger and its auxiliary facilities. The process of flue gas is as follows: The flue gas from the outlet of the low-temperature economizer of the waste heat boiler is sent into the plate heat exchanger placed in the flue. During the heat exchange process with the flue gas-water plate heat exchanger, the heat of the flue gas is absorbed by the intermediate water. After absorbing 40 MW of heat, the temperature drops to 30°C, and finally it enters the original chimney and is discharged into the atmosphere. Currently, most power plants use flue gas-water plate heat exchangers to carry out flue gas waste heat recovery work. Although the flue gas waste heat can be recovered, it is necessary to manually adjust the working parameters of the flue gas-water plate heat exchanger according to the actual operation conditions of the power plant, which may lead to untimely recovery sometimes, or sometimes being in the high-speed operation stage for a long time, resulting in waste of resources.

[0003] Therefore, the present invention provides a multi-objective PID automatic control method and system based on flue gas waste heat recovery. Summary of the Invention

[0004] The multi-objective PID automatic control method and system based on flue gas waste heat recovery of the present invention improve the degree of intelligence in the control process of flue gas waste heat recovery.

[0005] The present invention provides a multi-objective PID automatic control method based on flue gas waste heat recovery, including:

[0006] Step 1: Collect the temperature information and solution information generated during the flue gas waste heat recovery work at the work site;

[0007] Step 2: Use a deep learning model to deeply process the temperature information and the solution information to obtain several information parameter values of the work site;

[0008] Step 3: Use a PID controller to calculate the information parameter values, and deduce the flue gas waste heat recovery rate of the work site based on the calculation results;

[0009] Step 4: When the flue gas waste heat recovery rate is unqualified, adjust the function parameters of the designated unit at the work site based on the calculation results;

[0010] Step 5: Collect the automatic adjustment data of the work site, construct the intelligent control information of the work site and display it.

[0011] In an implementable manner,

[0012] The said Step 1 includes:

[0013] Step 11: Before performing flue gas waste heat recovery work at the work site, locate the on-site position of the flue gas waste heat recovery device at the work site, and add corresponding information collection devices for the work site at the on-site position.

[0014] Step 12: When performing flue gas waste heat recovery work at the work site, control each of the information collection devices to collect on-site information of the work site respectively.

[0015] Step 13: Analyze the on-site information to obtain the heat source temperature, cold source temperature, solution concentration, and solution temperature of the work site.

[0016] Step 14: Establish temperature information of the work site based on the heat source temperature and the cold source temperature, and establish solution information of the work site based on the solution concentration and the solution temperature.

[0017] In an implementable manner,

[0018] Step 2 includes:

[0019] Step 21: Perform hierarchical processing on the temperature information to obtain several sub-temperature information corresponding to each network layer, compare the sub-temperature information corresponding to the same network layer to generate the temperature change law of the work site, perform enhancement processing on the solution information to obtain the solution consumption law of the work site.

[0020] Step 22: Adjust the activation function of the deep learning model according to the temperature change law and the solution consumption law, input the temperature information and the solution information into the adjusted deep learning model for linear analysis to obtain several waste heat recovery linear relationships of the work site.

[0021] Step 23: Respectively use each waste heat recovery linear relationship to deduce the first parameter value range corresponding to the temperature information within a specified period and the second parameter value range corresponding to the solution information within a specified period.

[0022] Step 24: Perform parameter convergence training on the temperature information based on the first parameter range to obtain several temperature information parameter values of the work site, and perform parameter convergence training on the solution information based on the second parameter range to obtain several solution information parameter values of the work site.

[0023] In an implementable manner,

[0024] Step 23 includes:

[0025] Step 231: Based on the waste heat recovery linear relationship, construct a linear logic chart for the work site, make a first mark on the temperature information included in the linear logic chart, and make a second mark on the solution information included in the linear logic chart;

[0026] Step 232: Analyze all the first marks and all the second marks based on the overall logic information of the linear logic chart to obtain the temperature logic characteristics and solution logic characteristics of the site. Search for several current temperature values of the work site in the temperature information, and search for several current solution values of the work site in the solution information;

[0027] Step 233: Use the temperature logic characteristics to deduce the temperature transformation process of each current temperature value, construct a first parameter range for the work site within a specified period, use the solution logic characteristics to deduce the solution transformation process of each current solution value, and construct a second parameter range for the work site within a specified period.

[0028] In an implementable manner,

[0029] The said step 3 includes:

[0030] Step 31: Input the information parameter values into the PID controller, and use the feedback control algorithm to perform proportional control, integral control, and derivative control on the information parameter values respectively to obtain the proportional error, integral error, and derivative error of the work site;

[0031] Step 32: Adjust the numerical ratio between the information parameter values based on the proportional error, and perform comprehensive stability correction on each information parameter value based on the integral error and the derivative error respectively to generate the effective parameter values of the work site;

[0032] Step 33: Use the effective parameter values to deduce the flue gas waste heat generation law and flue gas waste heat recovery law of the work site, construct a flue gas waste heat circulation model for the work site, and capture several waste heat recovery moments of the work site in the flue gas waste heat circulation model;

[0033] Step 34: Conduct efficiency evaluation on each waste heat recovery moment respectively to generate the flue gas waste heat recovery rate of the work site.

[0034] In an implementable manner,

[0035] The said step 4 includes:

[0036] Step 41: When the flue gas waste heat recovery rate is unqualified, establish a parameter table of the work site according to the calculation results, identify several redundant flue gas positions included in the work site based on the parameter table, and determine the flue gas redundancy corresponding to each redundant flue gas position;

[0037] Step 42: Obtain the unit information corresponding to the redundant flue gas position, determine the specified unit to be adjusted, and obtain the current function parameters of the specified unit;

[0038] Step 43: Use the flue gas redundancy to correct the current function parameters to obtain the adaptation parameters corresponding to each specified unit;

[0039] Step 44: Input each adaptation parameter into the unit structure of the corresponding specified unit for training to obtain the synchronous adjustment parameters of each specified unit, and adjust the function parameters of the corresponding specified unit based on the adaptation parameters and the synchronous adjustment parameters.

[0040] In an implementable manner,

[0041] The said step 5 includes:

[0042] Step 51: During the process of adjusting the parameters of the specified unit, collect the automatic adjustment data of the work site and generate the adjustment progress information of the work site;

[0043] Step 52: Summarize the adjustment results corresponding to each unit time period in the adjustment progress information, generate the intelligent control information of the work site and display it.

[0044] In an implementable manner,

[0045] It also includes:

[0046] Perform auxiliary parameter adjustment on the corresponding specified unit according to the unit control instruction issued by the user, and deduce and display the unit function of the adjusted specified unit.

[0047] The present invention provides a multi-objective PID automatic control system based on flue gas waste heat recovery, including:

[0048] An information acquisition module, which is used to acquire the temperature information and solution information generated when the work site performs flue gas waste heat recovery work;

[0049] A deep processing module, which is used to perform deep processing on the temperature information and the solution information by using a deep learning model to obtain several information parameter values of the work site;

[0050] A calculation and derivation module, configured to calculate the information parameter values by using a PID controller and derive the flue gas waste heat recovery rate of the work site based on the calculation results;

[0051] A parameter adjustment module, configured to adjust the function parameters of a specified unit at the work site based on the calculation results when the flue gas waste heat recovery rate is unqualified;

[0052] An adjustment supervision module, configured to collect the automatic adjustment data of the work site, construct the intelligent control information of the work site and display it.

[0053] In an implementable manner,

[0054] The deep processing module includes:

[0055] A hierarchical processing unit, configured to perform hierarchical processing on the temperature information to obtain several sub-temperature information corresponding to each network layer, compare the sub-temperature information corresponding to the same network layer, generate the temperature change law of the work site, perform strengthening processing on the solution information to obtain the solution consumption law of the work site;

[0056] A deep learning unit, configured to adjust the activation function of the deep learning model according to the temperature change law and the solution consumption law, input the temperature information and the solution information into the adjusted deep learning model for linear analysis, and obtain several waste heat recovery linear relationships of the work site;

[0057] A parameter derivation unit, configured to respectively derive the first parameter value range corresponding to the temperature information within a specified period and the second parameter value range corresponding to the solution information within a specified period by using each waste heat recovery linear relationship;

[0058] A parameter training unit, configured to perform parameter convergence training on the temperature information based on the first parameter range to obtain several temperature information parameter values of the work site, and perform parameter convergence training on the solution information based on the second parameter range to obtain several solution information parameter values of the work site.

[0059] The achievable beneficial effects of the above technical solution are as follows: By deeply processing the temperature information and solution information generated during the flue gas waste heat recovery work at the work site to determine the information parameter values of the work site, further using a PID controller to calculate the information parameter values, and deriving the flue gas waste heat recovery rate of the work site. If the flue gas waste heat recovery rate is unqualified at this time, the function parameters of the unit for flue gas waste heat recovery at the work site are adjusted as needed. Finally, the automatic adjustment data of the work site is supervised to generate the intelligent control information of the work site. In this way, the flue gas waste heat recovery work at the work site can be supervised, the function parameters of the unit can be adjusted when the recovery rate is low and unqualified, and the adjustment link can be supervised to avoid over-adjustment. In this way, the intelligent level of the flue gas waste heat recovery control process can be improved, the manual participation can be reduced, the phenomenon that the unit is in a high-speed working state for a long time can be effectively avoided, and the economic performance of the flue gas waste heat recovery can be improved.

[0060] Other features and advantages of the present invention will be described in the following description, and in part will be obvious from the description, or will be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained by the structure specifically pointed out in the written description and the drawings.

[0061] The technical solution of the present invention will be further described in detail below through the drawings and embodiments. Description of the Drawings

[0062] The drawings are used to provide a further understanding of the present invention, and constitute a part of the description. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation to the present invention. In the drawings:

[0063] Figure 1 It is a schematic diagram of the working process of the multi-objective PID automatic control method based on flue gas waste heat recovery in the embodiment of the present invention;

[0064] Figure 2 It is a schematic diagram of the composition of the multi-objective PID automatic control system based on flue gas waste heat recovery in the embodiment of the present invention. Detailed Embodiments

[0065] The following describes the preferred embodiments of the present invention with reference to the drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0066] Embodiment 1 This embodiment provides a multi-objective PID automatic control method based on flue gas waste heat recovery, as Figure 1 shown, including:

[0067] Step 1: Collect the temperature information and solution information generated during the flue gas waste heat recovery work at the work site;

[0068] Step 2: Use a deep learning model to deeply process the temperature information and the solution information to obtain several information parameter values of the work site;

[0069] Step 3: Use a PID controller to calculate the information parameter values, and deduce the flue gas waste heat recovery rate of the work site based on the calculation results;

[0070] Step 4: When the flue gas waste heat recovery rate is unqualified, adjust the function parameters of the designated unit at the work site based on the calculation results;

[0071] Step 5: Collect the automatic adjustment data of the work site, construct the intelligent control information of the work site and display it.

[0072] In this example, the temperature information represents multiple pieces of information about temperature changes generated during the flue gas waste heat recovery work, and the solution information represents multiple pieces of information about solution changes generated during the flue gas waste heat recovery work;

[0073] In this example, the information parameter value represents the numerical value corresponding to each piece of information at the work site;

[0074] In this example, the flue gas waste heat recovery rate represents the ratio between the flue gas waste heat recovery amount and the total generation amount at the work site;

[0075] In this example, the function parameter represents the parameter set when the designated unit performs a function;

[0076] In this example, the flue gas waste heat recovery rate reaching 95% or more is qualified.

[0077] The working principle and beneficial effects of the above technical solution: By deeply processing the temperature information and solution information generated during the flue gas waste heat recovery work at the work site to determine the information parameter values of the work site, further using a PID controller to calculate the information parameter values, and deducing the flue gas waste heat recovery rate of the work site. If the flue gas waste heat recovery rate is unqualified at this time, adjust the function parameters of the unit related to flue gas waste heat recovery at the work site as needed. Finally, supervise the automatic adjustment data of the work site to generate the intelligent control information of the work site. In this way, the flue gas waste heat recovery work at the work site can be supervised, the function parameters of the unit can be adjusted when the recovery rate is low or unqualified, and the adjustment link can be supervised to avoid over-adjustment. In this way, the intelligent level of the flue gas waste heat recovery control process can be improved, the manual participation can be reduced, the phenomenon that the unit operates at a high speed for a long time can be effectively avoided, and the economic performance of the flue gas waste heat recovery can be improved.

[0078] Example 2

[0079] Based on Example 1, for the multi-objective PID automatic control method based on flue gas waste heat recovery, Step 1 includes:

[0080] Step 11: Before performing flue gas waste heat recovery work at the work site, search for the on-site location of the flue gas waste heat recovery device at the work site, and add corresponding information collection devices for the work site at the on-site location;

[0081] Step 12: When performing flue gas waste heat recovery work at the work site, control each information collection device to collect on-site information of the work site;

[0082] Step 13: Analyze the on-site information to obtain the heat source temperature, cold source temperature, solution concentration, and solution temperature of the work site;

[0083] Step 14: Establish the temperature information of the work site based on the heat source temperature and the cold source temperature, and establish the solution information of the work site based on the solution concentration and the solution temperature.

[0084] In this example, the heat source temperature represents the temperature of the heat source at the work site, and the cold source temperature represents the temperature of the cold source of the flue gas waste heat recovery unit;

[0085] In this example, the solution concentration represents the particulate matter concentration of the solution for waste heat recovery in the flue gas waste heat recovery unit, and the solution temperature represents the temperature of the solution for waste heat recovery in the flue gas waste heat recovery unit.

[0086] Working principle and beneficial effects of the above technical solution: In order to ensure the accuracy of information collection and improve the effect of automatic adjustment work, information collection devices are added to the corresponding on-site locations at the work site before flue gas waste heat recovery. When performing work at the work site, the on-site information of the work site is collected by using the information collection devices. In this way, the information of the work site can be collected at any time, laying a foundation for subsequent automatic control work.

[0087] Example 3

[0088] Based on Example 1, for the multi-objective PID automatic control method based on flue gas waste heat recovery, Step 2 includes:

[0089] Step 21: Perform hierarchical processing on the temperature information to obtain several sub-temperature information corresponding to each network layer, compare the sub-temperature information corresponding to the same network layer to generate the temperature change law of the work site, perform enhancement processing on the solution information to obtain the solution consumption law of the work site;

[0090] Step 22: Adjust the activation function of the deep learning model according to the temperature change law and the solution consumption law, input the temperature information and the solution information into the adjusted deep learning model for linear analysis, and obtain several waste heat recovery linear relationships at the work site;

[0091] Step 23: Respectively use each of the waste heat recovery linear relationships to deduce the range of the first parameter values corresponding to the temperature information within a specified period and the range of the second parameter values corresponding to the solution information within a specified period;

[0092] Step 24: Perform parameter convergence training on the temperature information based on the first parameter range to obtain several temperature information parameter values at the work site, and perform parameter convergence training on the solution information based on the second parameter range to obtain several solution information parameter values at the work site.

[0093] In this example, the temperature change law represents the law of temperature change at various locations in the work site, and the solution consumption law represents the law presented when the solution is consumed in the unit;

[0094] In this example, the waste heat recovery linear relationship represents the linear relationship regarding this work presented during the waste heat recovery of flue gas;

[0095] In this example, the specified period is 45 minutes;

[0096] In this example, the range of the first parameter values represents the reasonable range of temperature information change, that is, the temperatures within this range are all at normal levels, and the range of the second parameter values represents the reasonable range of solution data change, that is, the solution averages within this range are at normal levels;

[0097] In this example, the parameter convergence training represents the result of training the temperature information parameter values and the solution information parameter values to the corresponding parameter ranges.

[0098] Working principle and beneficial effects of the above technical solution: By performing hierarchical processing on temperature information to identify the temperature change law at the work site, using hierarchical processing technology to divide the temperature information into separate sub-temperature information, enhancing the stability of the comparison work, and at the same time strengthening the solution information to obtain the solution consumption law at the work site. Further, a deep learning model is used to perform linear analysis on the temperature information and solution information to determine the linear relationship of waste heat recovery at the work site, and then the parameter value ranges of the temperature and solution when the work site is operating normally are derived. Finally, the temperature information and solution information are subjected to parameter convergence training to obtain the parameter values of the temperature information and solution information at the work site. In this way, the temperature information and solution information at the work site can be digitized, facilitating observation and subsequent numerical analysis.

[0099] Example 4

[0100] Based on the multi-objective PID automatic control method for flue gas waste heat recovery in Example 3, step 23 includes:

[0101] Step 231: Construct a linear logic chart of the work site based on the waste heat recovery linear relationship, perform a first marking on the temperature information included in the linear logic chart, and perform a second marking on the solution information included in the linear logic chart;

[0102] Step 232: Analyze all the first markings and all the second markings based on the overall logic information of the linear logic chart to obtain the temperature logic characteristics and solution logic characteristics of the site, and find several current temperature values of the work site in the temperature information and several current solution values of the work site in the solution information;

[0103] Step 233: Use the temperature logic characteristics to deduce the temperature transformation process of each current temperature value, construct the first parameter range of the work site within a specified period, use the solution logic characteristics to deduce the solution transformation process of each current solution value, and construct the second parameter range of the work site within a specified period.

[0104] In this example, the first marking means performing the same type of marking on all temperature information, and the second marking means performing the same type of marking on all solution information.

[0105] Working principle and beneficial effects of the above technical solution: By constructing a linear logic diagram of the work site, and then retrieving the temperature logic with linear temperature and the solution logic of solution information therein to determine the current temperature value and the current solution value of the work site, and further combining the temperature logic and the solution logic to deduce the temperature change process and the solution change process, the first parameter range and the second parameter range of the work site are determined, which is convenient for subsequent parameter convergence training work.

[0106] Example 5

[0107] Based on Example 1, for the multi-objective PID automatic control method based on flue gas waste heat recovery, step 3 includes:

[0108] Step 31: Input the information parameter values into the PID controller, and use the feedback control algorithm to perform proportional control, integral control, and derivative control on the information parameter values respectively to obtain the proportional error, integral error, and derivative error of the work site;

[0109] Step 32: Adjust the numerical ratio between the information parameter values based on the proportional error, and perform comprehensive stability correction on each information parameter value based on the integral error and the derivative error respectively to generate the effective parameter values of the work site;

[0110] Step 33: Use the effective parameter values to deduce the flue gas waste heat generation law and the flue gas waste heat recovery law of the work site, construct the flue gas waste heat cycle model of the work site, and capture several waste heat recovery moments of the work site in the flue gas waste heat cycle model;

[0111] Step 34: Perform efficiency evaluation on each waste heat recovery moment respectively to generate the flue gas waste heat recovery rate of the work site.

[0112] In this example, proportional control means adjusting the numerical ratio between information parameter values so that the information parameter values are directly proportional to their errors;

[0113] In this example, integral control represents the result of eliminating the effect of the time integral of the error of the information parameter value;

[0114] In this example, derivative control represents the result of eliminating the oscillation of the information parameter value;

[0115] In this example, the waste heat recovery moment represents the time point for waste heat recovery;

[0116] In this example, efficiency evaluation represents the process of analyzing the waste heat recovery rate of each waste heat recovery moment.

[0117] Working principle and beneficial effects of the above technical solution: The PID controller is used to perform proportional control, integral control, and derivative control on the information parameter values, thereby adjusting the numerical ratio and comprehensive stability between the option parameters, so as to obtain the effective parameter values at the work site. Then, the law of flue gas waste heat generation and the law of flue gas waste heat recovery at the work site are deduced to establish a flue gas waste heat circulation model. Furthermore, the efficiency evaluation of the waste heat recovery moment at the work site is carried out to determine the flue gas waste heat recovery rate at the work site. In this way, the flue gas waste heat recovery process at the work site can be analyzed in detail, and the flue gas recovery status at different moments can be determined. Finally, the flue gas waste heat recovery rate at the work site is determined, achieving the effect of synchronous recovery and synchronous monitoring, and reducing the time delay of the automatic control work.

[0118] Example 6

[0119] Based on Example 1, for the multi-objective PID automatic control method based on flue gas waste heat recovery, Step 4 includes:

[0120] Step 41: When the flue gas waste heat recovery rate is unqualified, establish a parameter table of the work site according to the calculation results, identify several redundant flue gas positions included in the work site based on the parameter table, and determine the flue gas redundancy corresponding to each redundant flue gas position;

[0121] Step 42: Obtain the unit information corresponding to the redundant flue gas position, determine the specified unit to be adjusted, and obtain the current function parameters of the specified unit;

[0122] Step 43: Use the flue gas redundancy to correct the current function parameters to obtain the adaptation parameters corresponding to each specified unit;

[0123] Step 44: Input each adaptation parameter into the unit structure of the corresponding specified unit for training to obtain the synchronous adjustment parameters of each specified unit, and adjust the function parameters of the corresponding specified unit based on the adaptation parameters and the synchronous adjustment parameters.

[0124] In this example, the redundant flue gas position represents the position where flue gas waste heat accumulates;

[0125] In this example, the current function parameters represent the parameters currently set for the specified unit.

[0126] Working principle and beneficial effects of the above technical solution: When the flue gas waste heat recovery rate is unqualified, a parameter table of the work site is constructed to determine several redundant flue gas positions at the work site, and then each redundant flue gas position is analyzed to determine its designated unit and flue gas redundancy, and the current function parameters of the designated unit are determined. Further, the current function parameters are corrected to determine the adaptation parameters of the designated unit. Further, the function parameters of the designated unit are determined by training. In this way, the parameters of different units can be adjusted according to the actual situation of the work site to make them more suitable for the work site, achieving a tacit cooperation.

[0127] Example 7

[0128] Based on the multi-objective PID automatic control method for flue gas waste heat recovery in Example 1, step 5 includes:

[0129] Step 51: During the process of adjusting the parameters of the designated unit, collect the automatic adjustment data of the work site and generate the adjustment progress information of the work site;

[0130] Step 52: Summarize the adjustment results corresponding to each unit time period in the adjustment progress information, generate the intelligent control information of the work site and display it.

[0131] In this example, the unit time is 15 minutes.

[0132] Working principle and beneficial effects of the above technical solution: Supervise the automatic adjustment process of the work site and generate the intelligent control information of the work site for relevant personnel to view at any time.

[0133] Example 8

[0134] Based on Example 7, the multi-objective PID automatic control method for flue gas waste heat recovery further includes:

[0135] According to the unit control instruction issued by the user, perform auxiliary parameter adjustment on the corresponding designated unit, deduce the unit function of the adjusted designated unit and display it.

[0136] Working principle and beneficial effects of the above technical solution: When relevant personnel issue auxiliary parameter adjustment to the designated unit, first deduce the unit function according to the instruction and display it for relevant personnel to refer to.

[0137] Example 9

[0138] This embodiment provides a multi-objective PID automatic control system for flue gas waste heat recovery, as Figure 2 shown, including:

[0139] An information collection module for collecting temperature information and solution information generated during the flue gas waste heat recovery work at the work site:

[0140] A deep processing module for deeply processing the temperature information and the solution information by using a deep learning model to obtain several information parameter values of the work site;

[0141] A calculation and derivation module for calculating the information parameter values by using a PID controller and deriving the flue gas waste heat recovery rate of the work site based on the calculation results;

[0142] A parameter adjustment module for adjusting the function parameters of the specified unit at the work site based on the calculation results when the flue gas waste heat recovery rate is unqualified;

[0143] An adjustment supervision module for collecting the automatic adjustment data of the work site, constructing the intelligent control information of the work site and displaying it.

[0144] In this example, the temperature information represents multiple pieces of information about temperature changes generated during the flue gas waste heat recovery work, and the solution information represents multiple pieces of information about solution changes generated during the flue gas waste heat recovery work;

[0145] In this example, the information parameter value represents the numerical value corresponding to each piece of information at the work site;

[0146] In this example, the flue gas waste heat recovery rate represents the ratio between the flue gas waste heat recovery amount and the total generation amount at the work site;

[0147] In this example, the function parameter represents the parameter set when the specified unit executes a function;

[0148] In this example, the flue gas waste heat recovery rate reaching 95% or more is qualified.

[0149] Working principle and beneficial effects of the above technical solution: By deeply processing the temperature information and solution information generated during the flue gas waste heat recovery work at the work site to determine the information parameter values of the work site, and further using a PID controller to calculate the information parameter values, the flue gas waste heat recovery rate of the work site is deduced. If the flue gas waste heat recovery rate is unqualified at this time, the function parameters of the unit for flue gas waste heat recovery at the work site are adjusted as needed. Finally, the automatic adjustment data of the work site is supervised to generate the intelligent control information of the work site. In this way, the flue gas waste heat recovery work at the work site can be supervised, the function parameters of the unit can be adjusted when the recovery rate is low and unqualified, and the adjustment link can be supervised to avoid over-adjustment. In this way, the intelligent level of the flue gas waste heat recovery control process can be improved, the manual participation can be reduced, the phenomenon that the unit is in a high-speed working state for a long time can be effectively avoided, and the economic performance of the flue gas waste heat recovery can be improved.

[0150] Example 10

[0151] Based on the multi-objective PID automatic control system for flue gas waste heat recovery in Example 9, the deep processing module includes:

[0152] A hierarchical processing unit for hierarchically processing the temperature information to obtain several sub-temperature information corresponding to each network layer, comparing the sub-temperature information corresponding to the same network layer to generate the temperature change law of the work site, and strengthening the processing of the solution information to obtain the solution consumption law of the work site;

[0153] A deep learning unit for adjusting the activation function of the deep learning model according to the temperature change law and the solution consumption law, inputting the temperature information and the solution information into the adjusted deep learning model for linear analysis, and obtaining several waste heat recovery linear relationships of the work site;

[0154] A parameter derivation unit for respectively using each waste heat recovery linear relationship to derive the first parameter value range corresponding to the temperature information within a specified period and the second parameter value range corresponding to the solution information within a specified period;

[0155] A parameter training unit for performing parameter convergence training on the temperature information based on the first parameter range to obtain several temperature information parameter values of the work site, and performing parameter convergence training on the solution information based on the second parameter range to obtain several solution information parameter values of the work site.

[0156] In this example, the temperature change law represents the law of temperature changes at various locations in the work site, and the solution consumption law represents the law presented when the solution is consumed in the unit;

[0157] In this example, the linear relationship of waste heat recovery represents the linear relationship presented in the process of flue gas waste heat recovery for this work;

[0158] In this example, the specified period is 45 minutes;

[0159] In this example, the first parameter value range represents the reasonable range within which the temperature information changes, that is, the temperatures within this range are all within the normal level, and the second parameter value range represents the reasonable range within which the solution data changes, that is, the average value of the solution within this range is within the normal level;

[0160] In this example, the parameter convergence training represents the result of training the temperature information parameter value and the solution information parameter value to the corresponding parameter ranges.

[0161] The working principle and beneficial effects of the above technical solution: By performing hierarchical processing on the temperature information to identify the temperature change law at the work site, using the hierarchical processing technology to divide the temperature information into separate sub-temperature information, strengthening the stability of the comparison work, and at the same time strengthening the solution information to obtain the solution consumption law at the work site, further using a deep learning model to perform linear analysis on the temperature information and the solution information, determining the linear relationship of waste heat recovery at the work site, and then deriving the parameter value ranges of the temperature and the solution when the work site is operating normally, and finally performing parameter convergence training on the temperature information and the solution information to obtain the temperature information parameter value and the solution information parameter value at the work site. In this way, the temperature information and the solution information at the work site can be digitalized, which is convenient for observation and subsequent numerical analysis.

[0162] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these changes and modifications.

Claims

1. A multi-objective PID automatic control method based on flue gas waste heat recovery is characterized in that: include: Step 1: Collect the temperature information and solution information generated when performing flue gas waste heat recovery work at the work site; Step 2: Use a deep learning model to deeply process the temperature information and the solution information to obtain several information parameter values ​​of the work site; Step 3: Calculate the information parameter value using a PID controller, and derive the flue gas waste heat recovery rate of the work site based on the calculation result; Step 4: When the flue gas waste heat recovery rate is unqualified, adjusting the functional parameters of the designated unit at the work site based on the calculation result; Step 5: Collect the automatic adjustment data of the work site, construct and display the intelligent control information of the work site.

2. The multi-objective PID automatic control method based on flue gas waste heat recovery according to claim 1 is characterized in that: The step 1 comprises: Step 11: before performing the flue gas waste heat recovery work at the work site, find the on-site location of the flue gas waste heat recovery device at the work site, and add a corresponding information collection device for the work site in the on-site location; Step 12: When the flue gas waste heat recovery work is performed at the work site, each of the information collection devices is controlled to collect the on-site information of the work site; Step 13: Analyze the on-site information to obtain the heat source temperature, cold source temperature, solution concentration and solution temperature of the work site; Step 14: Establishing the temperature information of the work site according to the heat source temperature and the cold source temperature, and establishing the solution information of the work site according to the solution concentration and the solution temperature.

3. The multi-objective PID automatic control method based on flue gas waste heat recovery according to claim 1 is characterized in that: The step 2 comprises: Step 21: performing layered processing on the temperature information to obtain a plurality of sub-temperature information corresponding to each network layer, comparing the sub-temperature information corresponding to the same network layer, generating a temperature variation law of the work site, performing enhanced processing on the solution information, and obtaining a solution consumption law of the work site; Step 22: adjusting the activation function of the deep learning model according to the temperature variation law and the solution consumption law, inputting the temperature information and the solution information into the adjusted deep learning model for linear analysis, and obtaining several waste heat recovery linear relationships at the work site; Step 23: deriving a first parameter value range corresponding to the temperature information within a specified period and a second parameter value range corresponding to the solution information within a specified period by using each of the waste heat recovery linear relationships; Step 24: Based on the first parameter range, the temperature information is trained for parameter convergence to obtain several temperature information parameter values ​​of the work site; based on the second parameter range, the solution information is trained for parameter convergence to obtain several solution information parameter values ​​of the work site.

4. The multi-objective PID automatic control method based on flue gas waste heat recovery according to claim 3 is characterized in that: The step 23 comprises: Step 231: constructing a linear logic diagram of the work site based on the waste heat recovery linear relationship, performing a first mark on the temperature information contained in the linear logic diagram, and performing a second mark on the solution information contained in the linear logic diagram; Step 232: Analyze all the first marks and analyze all the second marks based on the overall logic information of the linear logic diagram to obtain the temperature logic characteristics and solution logic characteristics of the site, search for several current temperature values ​​of the work site in the temperature information, and search for several current solution values ​​of the work site in the solution information; Step 233: Utilize the temperature logic characteristic to derive the temperature transformation process of each current temperature value, and construct a first parameter range of the work site within a specified period; utilize the solution logic characteristic to derive the solution transformation process of each current solution value, and construct a second parameter range of the work site within a specified period.

5. The multi-objective PID automatic control method based on flue gas waste heat recovery according to claim 1 is characterized in that: The step 3 comprises: Step 31: input the information parameter value into the PID controller, and use the feedback control algorithm to perform proportional control, integral control and differential control on the information parameter value to obtain the proportional error, integral error and differential error of the work site; Step 32: adjusting the numerical ratio between the information parameter values ​​based on the proportional error, and performing comprehensive stability correction on each of the information parameter values ​​based on the integral error and the differential error to generate an effective parameter value for the work site; Step 33: using the effective parameter values ​​to deduce the flue gas waste heat generation law and the flue gas waste heat recovery law of the work site, constructing a flue gas waste heat cycle model of the work site, and capturing several waste heat recovery moments of the work site in the flue gas waste heat cycle model; Step 34: Evaluate the efficiency of each waste heat recovery moment to generate the flue gas waste heat recovery rate of the work site.

6. The multi-objective PID automatic control method based on flue gas waste heat recovery according to claim 1 is characterized in that: The step 4 comprises: Step 41: when the flue gas waste heat recovery rate is unqualified, a parameter table of the work site is established according to the calculation result, a plurality of redundant flue gas positions included in the work site are identified based on the parameter table, and a flue gas redundancy corresponding to each of the redundant flue gas positions is determined; Step 42: obtaining unit information corresponding to the redundant flue gas position, determining a designated unit to be adjusted, and obtaining current functional parameters of the designated unit; Step 43: using the flue gas redundancy to correct the current function parameters, and obtaining the adaptation parameters corresponding to each of the designated units; Step 44: Input each of the adaptation parameters into the unit structure of the corresponding designated unit for training, obtain the synchronous adjustment parameters of each of the designated units, and adjust the functional parameters of the corresponding designated unit based on the adaptation parameters and the synchronous adjustment parameters.

7. The multi-objective PID automatic control method based on flue gas waste heat recovery according to claim 1 is characterized in that: The step 5 comprises: Step 51: In the process of adjusting the parameters of the designated unit, collecting the automatic adjustment data of the work site and generating the adjustment progress information of the work site; Step 52: Summarize the adjustment results corresponding to each unit time period in the adjustment progress information, generate and display the intelligent control information of the work site.

8. The multi-objective PID automatic control method based on flue gas waste heat recovery according to claim 7 is characterized in that: Also includes: According to the unit control instruction issued by the user, auxiliary parameters of the corresponding designated unit are adjusted, and the unit function of the adjusted designated unit is derived and displayed.

9. The multi-objective PID automatic control system based on flue gas waste heat recovery is characterized by: include: The information collection module is used to collect the temperature information and solution information generated when the flue gas waste heat recovery work is performed at the work site; A deep processing module, used to perform deep processing on the temperature information and the solution information using a deep learning model to obtain a plurality of information parameter values ​​of the work site; A calculation and derivation module, used to calculate the information parameter value using a PID controller, and derive the flue gas waste heat recovery rate of the work site based on the calculation result; A parameter adjustment module, used for adjusting the functional parameters of the designated unit at the work site based on the calculation result when the flue gas waste heat recovery rate is unqualified; The adjustment supervision module is used to collect automatic adjustment data of the work site, construct intelligent control information of the work site and display it.

10. The multi-objective PID automatic control system based on flue gas waste heat recovery according to claim 9, characterized in that: The depth processing module comprises: A layered processing unit, used to perform layered processing on the temperature information to obtain a plurality of sub-temperature information corresponding to each network layer, compare the sub-temperature information corresponding to the same network layer, generate a temperature variation law of the work site, perform enhanced processing on the solution information, and obtain a solution consumption law of the work site; A deep learning unit, used to adjust the activation function of the deep learning model according to the temperature variation law and the solution consumption law, input the temperature information and the solution information into the adjusted deep learning model for linear analysis, and obtain a plurality of waste heat recovery linear relationships at the work site; A parameter derivation unit, used to derive a first parameter value range corresponding to the temperature information within a specified period and a second parameter value range corresponding to the solution information within a specified period by using each of the waste heat recovery linear relationships; A parameter training unit is used to perform parameter convergence training on the temperature information based on the first parameter range to obtain several temperature information parameter values ​​of the working site, and to perform parameter convergence training on the solution information based on the second parameter range to obtain several solution information parameter values ​​of the working site.

Images