Method for treating desulfurization wastewater by adding chemicals, electronic equipment, medium and product

Abstract

The invention provides a method for treating desulfurization wastewater by adding chemicals, electronic equipment, a medium and a product, and the method comprises the following steps: collecting water quality data of the desulfurization wastewater, and preprocessing the collected water quality data; performing dosage prediction on the water quality data by adopting a neural network model to obtain a first predicted dosage; predicting the dosage by adopting a fuzzy control model to obtain a second predicted dosage; calculating the optimal dosage according to the first predicted dosage and the second predicted dosage; and adding the optimal dosage into the desulfurization wastewater, collecting water quality data after dosing, and optimizing the neural network model and the fuzzy control model according to the water quality data after dosing. By monitoring multi-dimensional water quality data in real time and combining with an intelligent algorithm to dynamically predict the dosage, the problems of insufficient dosing precision, unstable treatment effect, serious medicament waste, low intelligent degree, poor real-time performance and the like are solved, and the efficiency, the stability and the economical efficiency of desulfurization wastewater treatment are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial wastewater treatment, and in particular to a method, electronic equipment, medium and product for treating desulfurization wastewater by adding chemicals. Background Art

[0002] With increasingly stringent environmental standards, the treatment of desulfurization wastewater has become a critical issue in the industrial sector. Traditional desulfurization wastewater treatment often relies on manual judgment of dosage, resulting in low dosing accuracy, unstable treatment results, and significant reagent waste. While some automated dosing systems exist, these systems lack intelligence and struggle to adjust dosage in real time based on water quality fluctuations, leading to suboptimal treatment results. Summary of the Invention

[0003] To address these challenges, the present invention proposes a method, electronic equipment, medium, and product for treating desulfurization wastewater through dosing. By monitoring multi-dimensional water quality data in real time and combining it with an intelligent algorithm to dynamically predict dosing dosage, the present invention addresses issues such as insufficient dosing accuracy, unstable treatment results, significant reagent waste, low intelligence, and poor real-time performance. This significantly improves the efficiency, stability, and economics of desulfurization wastewater treatment while reducing environmental impact.

[0004] The present invention provides a method for treating desulfurization wastewater by adding chemicals, comprising:

[0005] Collect water quality data of desulfurization wastewater and pre-process the collected water quality data;

[0006] The neural network model is used to predict the dosage of water quality data to obtain the first predicted dosage;

[0007] The dosage is predicted by using a fuzzy control model to obtain a second predicted dosage;

[0008] Calculating an optimal dosage according to the first predicted dosage and the second predicted dosage;

[0009] Add the optimal dosage to the desulfurization wastewater, collect the water quality data after the addition, and optimize the neural network model and the fuzzy control model based on the water quality data after the addition.

[0010] In addition, the pre-processing of the collected water quality data includes:

[0011] The water quality data is cleaned and standardized. The standardization process includes: outlier processing, null value filling, outlier removal and noise interference removal.

[0012] In addition, the outlier processing adopts an algorithm based on double thresholds to adjust quartile outliers to calculate the discrete range of water quality data.

[0013] In addition, the method of using a neural network model to predict the dosage of water quality data to obtain a first predicted dosage includes:

[0014] The adopted neural network includes: an input layer, a feature selection layer, a hidden layer and an output layer. The feature selection layer is set before the hidden layer, and the output of the output layer is the first predicted dosage.

[0015] In addition, the neural network input layer X=[x1,x2,…,x m ], where x i represents the i-th water quality data;

[0016] The feature selection layer uses the gate vector to realize feature selection. Let the gate vector be G = [g1, g2, ..., g m ], where m represents the number of water quality data, and the neural network output is expressed as:

[0017]

[0018] Among them, W l is the weight matrix of the output layer, b l is the bias term of the output layer, Z l-1 is the output of the previous layer, Z l is the output layer output, σ is the activation function;

[0019] Using the mean square error formula to calculate the variance And adjust the neural network based on the variance:

[0020]

[0021] in, is the output of the last layer of the neural network, That is, the first predicted dosage, Y is the standard data matrix, ||W l || 2,1 is the constraint matrix.

[0022] In addition, the method of using the fuzzy control model to predict the dosage to obtain the second predicted dosage includes:

[0023] Let X′ i =F fuzzy (x i ) means that the water quality data x i Convert to linguistic variable X′ in fuzzy set i , where F fuzzy (x i ) represents the fuzzy processing function;

[0024] Assume the fuzzy rule base is: Among them, Rule jrepresents the jth fuzzy rule, and B j Represent the linguistic variables of the input and output fuzzy sets respectively. According to the fuzzy rules, we can get:

[0025]

[0026] Among them, ∪ represents the fuzzy union operation, and X′ represents the fuzzified input variable;

[0027] Output the fuzzy dosage Y fuzzy Convert to a specific dosage value Y′, where Y′ is the second predicted dosage:

[0028] Y′=F defuzzy (Y fuzzy )

[0029] Among them, F defuzzy Represents the defuzzification processing function, which converts the fuzzy dosage output into a specific dosage value.

[0030] Furthermore, the calculating of the optimal dosage according to the first predicted dosage and the second predicted dosage includes:

[0031] Final dosage Y final :

[0032]

[0033] Among them, α is the weighting coefficient, is the first predicted dosage, and Y′ is the second predicted dosage.

[0034] The present invention further provides an electronic device, comprising:

[0035] at least one processor; and,

[0036] a memory communicatively connected to at least one of the processors; wherein,

[0037] The memory stores instructions that can be executed by at least one of the processors, and the instructions are executed by at least one of the processors to enable at least one of the processors to execute the method for treating desulfurization wastewater by adding chemicals as described in any one of the above items.

[0038] The present invention also proposes a storage medium, which stores computer instructions. When a computer executes the computer instructions, it is used to execute the method for treating desulfurization wastewater by adding chemicals as described in any of the above items.

[0039] The present invention also proposes a computer program product, including a computer program / instruction, characterized in that when the computer program / instruction is executed by a processor, the method for treating desulfurization wastewater by adding chemicals as described in any of the above items is implemented.

[0040] The present invention solves problems such as insufficient dosing accuracy, unstable treatment effect, serious waste of reagents, low intelligence level and poor real-time performance by real-time monitoring of multi-dimensional water quality data and combining it with intelligent algorithms to dynamically predict the dosage. It significantly improves the efficiency, stability and economy of desulfurization wastewater treatment while reducing the impact on the environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A flow chart of a method for treating desulfurization wastewater by adding chemicals provided in one embodiment of the present invention;

[0042] Figure 2 A neural network model provided by one embodiment of the present invention;

[0043] Figure 3 A schematic diagram of a method for treating desulfurization wastewater by adding chemicals according to an embodiment of the present invention;

[0044] Figure 4 A schematic diagram of the hardware structure of an electronic device provided in one embodiment of the present invention;

[0045] Figure 5 It is a neural network model in the prior art. DETAILED DESCRIPTION

[0046] The present invention is further described in detail below with reference to specific embodiments and accompanying drawings, which are intended only to elaborate on the specific embodiments of the present invention and do not impose any limitation on the present invention. The scope of protection of the present invention shall be subject to the claims.

[0047] Reference Figure 1 The present invention proposes a method for treating desulfurization wastewater by adding chemicals, comprising:

[0048] Step S001, collecting water quality data of desulfurization wastewater and preprocessing the collected water quality data;

[0049] Step S002, using a neural network model to predict the dosage of water quality data to obtain a first predicted dosage;

[0050] Step S003, using a fuzzy control model to predict the dosage to obtain a second predicted dosage;

[0051] Step S004, calculating the optimal dosage according to the first predicted dosage and the second predicted dosage;

[0052] Step S005: adding an optimal dosage of the drug to the desulfurization wastewater, collecting water quality data after the drug addition, and optimizing the neural network model and the fuzzy control model based on the water quality data after the drug addition.

[0053] The technical problems existing in the technical solutions for treating desulfurization wastewater in the prior art include:

[0054] 1. Traditional treatment method based on fixed dosage:

[0055] Fixed dosages are set based on manual experience or historical data, and dosages do not change with water quality fluctuations. For example, simple control logic is used, such as timed dosing or dosing based on water volume ratios. This approach cannot adapt to water quality fluctuations, resulting in over- or underdosing. The lack of real-time water quality data can delay dosage adjustments and lead to low dosing accuracy.

[0056] 2. Automatic dosing system based on simple feedback control:

[0057] A single sensor (such as a pH sensor or turbidity sensor) is used to monitor water quality and adjust the dosage according to a preset threshold. This method has a relatively simple control logic and usually adopts a PID (proportional-integral-differential) control algorithm. However, a simple feedback control system can only adjust the dosage according to a single parameter (such as pH value), which cannot fully consider the complexity of water quality. The correlation between multiple parameters is not fully utilized, and the control strategy is too simple, resulting in unstable treatment effects.

[0058] 3. Semi-automatic dosing system based on multi-parameter monitoring:

[0059] Multiple sensors are used to monitor water quality parameters (such as pH, COD, and SS), but dosing adjustments still rely on manual intervention or simple rule-based control. This approach lacks intelligent algorithm support and cannot achieve precise dynamic adjustment. Fixed dosing or simple feedback control lacks precise dosing prediction and dynamic adjustment mechanisms, which can easily lead to overuse of chemicals and increase treatment costs.

[0060] 4. Dosing system based on offline water quality analysis:

[0061] Regular water samples are collected and sent to the laboratory for analysis, and the dosage is adjusted based on the results. However, due to the long analysis cycle, the system cannot respond to water quality changes in real time. Existing systems rely heavily on manual experience or simple rules, which cannot achieve intelligent decision-making. Offline water quality analysis methods have a long data collection and analysis cycle and cannot respond to water quality changes in real time, resulting in delayed treatment effects and failing to meet real-time processing requirements.

[0062] The method proposed by the present invention is applicable to the treatment of desulfurization wastewater in industrial fields such as thermal power plants and chemical plants. Figure 3 A schematic diagram of the overall processing.

[0063] In step S001, water quality data of the desulfurized wastewater is collected, and the collected water quality data is preprocessed.

[0064] Water quality data collection stage:

[0065] Deploy a high-precision sensor array in the wastewater treatment system, including a pH sensor, a suspended solid monitor, a heavy metal ion detector, etc. When the wastewater passes through the sensors, key water quality indicators are captured in real time: water flow rate, pH value, COD concentration, suspended solid concentration (SS), heavy metal content (copper, cadmium, silver, lead, chromium), etc.

[0066] Data preprocessing stage:

[0067] The original water quality data is strictly cleaned and standardized. Optionally, it includes outlier processing, null value filling, etc., to eliminate outliers and noise interference, providing a reliable basis for subsequent algorithm analysis.

[0068] Taking outliers as an example, they are usually defined as observed values that are significantly different from most points in the dataset. To quickly identify the data distribution of each indicator for further analysis of the water quality of the wastewater. For example, some indicators may show a large numerical range and discreteness, while others may be relatively concentrated. Optionally, a double-threshold-adjusted interquartile range outlier judgment technique is adopted. Let Q1 be the first quartile, Q3 be the third quartile, and IQR be the interquartile range, i.e., IQR = Q3 - Q1. Then the outlier can be defined as outlier = {x|x < Q1 - 1.5×IQR - t1 or x > Q3 + 1.5×IQR + t2}

[0069] Among them, t1 and t2 are the user-defined lower and upper thresholds respectively, which are mainly used to flexibly control the outlier range in different scenarios. [[ID=2)0]]

[0070] [[ID=)22]]In step S002, a neural network model is used to predict the chemical dosage of the water quality data to obtain the first predicted chemical dosage.

[0071] A neural network model is used to predict the chemical dosage of the water quality data to obtain the first predicted chemical dosage.

[0072] In step S003, a fuzzy control model is used to predict the chemical dosage to obtain the second predicted chemical dosage.

[0073] In step S004, the optimal chemical dosage is calculated based on the first predicted chemical dosage and the second predicted chemical dosage.

[0074] According to the characteristics of the water quality data and the requirements of the application scenario, two algorithms, namely the comprehensive neural network and the fuzzy control model, are designed, which can effectively balance complex, non-linear data and scenario uncertainty problems.

[0075] For highly complex and non-linear data, the deep neural network of the intelligent algorithm in the present invention can achieve accurate prediction. If the water quality data shows ambiguity and uncertainty, the prior expert knowledge base and fuzzy control algorithm can be used in the intelligent algorithm of the present invention to improve the generalization performance of the system.

[0076] Step S005: Add the optimal dosage to the desulfurization wastewater, collect the water quality data after dosing, and optimize the neural network model and fuzzy control model according to the water quality data after dosing.

[0077] Collect the water quality data after dosing, and optimize the neural network model and fuzzy control model according to the water quality data after dosing, so that the prediction results of the two models are closer to the actual measured values.

[0078] By real-time monitoring of multi-dimensional water quality data and combining intelligent algorithms to dynamically predict the dosage, the present invention solves the problems of insufficient dosing accuracy, unstable treatment effect, serious waste of chemicals, low degree of intelligence and poor real-time performance, significantly improves the efficiency, stability and economy of desulfurization wastewater treatment, and at the same time reduces the impact on the environment.

[0079] In one of the embodiments, the preprocessing of the collected water quality data includes:

[0080] Clean and standardize the water quality data. The standardization process includes: outlier processing, null value filling, abnormal value removal and noise interference removal.

[0081] Taking outliers as an example, they are usually defined as observed values that are significantly different from most points in the data set. To quickly identify the data distribution of each index for further analysis of the water quality of the wastewater.

[0082] By preprocessing the water quality data, the subsequent input data becomes more accurate and standardized.

[0083] In one of the embodiments, the outlier processing uses an algorithm based on double-threshold adjusted quartile outliers to calculate the discrete range of the water quality data.

[0084] Optionally, use a double-threshold adjusted quartile outlier judgment technique. Let Q1 be the first quartile, Q3 be the third quartile, and IQR be the interquartile range, i.e., IQR = Q3 - Q1. Then the outlier can be defined as

[0085] Outlier = {x|x < Q1 - 1.5×IQR - t1 or x > Q3 + 1.5×IQR + t2}

[0086] Where, t1 and t2 are the user-defined lower and upper thresholds respectively, which are mainly used to flexibly control the outlier range in different scenarios.

[0087] like Figure 2 As shown, in one embodiment, the use of a neural network model to predict the dosage of water quality data to obtain a first predicted dosage includes:

[0088] The adopted neural network includes: an input layer, a feature selection layer, a hidden layer and an output layer. The feature selection layer is set before the hidden layer, and the output of the output layer is the first predicted dosage.

[0089] In wastewater quality monitoring, we usually face a large amount of water quality data, such as dosage, pH value, total nitrogen, total phosphorus, heavy metal content (copper, cadmium, silver, lead, chromium), suspended solids concentration, CODcr (chemical oxygen demand), TOC (total organic carbon), ammonia nitrogen and water intake, etc. There may be redundancy or correlation between these parameters. If all parameters are directly used as input features of the neural network, it may cause the model to overfit, that is, the model performs well on the training data, but has poor generalization ability on new and unseen data. To this end, it is necessary to perform selection operations on the input features. Through feature selection, we can remove redundant or unimportant features and retain the most representative features, thereby improving the generalization ability and prediction accuracy of the model. However, most of the existing feature selection technologies adopt a working mode separated from the target model (such as the neural network model), and the representative features selected are not the optimal features of the target model, which can easily affect the performance of the target model. To solve this problem, this scheme intends to embed feature selection technology into the neural network to realize the dynamic selection of wastewater indicator parameters. In Figure 5 A feature selection layer is added to the neural network model to improve the generalization ability and prediction accuracy of the model.

[0090] In one embodiment, the neural network input layer X=[x1, x2, ..., x m ], where x i represents the i-th water quality data;

[0091] The feature selection layer uses the gate vector to realize feature selection. Let the gate vector be G = [g1, g2, ..., g m ], where m represents the number of water quality data, and the neural network output is expressed as:

[0092]

[0093] Among them, W l is the weight matrix of the output layer, b l is the bias term of the output layer, Z l-1 is the output of the previous layer, Z l is the output layer output, σ is the activation function;

[0094] Using the mean square error formula to calculate the variance And adjust the neural network based on the variance:

[0095]

[0096] in, is the output of the last layer of the neural network, That is, the first predicted dosage, Y is the standard data matrix, ||W1|| 2,1 is the constraint matrix.

[0097] This embodiment intends to construct a neural network model through the following three steps:

[0098] First, a primitive neural network model is constructed. For a multilayer perceptron (MLP), this model has an input layer, a hidden layer, and an output layer. However, some water quality data may not be important for dosing prediction. Simply feeding all collected water quality data into the neural network model would inevitably cause network training oscillations and degrade prediction performance. To further optimize model performance, this proposal proposes introducing a feature selection mechanism into the neural network model.

[0099] Let the input layer X=[x1,x2,…,x m ], where x i Represents the i-th water quality data. Assuming the network has L layers, the output of the l-th layer is Z l =σ(W l Z l-1 +b l ), where Z l-1 Represents the network output of the l-1 layer. Let the final output of the network be but Assume that the label data matrix is ​​Y. This scheme intends to use mean square error to train the neural network:

[0100]

[0101] However, some water quality data may not be important for dosing prediction. If all collected water quality data is directly input into the neural network, it will inevitably cause network training oscillation and degrade prediction performance. To further optimize model performance, this solution proposes to introduce a feature selection mechanism into the neural network.

[0102] The neural network input layer used is X=[x1,x2,…,x m ], where x i represents the i-th water quality data;

[0103] The feature selection layer uses the gate vector to realize feature selection. Let the gate vector be G = [g1, g2, ..., g m ], where m represents the number of water quality data, and the neural network output is expressed as:

[0104]

[0105] Among them, W l is the weight matrix of the output layer, b l is the bias term of the output layer, Z l-1 is the output of the previous layer, Z l is the output layer output, σ is the activation function;

[0106] Using the mean square error formula to calculate the variance And adjust the neural network based on the variance:

[0107]

[0108] in, is the output of the last layer of the neural network, That is, the first predicted dosage, Y is the standard data matrix, ||W l || 2,1 is the constraint matrix.

[0109] where ||·|| 2,1 Indicates l 2,1 Norm, for a given matrix M=[m1,m2,…,m n ]∈R d×n , its 2,2 The norm can be expressed as:

[0110]

[0111] From the above formula we can see that l 2,1 The norm can realize the row-wise constraint matrix M, thereby suppressing the noise features in the data and effectively guiding the learning of important features in the neural network.

[0112] ||W l || 2,1 When calculating:

[0113] By introducing a feature selection layer, redundant water quality data is removed, thereby improving prediction accuracy.

[0114] In one embodiment, the step of using the fuzzy control model to predict the dosage to obtain the second predicted dosage includes:

[0115] Let X′ i =F fuzzy (x i ) means that the water quality data x i Convert to linguistic variable X′ in fuzzy set i , where F fuzzy (x i ) represents the fuzzy processing function;

[0116] Assume the fuzzy rule base is: Among them, Rule j represents the jth fuzzy rule, and B j Represent the linguistic variables of the input and output fuzzy sets respectively. According to the fuzzy rules, we can get:

[0117]

[0118] Among them, ∪ represents the fuzzy union operation, and X′ represents the fuzzified input variable;

[0119] Output the fuzzy dosage Y fuzzy Convert to a specific dosage value Y′, where Y′ is the second predicted dosage:

[0120] Y′=F defuzzy (Y fuzzy )

[0121] Among them, F defuzzy Represents the defuzzification processing function, which converts the fuzzy dosage output into a specific dosage value.

[0122] The neural network learns the complex mapping relationship between input water quality data and output dosage through training. Its output is a precise value representing the predicted dosage, which is calculated internally by the neural network based on the input water quality data. Neural networks are good at processing complex nonlinear mapping relationships, but they have difficulty handling uncertainty or fuzzy information and are easily interfered by external noise. To solve this problem, the solution proposes to introduce fuzzy control technology based on the neural network. Let X′ i =F fuzzy (x i ) means that the water quality index x i Convert to linguistic variable X′ in fuzzy set i , where F fuzzy (x i ) represents the fuzzification processing function.

[0123] Assume the fuzzy rule base is: Among them, Rule j represents the jth fuzzy rule, and B j Represent the linguistic variables of the input and output fuzzy sets respectively. According to the fuzzy rules, we can get:

[0124]

[0125] Among them, ∪ represents the fuzzy union operation, and X′ represents the fuzzified input variable.

[0126] Output the above fuzzy dosage Y fuzzy Convert to specific dosage value Y′:

[0127] Y′=F defuzzy (Y fuzzy )

[0128] Among them, F defuzzy Represents the defuzzification processing function, which converts the fuzzy dosage output into a specific dosage value.

[0129] By introducing the fuzzy control model, the entire model can handle uncertainty or fuzzy information, making it less susceptible to external noise interference.

[0130] In one embodiment, calculating the optimal dosage according to the first predicted dosage and the second predicted dosage includes:

[0131] Final dosage Y final :

[0132]

[0133] Among them, α is the weighting coefficient, is the first predicted dosage, and Y′ is the second predicted dosage.

[0134] Finally, the results of the neural network model and the fuzzy control model are combined to predict the dosage. If the input water quality data contains a large amount of fuzzy data, the weight of the second predicted dosage output by the fuzzy control model is increased; otherwise, the weight of the first predicted dosage output by the neural network model is increased. This embodiment allows the dosage prediction to be adjusted according to actual conditions, thereby better matching the actual situation and achieving more accurate predictions.

[0135] The present invention further provides an electronic device, comprising:

[0136] at least one processor; and,

[0137] a memory communicatively connected to at least one of the processors; wherein,

[0138] The memory stores instructions that can be executed by at least one of the processors, and the instructions are executed by at least one of the processors to enable at least one of the processors to execute the method for treating desulfurization wastewater by adding chemicals as described in any one of the above items.

[0139] The present invention solves problems such as insufficient dosing accuracy, unstable treatment effect, serious waste of reagents, low intelligence level and poor real-time performance by real-time monitoring of multi-dimensional water quality data and combining it with intelligent algorithms to dynamically predict the dosage. It significantly improves the efficiency, stability and economy of desulfurization wastewater treatment while reducing the impact on the environment.

[0140] The present invention also provides a storage medium storing computer instructions that, when executed by a computer, are used to perform any of the above methods for treating desulfurization wastewater by dosing. By monitoring multi-dimensional water quality data in real time and combining it with an intelligent algorithm to dynamically predict dosing dosage, the present invention addresses issues such as insufficient dosing accuracy, unstable treatment results, significant reagent waste, low intelligence, and poor real-time performance. This significantly improves the efficiency, stability, and cost-effectiveness of desulfurization wastewater treatment while reducing environmental impact.

[0141] The present invention also provides a computer program product comprising a computer program / instructions, characterized in that, when executed by a processor, the computer program / instructions implement the method for treating desulfurization wastewater by dosing as described in any of the above items. By monitoring multi-dimensional water quality data in real time and combining it with an intelligent algorithm to dynamically predict dosing dosage, the present invention addresses issues such as insufficient dosing accuracy, unstable treatment results, significant reagent waste, low intelligence, and poor real-time performance. This significantly improves the efficiency, stability, and economic efficiency of desulfurization wastewater treatment while reducing environmental impact.

[0142] Reference Figure 4 The present invention also provides a hardware structure diagram of an electronic device, including:

[0143] at least one processor 301; and,

[0144] A memory 302 in communication with at least one of the processors 301; wherein,

[0145] The memory 302 stores instructions that can be executed by at least one processor. The instructions are executed by the at least one processor so that the at least one processor can execute the method for treating desulfurization wastewater by adding chemicals as described above.

[0146] Figure 4 A processor 301 is taken as an example.

[0147] The electronic device is preferably a controller of the vehicle. The electronic device may further include: an input device 303 and a display device 304 .

[0148] The processor 301 , the memory 302 , the input device 303 and the display device 304 may be connected via a bus or other means, with the bus connection being used as an example in the figure.

[0149] The memory 302 is a non-volatile computer-readable storage medium that can be used to store non-volatile software programs, non-volatile computer executable programs and modules, such as the program instructions / modules corresponding to the method for treating desulfurization wastewater by adding chemicals in the embodiment of the present application, for example, Figure 1 The processor 301 executes various functional applications and data processing by running the non-volatile software programs, instructions and modules stored in the memory 302, that is, implementing the method for treating desulfurization wastewater by adding chemicals in the above embodiment.

[0150] The memory 302 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created according to the method for treating desulfurization wastewater by adding chemicals, etc. In addition, the memory 302 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 302 may optionally include a memory remotely located relative to the processor 301, and these remote memories may be connected to the device for executing the method for treating desulfurization wastewater by adding chemicals via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0151] The input device 303 can receive user clicks and generate signal input related to user settings and function control of the method for treating desulfurization wastewater by adding chemicals. The display device 304 can include a display device such as a display screen.

[0152] One or more modules are stored in the memory 302 and, when executed by one or more processors 301 , execute the method for treating desulfurization wastewater by adding chemicals in any of the above method embodiments.

[0153] An embodiment of the present invention provides a storage medium, which stores computer instructions. When a computer executes the computer instructions, it is used to execute all steps of the method for treating desulfurization wastewater by adding chemicals as described above.

[0154] In the context of the present disclosure, a storage medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or apparatus. The storage medium may be a machine-readable signal medium or a machine-readable storage medium. Alternatively, the storage medium may be a non-transitory computer-readable storage medium, for example, a non-transitory computer-readable storage medium may be a ROM, a random access memory (RAM), a compact disc read-only memory (CD-ROM), a magnetic tape, a floppy disk, and an optical data storage device.

[0155] The above description is only the principle and preferred embodiment of the present invention. It should be noted that for those skilled in the art, several other variations can be made based on the principle of the present invention, which should also be considered as the scope of protection of the present invention.

Claims

1. A method for treating desulfurization wastewater by adding chemicals, characterized in that: include: Collect water quality data of desulfurization wastewater and pre-process the collected water quality data; The neural network model is used to predict the dosage of water quality data to obtain the first predicted dosage; The dosage is predicted by using a fuzzy control model to obtain a second predicted dosage; Calculating an optimal dosage according to the first predicted dosage and the second predicted dosage; Add the optimal dosage to the desulfurization wastewater, collect the water quality data after the addition, and optimize the neural network model and the fuzzy control model based on the water quality data after the addition.

2. The method for treating desulfurization wastewater by adding chemicals according to claim 1, characterized in that: The pre-processing of the collected water quality data includes: The water quality data is cleaned and standardized. The standardization process includes: outlier processing, null value filling, outlier removal and noise interference removal.

3. The method for treating desulfurization wastewater by adding chemicals according to claim 2, characterized in that: The outlier processing adopts an algorithm based on double thresholds to adjust quartile outliers to calculate the discrete range of water quality data.

4. The method for treating desulfurization wastewater by adding chemicals according to claim 1, characterized in that: The method of using a neural network model to predict the dosage of water quality data to obtain a first predicted dosage includes: The adopted neural network includes: an input layer, a feature selection layer, a hidden layer and an output layer. The feature selection layer is set before the hidden layer, and the output of the output layer is the first predicted dosage.

5. The method for treating desulfurization wastewater by adding chemicals according to claim 4, characterized in that: The neural network input layer used is X=[x1,x2,…,x m ], where x i represents the i-th water quality data; The feature selection layer uses the gate vector to realize feature selection. Let the gate vector be G = [g1, g2, ..., g m ], where m represents the number of water quality data, and the neural network output is expressed as: Among them, W l is the weight matrix of the output layer, b l is the bias term of the output layer, Z l-1 is the output of the previous layer, Z l is the output layer output, σ is the activation function; Using the mean square error formula to calculate the variance And adjust the neural network based on the variance: in, is the output of the last layer of the neural network, That is, the first predicted dosage, Y is the standard data matrix, ||W l || 2,1 is the constraint matrix.

6. The method for treating desulfurization wastewater by adding chemicals according to claim 1, characterized in that: The method of using the fuzzy control model to predict the dosage to obtain the second predicted dosage includes: Let X′ i =F fuzzy (x i ) means that the water quality data x i Convert to linguistic variable X′ in fuzzy set i , where F fuzzy (x i ) represents the fuzzy processing function; Assume the fuzzy rule base is: Among them, Rule j represents the jth fuzzy rule, and B j Represent the linguistic variables of the input and output fuzzy sets respectively. According to the fuzzy rules, we can get: Among them, ∪ represents the fuzzy union operation, X ′ represents the input variable after fuzzification; Output the fuzzy dosage Y fuzzy Convert to specific dosage value Y ′ , Y ′ That is the second predicted dosage: Y ′ =F defuzzy (Y fuzzy ) Among them, F defuzzy Represents the defuzzification processing function, which converts the fuzzy dosage output into a specific dosage value.

7. The method for treating desulfurization wastewater by adding chemicals according to claim 1, characterized in that: Calculating the optimal dosage according to the first predicted dosage and the second predicted dosage includes: Final dosage Y final : Among them, α is the weighting coefficient, is the first predicted dosage, Y ′ This is the second predicted dosage.

8. An electronic device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to at least one of the processors; wherein, The memory stores instructions that can be executed by at least one of the processors, and the instructions are executed by at least one of the processors to enable at least one of the processors to execute the method for treating desulfurization wastewater by adding chemicals as described in any one of claims 1 to 7.

9. A storage medium, characterized in that: The storage medium stores computer instructions, and when a computer executes the computer instructions, it is used to execute the method for treating desulfurization wastewater by adding chemicals as described in any one of claims 1 to 7.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the method for treating desulfurization wastewater by adding chemicals as described in any one of claims 1 to 7 is implemented.

Images