A full-load desulfurization wastewater zero discharge system and method

Abstract

This invention relates to the field of zero discharge technology for desulfurization wastewater, and discloses a full-load desulfurization wastewater zero discharge system and method. The system includes: a data acquisition module for constructing a first control parameter group, a second control parameter group, and a third control parameter group for the full load of the unit, and using them as training data for neural network training to obtain corresponding control parameter prediction models; a generation module for acquiring real-time load values ​​and inputting them into the control parameter prediction models to obtain several predicted control parameters and generate desulfurization wastewater treatment instructions; an evaluation module for acquiring real-time monitoring parameters of the wastewater based on monitoring time nodes, and generating real-time operation evaluation values ​​for the corresponding modules based on the real-time monitoring parameters; and an optimization module for determining whether optimization is needed based on the real-time operation evaluation values. If so, the module selects the control parameters to be optimized and generates optimization instructions to achieve full-load zero discharge of desulfurization wastewater from the desulfurization wastewater system, meeting wastewater discharge requirements and reducing environmental pollution.

Description

Technical Field

[0001] This application relates to the field of zero discharge technology for desulfurization wastewater, and in particular to a full-load zero discharge system and method for desulfurization wastewater. Background Technology

[0002] The current zero-discharge technology for desulfurization wastewater in thermal power plants, which combines "low-temperature flue gas concentration + heavy metal harmless treatment + desulfurization wastewater bypass flue gas evaporation", is limited by the unit load. The evaporation flue gas is taken from before the air preheater and returned to after the air preheater. When the unit is under low load, the wastewater in the evaporator cannot be well atomized due to the low differential pressure of the air preheater, and it clumps, forcing it to shut down. The concentration system is also forced to shut down, resulting in severe corrosion of the flue gas duct of the concentration system and affecting the normal operation of the zero-discharge wastewater system.

[0003] Therefore, current zero-discharge technology for desulfurization wastewater has not achieved zero discharge of desulfurization wastewater at full load. Thus, there is an urgent need for a zero-discharge system and method for desulfurization wastewater at full load to meet the requirements for full-load wastewater discharge, improve system control efficiency, and reduce environmental pollution. Summary of the Invention

[0004] To address the aforementioned technical issues, this application provides a full-load desulfurization wastewater zero-discharge system and method. By constructing a control parameter prediction module under full-load conditions and combining it with real-time load values, the predicted control parameters of the current module are determined and desulfurization wastewater treatment instructions are generated. Based on real-time operation evaluation values, it is determined whether optimization is needed. If so, the control parameters to be optimized are determined and optimization instructions are generated, thereby achieving full-load desulfurization wastewater zero discharge from the desulfurization wastewater system, improving the control efficiency and accuracy of the zero-discharge system, meeting wastewater discharge requirements, and reducing environmental pollution.

[0005] In some embodiments of this application, a full-load desulfurization wastewater zero-discharge system is provided, comprising:

[0006] The data acquisition module is used to construct the first, second, and third control parameter sets at full load of the unit, and use them as training data to train the neural network and obtain the corresponding control parameter prediction model.

[0007] The generation module is used to obtain real-time load values ​​and input them into the control parameter prediction model to obtain several predicted control parameters, and generate desulfurization wastewater treatment instructions according to the predicted control parameters.

[0008] The evaluation module is used to obtain real-time monitoring parameters of wastewater based on pre-set monitoring time nodes, and generate real-time operation evaluation values ​​for the corresponding modules based on the real-time monitoring parameters.

[0009] The optimization module is used to determine whether optimization is needed based on real-time performance evaluation values. If so, it filters out the control parameters to be optimized for the current module and generates optimization instructions.

[0010] In some embodiments of this application, the first control parameter group includes several first control parameters of the low-temperature flue gas concentration module, and the several first control parameters include the flue gas temperature, flue gas flow rate, wastewater flow rate and concentrate concentration of the low-temperature flue gas concentration module.

[0011] The second control parameter group includes several second control parameters of the heavy metal treatment module, including reaction pH value, reagent addition parameters of calcium carbonate slurry, reaction temperature, stirring intensity and reaction time;

[0012] The third control parameter group includes several third control parameters of the high-temperature bypass flue gas evaporation module, including the evaporation temperature of the high-temperature bypass flue gas evaporation module, the atomized particle size of wastewater, the moisture content of flue gas, and the atomization effect.

[0013] In some embodiments of this application, before constructing the first control parameter group, the second control parameter group, and the third control parameter group at full unit load, the following steps are also included:

[0014] Obtain historical wastewater treatment logs, filter out preferred wastewater treatment logs based on historical wastewater treatment results, and use the historical treatment duration of each module in each preferred wastewater treatment log as a time reference line;

[0015] The historical data collection nodes for each module are set based on the preset time interval and the historical processing time of each module.

[0016] Historical unit loads and corresponding module historical control parameters are collected according to historical acquisition nodes, and historical unit load change curves and historical control parameter change curves are constructed for each module.

[0017] The unit's full load is pre-divided into several preset sub-load intervals;

[0018] Based on the boundary load value of each preset sub-load interval, the historical unit load change curve corresponding to each module is marked to obtain several historical unit load change curve segments. Each historical unit load change curve segment is mapped one by one to the corresponding preset sub-load interval.

[0019] The historical control parameter variation curves at the same historical acquisition node for each historical unit load variation curve segment are analyzed to obtain the historical fluctuation characteristics of each historical control parameter at several historical acquisition nodes for each historical unit load variation curve segment.

[0020] Based on historical fluctuation characteristics, the historical fluctuation evaluation value of the corresponding historical control parameter in the corresponding historical unit load change curve segment is generated. Based on the historical fluctuation evaluation value of each historical control parameter, the evaluation value of the operation impact of the preset sub-load interval mapped by the corresponding historical load change curve segment on the corresponding module is generated.

[0021] Several preset sub-load intervals are generated sequentially to evaluate the impact of each module's operation on the load.

[0022] Pre-set the first preset operational impact evaluation value range, the second preset operational impact evaluation value range, and the third preset operational impact evaluation value range;

[0023] Based on the relationship between the operational impact assessment value and the preset operational impact assessment value range, determine whether it is necessary to adjust the corresponding preset sub-load range;

[0024] When the operational impact assessment value is within the first preset operational impact assessment value range, the historical change characteristics of each historical control parameter in the corresponding preset sub-load range are compared with the historical change characteristics of the corresponding historical control parameters in the adjacent preset sub-load range to obtain the similarity.

[0025] Merge the corresponding preset sub-load intervals into the adjacent preset sub-load intervals with the highest similarity;

[0026] When the operational impact assessment value is within the second preset operational impact assessment value range, the preset sub-load range will not be adjusted;

[0027] When the operational impact assessment value falls within the third preset operational impact assessment value range, the corresponding preset sub-load range is further divided.

[0028] Based on the adjustment results, several adjustment sub-load intervals are generated for each module.

[0029] In some embodiments of this application, the operational impact evaluation value includes:

[0030] The historical fluctuation characteristics include the degree of historical fluctuation, the magnitude of historical fluctuation, and the rate of historical fluctuation at several historical data collection nodes corresponding to the historical unit load change curve segment.

[0031] By comparing historical fluctuation characteristics with preset fluctuation characteristics, the differences in historical fluctuation characteristics are obtained and quantified to obtain the first value, the second value, and the third value.

[0032] Historical fluctuation evaluation values ​​are generated based on the first, second, and third values.

[0033] The historical fluctuation evaluation values ​​of multiple historical control parameters in each preset sub-load interval of each module are compared with the preset fluctuation evaluation value threshold. Based on the comparison results, the number of historical control parameters whose historical fluctuation evaluation values ​​are greater than the preset fluctuation evaluation value threshold is selected, and the difference in historical fluctuation evaluation values ​​is calculated.

[0034] Based on the number of selected historical control parameters and the difference between the corresponding historical fluctuation evaluation values, an evaluation value for the operational impact of the preset sub-load interval on the corresponding module is generated.

[0035] The formula for calculating the operational impact evaluation value is as follows:

[0036] ;

[0037] Where Y is the operational impact evaluation value, y1 is the operational impact conversion coefficient, n1 is the number of historical control parameters selected, and n2 is the total number of historical control parameters for the corresponding module. The difference in historical fluctuation evaluation values ​​for the selected i-th historical control parameter.

[0038] In some embodiments of this application, a first set of control parameters, a second set of control parameters, and a third set of control parameters at full load of the unit are constructed and used as training data for neural network training to obtain the corresponding control parameter prediction model, including:

[0039] The operation impact ratio of the corresponding adjustment sub-load interval is generated based on the load change value and operation impact evaluation value of each adjustment sub-load interval in each module;

[0040] A first preset operating impact ratio threshold, a second preset operating impact ratio threshold, and a third preset operating impact ratio threshold are preset.

[0041] When the operating impact ratio is less than the first preset operating impact ratio threshold, the standard number of samples for the corresponding adjustment sub-load interval is set to the fourth preset number of samples.

[0042] When the operating impact ratio is between the first preset operating impact ratio threshold and the second preset operating impact ratio threshold, the standard number of samples for the corresponding adjustment sub-load interval is set to the third preset number of samples.

[0043] When the operating impact ratio is between the second preset operating impact ratio threshold and the third preset operating impact ratio threshold, the standard number of samples for the corresponding adjustment sub-load interval is set to the second preset number of samples.

[0044] When the operating impact ratio is greater than the third preset operating impact ratio threshold, the standard number of samples for the corresponding adjustment sub-load interval is set to the first preset number of samples.

[0045] According to the standard number of data acquisitions, data acquisitions and water quality characteristic acquisitions are performed on the historical control parameter change curves at several historical acquisition nodes in the historical load change curve segment mapped by the adjustment sub-load interval. This yields several first control parameters of the low-temperature flue gas concentration module, several second control parameters of the heavy metal treatment module, several third control parameters of the high-temperature bypass flue gas evaporation module, and wastewater quality characteristics.

[0046] Among them, the plurality of first control parameters, the plurality of second control parameters and the plurality of third control parameters have a mapping relationship with the corresponding adjustment sub-load range of the corresponding module and the wastewater quality characteristics;

[0047] The training input data consists of several adjustment sub-loads of each module and wastewater quality characteristics. The training output data consists of several first control parameters, several second control parameters and several third control parameters of each module with a mapping relationship between the adjustment sub-loads.

[0048] The neural network is trained based on the training input data and training output data to obtain the control parameter prediction model.

[0049] In some embodiments of this application, generating desulfurization wastewater treatment instructions according to predictive control parameters includes:

[0050] Obtain the real-time load value and real-time wastewater quality characteristics of the current module, and determine the adjustment sub-load range in which the real-time load value is located;

[0051] Based on the control parameter prediction model, predictive control parameters are used to determine the adjustment sub-load range of the current module's real-time load value and the real-time wastewater quality characteristics.

[0052] Generate desulfurization wastewater treatment instructions for the corresponding module based on the predictive control parameters of the current module, and control the current module according to the desulfurization wastewater treatment instructions.

[0053] In some embodiments of this application, generating real-time operational evaluation values ​​for corresponding modules based on real-time monitoring parameters includes:

[0054] The current module's operation and inspection period is preset, and several monitoring time nodes are set based on the preset time interval and the length of the operation and inspection period;

[0055] The real-time monitoring parameters of the current module are obtained according to the monitoring time node. The real-time monitoring parameters include the actual control parameters of the current module, the actual wastewater quality characteristics, and the actual change characteristics of the actual wastewater quality.

[0056] The actual wastewater quality characteristics are compared with the standard wastewater quality characteristics of the corresponding module to obtain the differences in actual wastewater quality characteristics. The actual changes in characteristics are compared with the standard changes in the standard wastewater quality characteristics of the corresponding module to obtain the differences in actual changes in characteristics.

[0057] Initial operational evaluation values ​​are generated based on the actual differences in wastewater quality characteristics, and compensation coefficients are generated based on the actual differences in change characteristics.

[0058] The real-time operational evaluation value of the current module is generated based on the initial operational evaluation value and the corresponding compensation coefficient.

[0059] In some embodiments of this application, the process of filtering out the control parameters to be optimized for the current module and generating optimization instructions includes:

[0060] Pre-set the threshold value for the operational evaluation value during the current module's operational testing period;

[0061] If the real-time performance evaluation value is greater than the performance evaluation value threshold, then the current module does not need to be optimized.

[0062] If the real-time evaluation value is less than the evaluation value threshold, the wastewater quality characteristics whose actual wastewater quality characteristics differ from the preset water quality characteristics threshold or whose actual changes in wastewater quality characteristics differ from the preset changes in water quality characteristics are set as water quality characteristics to be optimized, and the corresponding differences to be optimized are calculated.

[0063] Based on historical wastewater treatment logs, historical control parameters associated with the water quality characteristics to be optimized are determined, and a reference library of control parameters for the water quality characteristics to be optimized is constructed.

[0064] The control parameter reference library includes preset difference features corresponding to the water quality characteristics to be optimized, and each preset difference feature is associated with a preset adjustment value of a corresponding historical control parameter.

[0065] The control parameters to be optimized for the current module are determined based on the historical control parameters associated with the water quality characteristics to be optimized, and the adjustment values ​​of the control parameters to be optimized are determined based on the control parameter reference library.

[0066] Optimization instructions are generated based on the control parameters to be optimized and the values ​​to be adjusted, and the control parameter prediction model is iteratively trained based on the adjusted control parameters.

[0067] In some embodiments of this application, a method for zero discharge of desulfurization wastewater at full load is also included:

[0068] The first, second, and third sets of control parameters for the unit at full load are constructed and used as training data to train a neural network, thereby obtaining the corresponding control parameter prediction model.

[0069] The real-time load value is obtained and input into the control parameter prediction model to obtain several predicted control parameters. The desulfurization wastewater treatment instruction is generated according to the predicted control parameters.

[0070] Real-time monitoring parameters of wastewater are obtained based on pre-set monitoring time nodes, and real-time operation evaluation values ​​of corresponding modules are generated based on the real-time monitoring parameters.

[0071] Determine whether optimization is needed based on real-time performance evaluation values. If so, filter out the control parameters to be optimized for the current module and generate optimization instructions.

[0072] The full-load desulfurization wastewater zero-discharge system and method of this application, compared with the prior art, has the following advantages:

[0073] By constructing a control parameter prediction module under full load conditions and combining it with real-time load values, the predicted control parameters of the current module are determined and desulfurization wastewater treatment instructions are generated. Based on the real-time operation evaluation values, it is determined whether optimization is needed. If so, the control parameters to be optimized are determined and optimization instructions are generated, thereby achieving zero discharge of desulfurization wastewater under full load in the desulfurization wastewater system, improving the control efficiency and control accuracy of the zero discharge system, meeting wastewater discharge requirements, and reducing environmental pollution. Attached Figure Description

[0074] Figure 1 This is a schematic diagram of a full-load desulfurization wastewater zero-discharge system in an embodiment of this application;

[0075] Figure 2 This is a schematic flowchart of a method for zero discharge of desulfurization wastewater under full load, as described in an embodiment of this application. Detailed Implementation

[0076] The specific embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The following examples are used to illustrate this application, but are not intended to limit the scope of this application.

[0077] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0078] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0079] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0080] like Figure 1 As shown in the figure, a full-load desulfurization wastewater zero-discharge system according to an embodiment of this application includes:

[0081] The data acquisition module is used to construct the first, second, and third control parameter sets at full load of the unit, and use them as training data to train the neural network and obtain the corresponding control parameter prediction model.

[0082] The generation module is used to obtain real-time load values ​​and input them into the control parameter prediction model to obtain several predicted control parameters, and generate desulfurization wastewater treatment instructions according to the predicted control parameters.

[0083] The evaluation module is used to obtain real-time monitoring parameters of wastewater based on pre-set monitoring time nodes, and generate real-time operation evaluation values ​​for the corresponding modules based on the real-time monitoring parameters.

[0084] The optimization module is used to determine whether optimization is needed based on real-time performance evaluation values. If so, it filters out the control parameters to be optimized for the current module and generates optimization instructions.

[0085] In this embodiment, the full load of the unit includes both high load and low load under deep adjustment. This application focuses more on the system control parameters under low load. By constructing a full load control parameter prediction model for each module, the normal operation of the wastewater zero discharge system and the wastewater discharge requirements are achieved, and environmental pollution is reduced.

[0086] In some embodiments of this application, the first control parameter group includes several first control parameters of the low-temperature flue gas concentration module, and the several first control parameters include the flue gas temperature, flue gas flow rate, wastewater flow rate and concentrate concentration of the low-temperature flue gas concentration module.

[0087] The second control parameter group includes several second control parameters of the heavy metal treatment module, including reaction pH value, reagent addition parameters of calcium carbonate slurry, reaction temperature, stirring intensity and reaction time;

[0088] The third control parameter group includes several third control parameters of the high-temperature bypass flue gas evaporation module, including the evaporation temperature of the high-temperature bypass flue gas evaporation module, the atomized particle size of wastewater, the moisture content of flue gas, and the atomization effect.

[0089] In this embodiment, the low-temperature flue gas concentration module includes a pre-settler, a wastewater tank, a low-temperature flue gas concentration tower, and a booster fan. The operation process includes, but is not limited to: one path of desulfurization wastewater enters the pre-settler and then the wastewater tank, while the other path directly enters the wastewater tank through a bypass electric valve to adjust the solids content of the wastewater entering the tank to 1%-2%. The wastewater from the wastewater tank enters the low-temperature flue gas concentration tower. On the water side, it is sprayed onto the top of the concentration tower by a circulating pump. Flue gas at 90-110°C from the induced draft fan enters the concentration tower through the booster fan, evaporating and demisting a portion of the sprayed wastewater before returning it to the FGD inlet. After FGD desulfurization, it is discharged into the atmosphere through the chimney. The solids content of the concentrated wastewater is approximately 6-8%. This concentration of solids content can improve seed crystal formation, making it less likely for the wastewater to crystallize on the pipeline.

[0090] In this embodiment, the heavy metal treatment module includes a heavy metal treatment device and a mercury removal device. The operation process includes, but is not limited to: wastewater entering the heavy metal treatment device, adjusting the pH of the wastewater to 5-6 using calcium carbonate slurry; simultaneously, a mercury removal device is installed, which adjusts the pH to around 9 by adding limestone, removing most of the mercury to form precipitates, which are then filtered out by an inclined tube sedimentation tank. The wastewater after heavy metal treatment enters a spray tank and is pumped into a bypass flue gas evaporator. The concentration inlet and outlet flues are made of highly corrosion-resistant glass flakes suitable for humid and strongly acidic environments to ensure that the flues are not corroded.

[0091] In this embodiment, the high-temperature bypass flue gas evaporation module includes an air preheater, a high-temperature bypass flue gas evaporator, a spray water tank, etc. The operation process includes, but is not limited to: extracting 5-10% of the high-temperature flue gas with a temperature of up to 300-350℃ from the air preheater inlet to the high-temperature bypass flue gas evaporator; the wastewater from the spray water tank is vaporized by the high-temperature flue gas after being atomized by high-speed rotating spray; and the flue gas is returned to the electrostatic precipitator. Since the flue gas differential pressure of the bypass flue gas evaporator is at least 0.6 kPa, even if the unit load is 30% of the rated load deep peak load, there is enough power and flue gas volume to achieve full-load evaporation.

[0092] In some embodiments of this application, before constructing the first control parameter group, the second control parameter group, and the third control parameter group at full unit load, the following steps are also included:

[0093] Obtain historical wastewater treatment logs, filter out preferred wastewater treatment logs based on historical wastewater treatment results, and use the historical treatment duration of each module in each preferred wastewater treatment log as a time reference line;

[0094] The historical data collection nodes for each module are set based on the preset time interval and the historical processing time of each module.

[0095] Historical unit loads and corresponding module historical control parameters are collected according to historical acquisition nodes, and historical unit load change curves and historical control parameter change curves are constructed for each module.

[0096] The unit's full load is pre-divided into several preset sub-load intervals;

[0097] Based on the boundary load value of each preset sub-load interval, the historical unit load change curve corresponding to each module is marked to obtain several historical unit load change curve segments. Each historical unit load change curve segment is mapped one by one to the corresponding preset sub-load interval.

[0098] The historical control parameter variation curves at the same historical acquisition node for each historical unit load variation curve segment are analyzed to obtain the historical fluctuation characteristics of each historical control parameter at several historical acquisition nodes for each historical unit load variation curve segment.

[0099] Based on historical fluctuation characteristics, the historical fluctuation evaluation value of the corresponding historical control parameter in the corresponding historical unit load change curve segment is generated. Based on the historical fluctuation evaluation value of each historical control parameter, the evaluation value of the operation impact of the preset sub-load interval mapped by the corresponding historical load change curve segment on the corresponding module is generated.

[0100] Several preset sub-load intervals are generated sequentially to evaluate the impact of each module's operation on the load.

[0101] Pre-set the first preset operational impact evaluation value range, the second preset operational impact evaluation value range, and the third preset operational impact evaluation value range;

[0102] Based on the relationship between the operational impact assessment value and the preset operational impact assessment value range, determine whether it is necessary to adjust the corresponding preset sub-load range;

[0103] When the operational impact assessment value is within the first preset operational impact assessment value range, the historical change characteristics of each historical control parameter in the corresponding preset sub-load range are compared with the historical change characteristics of the corresponding historical control parameters in the adjacent preset sub-load range to obtain the similarity.

[0104] Merge the corresponding preset sub-load intervals into the adjacent preset sub-load intervals with the highest similarity;

[0105] When the operational impact assessment value is within the second preset operational impact assessment value range, the preset sub-load range will not be adjusted;

[0106] When the operational impact assessment value falls within the third preset operational impact assessment value range, the corresponding preset sub-load range is further divided.

[0107] Based on the adjustment results, several adjustment sub-load intervals are generated for each module.

[0108] In this embodiment, the preferred wastewater treatment log refers to a historical wastewater treatment log with high wastewater treatment efficiency and wastewater discharge meeting emission standards.

[0109] In this embodiment, the first preset operating impact evaluation value range < the second preset operating impact evaluation value range < the third preset operating impact evaluation value range. The operating impact evaluation value refers to the degree of influence of the corresponding preset sub-load range on the control parameters of the corresponding module. When the operating impact evaluation value is larger, it means that the degree of influence is greater, and the corresponding preset sub-load range should be further divided. The re-division node is set according to the actual situation. Conversely, it means that the degree of influence is smaller, and the corresponding preset sub-load range is merged. This achieves the accuracy of the division of the adjustment sub-load range for different modules, lays the foundation for the subsequent construction of the control parameter prediction module, and improves the control efficiency and control accuracy of the desulfurization wastewater zero discharge system.

[0110] In some embodiments of this application, the operational impact evaluation value includes:

[0111] The historical fluctuation characteristics include the degree of historical fluctuation, the magnitude of historical fluctuation, and the rate of historical fluctuation at several historical data collection nodes corresponding to the historical unit load change curve segment.

[0112] By comparing historical fluctuation characteristics with preset fluctuation characteristics, the differences in historical fluctuation characteristics are obtained and quantified to obtain the first value, the second value, and the third value.

[0113] Historical fluctuation evaluation values ​​are generated based on the first, second, and third values.

[0114] The historical fluctuation evaluation values ​​of multiple historical control parameters in each preset sub-load interval of each module are compared with the preset fluctuation evaluation value threshold. Based on the comparison results, the number of historical control parameters whose historical fluctuation evaluation values ​​are greater than the preset fluctuation evaluation value threshold is selected, and the difference in historical fluctuation evaluation values ​​is calculated.

[0115] Based on the number of selected historical control parameters and the difference between the corresponding historical fluctuation evaluation values, an evaluation value for the operational impact of the preset sub-load interval on the corresponding module is generated.

[0116] The formula for calculating the operational impact evaluation value is as follows:

[0117] ;

[0118] Where Y is the operational impact evaluation value, y1 is the operational impact conversion coefficient, n1 is the number of historical control parameters selected, and n2 is the total number of historical control parameters for the corresponding module. The difference in historical fluctuation evaluation values ​​for the selected i-th historical control parameter.

[0119] In this embodiment, the first value, the second value, and the third value refer to the differences in degree, value, and volatility between the historical volatility degree, historical volatility value, and historical volatility rate in the historical volatility characteristics and the standard volatility degree, standard volatility value, and standard volatility rate in the standard volatility characteristics, respectively. When the degree difference, value difference, and volatility rate difference are greater, the corresponding first value, second value, and third value are also greater, that is, the historical volatility evaluation value of the corresponding historical control parameter is greater, and vice versa.

[0120] In this embodiment, the influence of the corresponding preset sub-load interval on the control parameters in the corresponding module is generated based on the historical fluctuation evaluation value of the historical control parameters in each preset sub-load interval. This lays the foundation for determining the number of standard data collections and building a control parameter prediction model, thereby improving the control accuracy and efficiency of the desulfurization wastewater zero discharge system under full load conditions, meeting wastewater discharge standards, and reducing environmental pollution.

[0121] In some embodiments of this application, a first set of control parameters, a second set of control parameters, and a third set of control parameters at full load of the unit are constructed and used as training data for neural network training to obtain the corresponding control parameter prediction model, including:

[0122] The operation impact ratio of the corresponding adjustment sub-load interval is generated based on the load change value and operation impact evaluation value of each adjustment sub-load interval in each module;

[0123] A first preset operating impact ratio threshold, a second preset operating impact ratio threshold, and a third preset operating impact ratio threshold are preset.

[0124] When the operating impact ratio is less than the first preset operating impact ratio threshold, the standard number of samples for the corresponding adjustment sub-load interval is set to the fourth preset number of samples.

[0125] When the operating impact ratio is between the first preset operating impact ratio threshold and the second preset operating impact ratio threshold, the standard number of samples for the corresponding adjustment sub-load interval is set to the third preset number of samples.

[0126] When the operating impact ratio is between the second preset operating impact ratio threshold and the third preset operating impact ratio threshold, the standard number of samples for the corresponding adjustment sub-load interval is set to the second preset number of samples.

[0127] When the operating impact ratio is greater than the third preset operating impact ratio threshold, the standard number of samples for the corresponding adjustment sub-load interval is set to the first preset number of samples.

[0128] According to the standard number of data acquisitions, data acquisitions and water quality characteristic acquisitions are performed on the historical control parameter change curves at several historical acquisition nodes in the historical load change curve segment mapped by the adjustment sub-load interval. This yields several first control parameters of the low-temperature flue gas concentration module, several second control parameters of the heavy metal treatment module, several third control parameters of the high-temperature bypass flue gas evaporation module, and wastewater quality characteristics.

[0129] Among them, the plurality of first control parameters, the plurality of second control parameters and the plurality of third control parameters have a mapping relationship with the corresponding adjustment sub-load range of the corresponding module and the wastewater quality characteristics;

[0130] The training input data consists of several adjustment sub-loads of each module and wastewater quality characteristics. The training output data consists of several first control parameters, several second control parameters and several third control parameters of each module with a mapping relationship between the adjustment sub-loads.

[0131] The neural network is trained based on the training input data and training output data to obtain the control parameter prediction model.

[0132] In this embodiment, the operating impact ratio = load change value / operating impact evaluation value. The smaller the operating impact ratio, the smaller the load change value or the larger the operating impact evaluation value. That is, a small load change causes a large change in the control parameters. In this case, the number of standard data collections for the corresponding adjustment sub-load interval is more, and vice versa.

[0133] In this embodiment, the first preset operating influence ratio threshold is less than the second preset operating influence ratio threshold and the third preset operating influence ratio threshold, and the first preset number of samples is less than the second preset number of samples, the third preset number of samples, and the fourth preset number of samples.

[0134] In this embodiment, the wastewater quality characteristics include heavy metal content, suspended solids concentration, and salt content.

[0135] In this embodiment, by determining the standard number of data collections for each adjustment sub-load interval, and determining several first control parameters, second control parameters, and third control parameters, and constructing a control parameter prediction model, accurate prediction of control parameters for the desulfurization wastewater zero discharge system under full load conditions is achieved, thereby improving control efficiency, meeting wastewater discharge requirements, and reducing environmental pollution.

[0136] In some embodiments of this application, generating desulfurization wastewater treatment instructions according to predictive control parameters includes:

[0137] Obtain the real-time load value and real-time wastewater quality characteristics of the current module, and determine the adjustment sub-load range in which the real-time load value is located;

[0138] Based on the control parameter prediction model, predictive control parameters are used to determine the adjustment sub-load range of the current module's real-time load value and the real-time wastewater quality characteristics.

[0139] Generate desulfurization wastewater treatment instructions for the corresponding module based on the predictive control parameters of the current module, and control the current module according to the desulfurization wastewater treatment instructions.

[0140] In some embodiments of this application, generating real-time operational evaluation values ​​for corresponding modules based on real-time monitoring parameters includes:

[0141] The current module's operation and inspection period is preset, and several monitoring time nodes are set based on the preset time interval and the length of the operation and inspection period;

[0142] The real-time monitoring parameters of the current module are obtained according to the monitoring time node. The real-time monitoring parameters include the actual control parameters of the current module, the actual wastewater quality characteristics, and the actual change characteristics of the actual wastewater quality.

[0143] The actual wastewater quality characteristics are compared with the standard wastewater quality characteristics of the corresponding module to obtain the differences in actual wastewater quality characteristics. The actual changes in characteristics are compared with the standard changes in the standard wastewater quality characteristics of the corresponding module to obtain the differences in actual changes in characteristics.

[0144] Initial operational evaluation values ​​are generated based on the actual differences in wastewater quality characteristics, and compensation coefficients are generated based on the actual differences in change characteristics.

[0145] The real-time operational evaluation value of the current module is generated based on the initial operational evaluation value and the corresponding compensation coefficient.

[0146] In this embodiment, the operation and verification period is set according to the historical average processing time of the adjustment sub-load interval in which the real-time load value of the current module is located. The operation and verification period is half of the historical average processing time and is used to verify the application effect of the predictive control parameters of the current module.

[0147] In this embodiment, the actual change characteristics include the actual change trend, the actual change value, and the actual change rate, which are used to evaluate whether each wastewater quality characteristic meets the standard change characteristics of the current module's wastewater discharge standard. When the difference in actual change characteristics is smaller, the corresponding compensation coefficient is larger, and vice versa. The range of the compensation coefficient is (0.8, 1.2).

[0148] In this embodiment, the greater the difference in actual wastewater quality characteristics, the smaller the corresponding initial operational evaluation value. When the initial operational evaluation value is smaller and the compensation coefficient is smaller, the corresponding real-time operational evaluation value is smaller, and vice versa.

[0149] In some embodiments of this application, the process of filtering out the control parameters to be optimized for the current module and generating optimization instructions includes:

[0150] Pre-set the threshold value for the operational evaluation value during the current module's operational testing period;

[0151] If the real-time performance evaluation value is greater than the performance evaluation value threshold, then the current module does not need to be optimized.

[0152] If the real-time evaluation value is less than the evaluation value threshold, the wastewater quality characteristics whose actual wastewater quality characteristics differ from the preset water quality characteristics threshold or whose actual changes in wastewater quality characteristics differ from the preset changes in water quality characteristics are set as water quality characteristics to be optimized, and the corresponding differences to be optimized are calculated.

[0153] Based on historical wastewater treatment logs, historical control parameters associated with the water quality characteristics to be optimized are determined, and a reference library of control parameters for the water quality characteristics to be optimized is constructed.

[0154] The control parameter reference library includes preset difference features corresponding to the water quality characteristics to be optimized, and each preset difference feature is associated with a preset adjustment value of a corresponding historical control parameter.

[0155] The control parameters to be optimized for the current module are determined based on the historical control parameters associated with the water quality characteristics to be optimized, and the adjustment values ​​of the control parameters to be optimized are determined based on the control parameter reference library.

[0156] Optimization instructions are generated based on the control parameters to be optimized and the values ​​to be adjusted, and the control parameter prediction model is iteratively trained based on the adjusted control parameters.

[0157] In this embodiment, the difference feature to be optimized and the control parameter to be optimized are input into the control parameter reference library corresponding to the water quality feature to be optimized, and the similarity between the difference feature to be optimized and the preset difference feature is obtained. The preset adjustment value of the corresponding control parameter to be optimized under the preset difference feature with the highest similarity is set as the adjustment value.

[0158] In some embodiments of this application, such as Figure 2 As shown, it also includes a method for zero discharge of desulfurization wastewater at full load:

[0159] Step S201: Construct the first control parameter group, the second control parameter group, and the third control parameter group at full load of the unit, and use them as training data to train the neural network to obtain the corresponding control parameter prediction model;

[0160] Step S202: Obtain the real-time load value and input it into the control parameter prediction model to obtain several predicted control parameters, and generate desulfurization wastewater treatment instructions according to the predicted control parameters;

[0161] Step S203: Obtain real-time monitoring parameters of wastewater based on pre-set monitoring time nodes, and generate real-time operation evaluation values ​​of the corresponding modules based on the real-time monitoring parameters;

[0162] Step S204: Determine whether optimization is needed based on the real-time running evaluation value. If so, filter out the control parameters to be optimized for the current module and generate optimization instructions.

[0163] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and substitutions can be made without departing from the technical principles of this application, and these improvements and substitutions should also be considered within the scope of protection of this application.

Claims

1. A full load desulfurization wastewater zero discharge method, characterized in that, include: The first, second, and third sets of control parameters for the unit at full load are constructed and used as training data to train a neural network, thereby obtaining the corresponding control parameter prediction model. The real-time load value is obtained and input into the control parameter prediction model to obtain several predicted control parameters. The desulfurization wastewater treatment instruction is generated according to the predicted control parameters. Real-time monitoring parameters of wastewater are obtained based on pre-set monitoring time nodes, and real-time operation evaluation values ​​of corresponding modules are generated based on the real-time monitoring parameters. Determine whether optimization is needed based on real-time performance evaluation values. If so, filter out the control parameters to be optimized for the current module and generate optimization instructions. Before constructing the first, second, and third control parameter sets for the unit at full load, the following steps are also included: Obtain historical wastewater treatment logs, filter out preferred wastewater treatment logs based on historical wastewater treatment results, and use the historical treatment duration of each module in each preferred wastewater treatment log as a time reference line; The historical data collection nodes for each module are set based on the preset time interval and the historical processing time of each module. Historical unit loads and corresponding module historical control parameters are collected according to historical acquisition nodes, and historical unit load change curves and historical control parameter change curves are constructed for each module. The unit's full load is pre-divided into several preset sub-load intervals; Based on the boundary load value of each preset sub-load interval, the historical unit load change curve corresponding to each module is marked to obtain several historical unit load change curve segments. Each historical unit load change curve segment is mapped one by one to the corresponding preset sub-load interval. The historical control parameter variation curves at the same historical acquisition node for each historical unit load variation curve segment are analyzed to obtain the historical fluctuation characteristics of each historical control parameter at several historical acquisition nodes for each historical unit load variation curve segment. Based on the historical fluctuation characteristics, the historical fluctuation evaluation value of the corresponding historical control parameter in the corresponding historical unit load change curve segment is generated. Based on the historical fluctuation evaluation value of each historical control parameter, the evaluation value of the operation impact of the preset sub-load interval mapped by the corresponding historical load change curve segment on the corresponding module is generated. Several preset sub-load intervals are generated sequentially to evaluate the impact of each module's operation on the load. Pre-set the first preset operational impact evaluation value range, the second preset operational impact evaluation value range, and the third preset operational impact evaluation value range; Based on the relationship between the operational impact assessment value and the preset operational impact assessment value range, determine whether it is necessary to adjust the corresponding preset sub-load range; When the operational impact assessment value is within the first preset operational impact assessment value range, the historical change characteristics of each historical control parameter in the corresponding preset sub-load range are compared with the historical change characteristics of the corresponding historical control parameters in the adjacent preset sub-load range to obtain the similarity. Merge the corresponding preset sub-load intervals into the adjacent preset sub-load intervals with the highest similarity; When the operational impact assessment value is within the second preset operational impact assessment value range, the preset sub-load range will not be adjusted; When the operational impact assessment value falls within the third preset operational impact assessment value range, the corresponding preset sub-load range is further divided. Based on the adjustment results, several adjustment sub-load intervals are generated for each module; The first control parameter group includes several first control parameters of the low-temperature flue gas concentration module, including the flue gas temperature, flue gas flow rate, wastewater flow rate, and concentrate concentration of the low-temperature flue gas concentration module. The second control parameter group includes several second control parameters of the heavy metal treatment module, including reaction pH value, reagent addition parameters of calcium carbonate slurry, reaction temperature, stirring intensity and reaction time; The third control parameter group includes several third control parameters of the high-temperature bypass flue gas evaporation module, including the evaporation temperature of the high-temperature bypass flue gas evaporation module, the atomized particle size of wastewater, the moisture content of flue gas, and the atomization effect.

2. The full load desulfurized wastewater zero emission method according to claim 1, characterized by, The operational impact evaluation values ​​include: The historical fluctuation characteristics include the degree of historical fluctuation, the magnitude of historical fluctuation, and the rate of historical fluctuation at several historical data collection nodes corresponding to the historical unit load change curve segment. By comparing historical fluctuation characteristics with preset fluctuation characteristics, the differences in historical fluctuation characteristics are obtained and quantified to obtain the first value, the second value, and the third value. Historical fluctuation evaluation values ​​are generated based on the first, second, and third values. The historical fluctuation evaluation values ​​of multiple historical control parameters in each preset sub-load interval of each module are compared with the preset fluctuation evaluation value threshold. Based on the comparison results, the number of historical control parameters whose historical fluctuation evaluation values ​​are greater than the preset fluctuation evaluation value threshold is selected, and the difference in historical fluctuation evaluation values ​​is calculated. Based on the number of selected historical control parameters and the difference between the corresponding historical fluctuation evaluation values, an evaluation value for the operational impact of the preset sub-load interval on the corresponding module is generated. The formula for calculating the operational impact evaluation value is as follows: ； Where Y is the operational impact evaluation value, y1 is the operational impact conversion coefficient, n1 is the number of historical control parameters selected, and n2 is the total number of historical control parameters for the corresponding module. The difference in historical fluctuation evaluation values ​​for the selected i-th historical control parameter.

3. The full load desulfurized wastewater zero emission method according to claim 2, characterized by, Construct the first, second, and third sets of control parameters for the unit at full load, and use them as training data to train a neural network, obtaining the corresponding control parameter prediction model, including: The operation impact ratio of the corresponding adjustment sub-load interval is generated based on the load change value and operation impact evaluation value of each adjustment sub-load interval in each module; A first preset operating impact ratio threshold, a second preset operating impact ratio threshold, and a third preset operating impact ratio threshold are preset. When the operating impact ratio is less than the first preset operating impact ratio threshold, the standard number of samples for the corresponding adjustment sub-load interval is set to the fourth preset number of samples. When the operating impact ratio is between the first preset operating impact ratio threshold and the second preset operating impact ratio threshold, the standard number of samples for the corresponding adjustment sub-load interval is set to the third preset number of samples. When the operating impact ratio is between the second preset operating impact ratio threshold and the third preset operating impact ratio threshold, the standard number of samples for the corresponding adjustment sub-load interval is set to the second preset number of samples. When the operating impact ratio is greater than the third preset operating impact ratio threshold, the standard number of samples for the corresponding adjustment sub-load interval is set to the first preset number of samples. According to the standard number of data acquisitions, data acquisitions and water quality characteristic acquisitions are performed on the historical control parameter change curves at several historical acquisition nodes in the historical load change curve segment mapped by the adjustment sub-load interval. This yields several first control parameters of the low-temperature flue gas concentration module, several second control parameters of the heavy metal treatment module, several third control parameters of the high-temperature bypass flue gas evaporation module, and wastewater quality characteristics. Among them, the plurality of first control parameters, the plurality of second control parameters and the plurality of third control parameters have a mapping relationship with the corresponding adjustment sub-load range of the corresponding module and the wastewater quality characteristics; The training input data consists of several adjustment sub-loads of each module and wastewater quality characteristics. The training output data consists of several first control parameters, several second control parameters and several third control parameters of each module with a mapping relationship between the adjustment sub-loads. The neural network is trained based on the training input data and training output data to obtain the control parameter prediction model.

4. The full load desulfurized wastewater zero emission method according to claim 3, characterized by, Generate desulfurization wastewater treatment instructions based on predicted control parameters, including: Obtain the real-time load value and real-time wastewater quality characteristics of the current module, and determine the adjustment sub-load range in which the real-time load value is located; Based on the control parameter prediction model, predictive control parameters are used to determine the adjustment sub-load range of the current module's real-time load value and the real-time wastewater quality characteristics. Generate desulfurization wastewater treatment instructions for the corresponding module based on the predictive control parameters of the current module, and control the current module according to the desulfurization wastewater treatment instructions.

5. The method for zero discharge of desulfurization wastewater under full load as described in claim 4, characterized in that, Real-time operational evaluation values ​​for the corresponding modules are generated based on real-time monitoring parameters, including: The current module's operation and inspection period is preset, and several monitoring time nodes are set based on the preset time interval and the length of the operation and inspection period; The real-time monitoring parameters of the current module are obtained according to the monitoring time node. The real-time monitoring parameters include the actual control parameters of the current module, the actual wastewater quality characteristics, and the actual change characteristics of the actual wastewater quality. The actual wastewater quality characteristics are compared with the standard wastewater quality characteristics of the corresponding module to obtain the differences in actual wastewater quality characteristics. The actual changes in characteristics are compared with the standard changes in the standard wastewater quality characteristics of the corresponding module to obtain the differences in actual changes in characteristics. Initial operational evaluation values ​​are generated based on the actual differences in wastewater quality characteristics, and compensation coefficients are generated based on the actual differences in change characteristics. The real-time operational evaluation value of the current module is generated based on the initial operational evaluation value and the corresponding compensation coefficient.

6. The full load desulfurized wastewater zero emission method according to claim 5, characterized by, Filter out the control parameters to be optimized for the current module and generate optimization instructions, including: Pre-set the threshold value for the operational evaluation value during the current module's operational testing period; If the real-time performance evaluation value is greater than the performance evaluation value threshold, then the current module does not need to be optimized. If the real-time evaluation value is less than the evaluation value threshold, the wastewater quality characteristics whose actual wastewater quality characteristics differ from the preset water quality characteristics threshold or whose actual changes in wastewater quality characteristics differ from the preset changes in water quality characteristics are set as water quality characteristics to be optimized, and the corresponding differences to be optimized are calculated. Based on historical wastewater treatment logs, historical control parameters associated with the water quality characteristics to be optimized are determined, and a reference library of control parameters for the water quality characteristics to be optimized is constructed. The control parameter reference library includes preset difference features corresponding to the water quality characteristics to be optimized, and each preset difference feature is associated with a preset adjustment value of a corresponding historical control parameter. The control parameters to be optimized for the current module are determined based on the historical control parameters associated with the water quality characteristics to be optimized, and the adjustment values ​​of the control parameters to be optimized are determined based on the control parameter reference library. Optimization instructions are generated based on the control parameters to be optimized and the values ​​to be adjusted, and the control parameter prediction model is iteratively trained based on the adjusted control parameters.

Images