CN113094992A - Desulfurizing tower optimization method based on neural network discrimination - Google Patents

Desulfurizing tower optimization method based on neural network discrimination Download PDF

Info

Publication number
CN113094992A
CN113094992A CN202110386338.5A CN202110386338A CN113094992A CN 113094992 A CN113094992 A CN 113094992A CN 202110386338 A CN202110386338 A CN 202110386338A CN 113094992 A CN113094992 A CN 113094992A
Authority
CN
China
Prior art keywords
neural network
desulfurizing tower
model
output
discrimination
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202110386338.5A
Other languages
Chinese (zh)
Inventor
张志勇
阿茹娜
赵全中
姜冉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Inner Mongolia Electric Power Research Institute of Inner Mongolia Power Group Co Ltd
Original Assignee
Inner Mongolia Electric Power Research Institute of Inner Mongolia Power Group Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Inner Mongolia Electric Power Research Institute of Inner Mongolia Power Group Co Ltd filed Critical Inner Mongolia Electric Power Research Institute of Inner Mongolia Power Group Co Ltd
Priority to CN202110386338.5A priority Critical patent/CN113094992A/en
Publication of CN113094992A publication Critical patent/CN113094992A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2113/00—Details relating to the application field
    • G06F2113/08—Fluids

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biophysics (AREA)
  • Mathematical Physics (AREA)
  • Computing Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Geometry (AREA)
  • Medical Informatics (AREA)
  • Computer Hardware Design (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Treating Waste Gases (AREA)

Abstract

The invention belongs to the technical field of computers, and particularly relates to a desulfurizing tower optimization method based on neural network discrimination. According to the method, under the condition that the structure and the reaction mechanism of the absorption tower do not need to be researched, the input and output vectors of a neural network model are determined through the judgment of input and output operation parameters with close correlation of the desulfurization tower, the neural network structure of a single input layer, a single output layer and a single hidden layer is constructed, a training sample set covering the complete operation condition of the desulfurization tower is included, the optimal operation model of the desulfurization tower based on the neural network judgment is finally established, and the guide suggestion of operation control is provided for the real-time operation process of the desulfurization tower under different working conditions, so that the purposes of optimizing the operation parameters of the desulfurization tower, ensuring the stability of desulfurization efficiency and reducing the operation consumption are achieved.

Description

Desulfurizing tower optimization method based on neural network discrimination
Technical Field
The invention belongs to the technical field of computers, and particularly relates to a desulfurizing tower optimization method based on neural network discrimination.
Background
The reaction device in the wet flue gas desulfurization system is a desulfurization absorption tower which consists of a cylindrical tower body, a slurry pool in the tower, a spraying absorption zone (comprising a spraying layer and a nozzle), a mechanical demister, tower top wet electricity and the like. Wherein the heat and mass transfer process of the core mainly occurs in the spraying absorption area and the slurry pool in the tower. Therefore, how to accurately predict the heat and mass transfer processes of the spray absorption area and the slurry pool in the tower is further reflected on various operation parameters output by the model, and becomes the main content of the optimized operation modeling of the desulfurization tower.
At present, the modeling of the heat and mass transfer process of the desulfurizing tower is mainly based on the traditional double-membrane theory of gas-liquid phase reaction, a heat transfer basic equation and a bias mechanism model research, and the modeling is applied to actual operation and has certain deviation with actual operation parameters in the aspect of model precision. Different researchers and applications have explored this problem from different perspectives.
Researchers at the university of north china electric power, Zhang Xiaodong and the like establish a one-dimensional numerical calculation model for wet flue gas desulfurization. The study was directed to the absorption process of SO2 in a spray tower, making a one-dimensional simplifying assumption on the change in the composition of the material in the absorption zone. On the basis of discretizing the calculation region, a one-dimensional numerical calculation model of the desulfurization process is established by using a control volume equation reflecting material balance between gas-liquid two phases and chemical reaction balance in a liquid phase, and the setting mode of model parameters is discussed. The output of the model is the desulfurization efficiency.
The researchers of Shenxiang and the like of the university of southeast utilize a double-membrane theory as a basis to simulate the chemical reaction and the interphase mass transfer process inside the desulfurization spray tower and establish a mathematical model containing the concentration and the mass transfer rate of all sulfur-containing components. The obtained constant coefficient differential equation is solved by a Runge-Kutta method. Finally, a desulfurization efficiency model is obtained, and the model output is the change of the desulfurization efficiency.
The established mechanism model and numerical calculation model can be used for optimizing the desulfurization design process, but have no real-time performance, and meanwhile, the model needs to be based on a series of assumptions on the desulfurization system model condition and simplification of an object, is limited in model precision, cannot cover all working condition changes of the desulfurization operation process, and is not suitable for the field of desulfurization tower operation optimization.
In order to guide fault maintenance and operation optimization of a desulfurization system, researchers such as xudan at university of southeast take a limestone-gypsum wet desulfurization system as an object, discuss common fault modes and fault reasons of a slurry circulating pump, and establish a fault diagnosis model of the slurry circulating pump and an optimization model of limestone-gypsum wet desulfurization efficiency by using a fuzzy theory. The model is based on a fuzzy mathematical theory, a fuzzy rule criterion is adopted to diagnose and optimize the operation state of the circulating pump and the desulfurization efficiency, the fuzzy rule criterion is derived from the summary induction of a model builder on the correlation of the parameters of the operation process of the circulating pump and the desulfurization tower and the operation parameter range, the reasonability of the rule and the selection of the parameter range are both dependent on the subjective selection and experience of the model builder, so the precision and the accuracy of the model have uncertainty, in addition, the model has larger accuracy deviation aiming at different desulfurization system objects, and the fuzzy rule criterion and the parameter range are required to be reformulated under the manual control. The model has poor adaptability.
Researchers such as the huoluong of North China electric university deduce mathematical models of booster fan, oxidation fan and circulating slurry pump by studying the running conditions of equipment such as a gas-gas heat exchanger (GGH), a demister, a booster fan, a circulating slurry pump and an oxidation fan and combining the basic principle of fluid mechanics. And obtaining a characteristic curve of the blocking rate of the desulfurization system, the GGH differential pressure and the demister differential pressure and a relation of the total resistance coefficient, the GGH blocking rate and the demister blocking rate. For guiding the operation optimization of the desulfurization energy-consuming equipment. The model is mainly a single equipment model for desulfurizing each electric equipment for rotary machinery, is used for saving energy and reducing consumption of the equipment, does not relate to the optimization and control of core operation parameters of a desulfurizing tower, and is not suitable for the field of operation optimization of the desulfurizing tower.
Disclosure of Invention
Aiming at the defects in the prior art, the technical problems to be solved by the invention are as follows: the invention provides a desulfurization tower optimization method based on neural network discrimination, which is used for predicting main operation parameters of a desulfurization tower and guiding the operation optimization of the desulfurization tower, and aims to solve the problems that the operation of a desulfurization core reactor is relatively coarsely adjusted and controlled in the operation process and the operation of the desulfurization core reactor of the desulfurization system is not properly adjusted due to the fact that a wet flue gas desulfurization system matched with a boiler of a thermal power plant is lack of an accurate desulfurization core reactor reaction state prediction and optimization model guidance device to operate.
The invention adopts the following technical scheme:
the desulfurizing tower optimizing method based on neural network discrimination includes the following steps,
step one, acquiring relevant operation parameters of a desulfurizing tower and processing data;
and step two, inputting the processed data serving as an input vector into the neural network model, and obtaining an output vector.
According to the further optimization of the technical scheme, the relevant operation parameters of the desulfurizing tower comprise the flue gas volume, the original flue gas temperature, the limestone slurry flow, the total flow of the circulating pump in transportation and the gypsum slurry density.
According to the further optimization of the technical scheme, the output vector of the neural network model comprises the desulfurization efficiency, the slurry pH value and the clean flue gas temperature.
In the further optimization of the technical scheme, in the first step, the data is normalized, and the normalization processing formula is as follows:
Figure 872342DEST_PATH_IMAGE001
wherein a and b are constants, xmax、xminActual maximum and minimum values for each set of factor variables; x is the number ofiAnd xi ’The values of the factor variables before and after normalization, respectively.
In the further optimization of the technical scheme, the neural network model is a BP neural network.
In the further optimization of the technical scheme, the structure of the neural network model is a BP neural network structure with a single input layer, a single output layer and a 1-layer hidden layer.
In the further optimization of the technical scheme, the number of input layer nodes of the neural network model is 5 corresponding to the number of input vectors, the number of output layer nodes of the neural network model is 3 corresponding to the number of output vectors, and the number of initial hidden layer nodes of the neural network model is 7.
In the further optimization of the technical scheme, the output transformation function of the hidden layer neuron of the neural network model adopts a nonlinear sigmoid function, and the function is specifically as follows:
Figure 744483DEST_PATH_IMAGE002
where x is the input to the neuron and y is the output of the neuron.
In the further optimization of the technical scheme, the neural network model respectively corrects the weight values and the threshold values of the connecting lines of each network layer by adopting a first-order gradient method and a logarithmic reduction method until the error is smaller than a preset value.
Compared with the prior art, the model can be used for real-time prediction of the operation process of the desulfurizing tower after being trained and shaped, and the applicability of the model to the full-working-condition operation of the desulfurizing tower is ensured through the full-working-condition selection of the training sample. The model output parameter predicted value also comprises the pH value of slurry in the tower and the temperature of flue gas at the outlet of the desulfurization tower besides the desulfurization efficiency, so that the main symbolic operation parameters of the heat and mass transfer process of the desulfurization tower are covered, the model output is more comprehensive, the prediction accuracy can meet the requirement of the actual optimized operation process, and the model output parameter predicted value can be combined with the existing closed-loop control system of the flue gas desulfurization device to optimize the control and regulation process.
Drawings
FIG. 1 is a diagram of a desulfurizing tower neural network optimization model;
FIG. 2 is a graph of different numbers of iterative error distributions for hidden layer neurons;
FIG. 3 is a comparison graph of the predicted values of the output parameters and the measured values of the test samples;
fig. 4 is an iterative error plot.
Detailed Description
To further illustrate the various embodiments, the invention provides the accompanying drawings. The accompanying drawings, which are incorporated in and constitute a part of this disclosure, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the embodiments. Those skilled in the art will appreciate still other possible embodiments and advantages of the present invention with reference to these figures. Elements in the figures are not drawn to scale and like reference numerals are generally used to indicate like elements.
The invention will now be further described with reference to the accompanying drawings and detailed description.
The invention provides a desulfurization tower optimization method based on neural network discrimination, which determines input vectors of a neural network model as { Q (smoke flow), Tin (original smoke temperature), F (limestone slurry flow), Qx (total flow of a circulating pump in transportation) and M (gypsum slurry density) } through the discrimination of input and output operation parameters with close correlation of each desulfurization tower under the condition of not researching an absorption tower structure and a reaction mechanism. The output vector is { η (desulfurization efficiency), PH (slurry PH), Tout (clean flue gas temperature) }. The method comprises the steps of constructing a neural network structure of a single input layer, a single output layer and a single hidden layer, bringing a training sample set covering the complete operation condition of the desulfurizing tower into consideration, finally establishing a desulfurizing tower optimal operation model based on neural network discrimination, and providing guidance suggestions of operation control for the real-time operation process of the desulfurizing tower under different conditions, so that the aims of optimizing the operation parameters of the desulfurizing tower, ensuring stable desulfurizing efficiency and reducing operation consumption are fulfilled.
The embodiment relates to a desulfurizing tower optimizing method based on neural network discrimination, which comprises the following steps,
step one, obtaining relevant operation parameters of the desulfurizing tower, carrying out normalization processing on data,
the normalization processing formula is adopted as follows:
Figure 206688DEST_PATH_IMAGE003
wherein a and b are constants, xmax、xminActual maximum and minimum values for each set of factor variables; x is the number ofiAnd xi ’The values of the factor variables before and after normalization, respectively.
And step two, inputting the processed data serving as an input vector into the neural network model, and obtaining an output vector.
The establishment of the neural network model comprises the following steps:
first, determining an input vector and an output vector:
the embodiment determines the input vector of the neural network model of the invention as { Q (flue gas volume), T (flue gas volume) aiming at the correlation between the main operation parameters of the wet desulphurization absorption towerin(original flue gas temperature), F (limestone slurry flow), Qx (total flow of circulating pump in transportation), M (gypsum slurry density) }, and output vector is { eta (desulfurization efficiency), PH (slurry pH value), and Tout(clean flue gas temperature).
Secondly, determining a neural network model structure:
the optimization model adopts a BP neural network structure of a single input layer, a single output layer and 1 hidden layer, the number of nodes of the input layer is 5 corresponding to that of input vectors, the number of nodes of the output layer is 3 corresponding to that of output vectors, the number of nodes of the hidden layer can not be directly determined by theoretical calculation initially, the number of the initial hidden layer nodes is set to be 7 according to experience, the basic principle is to reduce the number of layers of the hidden layer so as to reduce the complexity of the neural network model structure, and lower errors are obtained by increasing the number of the nodes of the hidden layer. The number of training samples satisfies 5 times the number of connection weights of the neural network structure. Fig. 1 shows a diagram of a neural network optimization model of a desulfurizing tower.
Step three, neural network training:
the number of the model training samples is set to 300 groups, wherein 50 groups of low smoke load data samples (the smoke volume is 50% and below of the total smoke volume of desulfurization design), 100 groups of medium smoke load data samples (the smoke volume is 50% -75% of the total smoke volume of desulfurization design), 100 groups of high smoke load data samples (the smoke volume is 75% -100% of the total smoke volume of desulfurization design), and 50 groups of full smoke load data samples (the smoke volume is the total smoke volume of desulfurization design). Before training sample data enters model training, normalization processing is required.
The output transformation function of the hidden layer neuron adopts a nonlinear sigmoid function, and the function is specifically as follows:
Figure 226597DEST_PATH_IMAGE004
wherein x is the input to the neuron; y is the output of the neuron.
And establishing the basic structure of the optimized operation model of the desulfurizing tower.
When the training sample mentioned above is used for model training, the working signal is transmitted in the forward direction, the error is transmitted in the reverse direction, and the first-order gradient method and the logarithmic reduction method are used for respectively and continuously correcting the weight value and the threshold value of each network connection line until the error is less than the preset value by less than 5%, so that the requirement of model precision is met. The final neuron number of the hidden layer is determined by the iterative error. The above training sample distribution grouping training is adopted, and the iteration error result is shown in fig. 2 and is an iteration error distribution diagram for different numbers of hidden layer neurons. The different numbers of neurons and the different numbers of hidden layers cause different computation complexity and precision, and it can be seen from the figure that the number of neurons in the hidden layers is different, the iteration precision of the neural network model after training is different, and under the condition of adopting 7 neurons, the iteration precision is the highest, reaching the level of 3.9%, and lower than the preset value, so that the number of neurons in the hidden layers of the optimized neural network model of the final desulfurizing tower is determined to be 7.
Step four, model testing:
and in order to verify the reliability of the model, testing the established model after the training is finished. Similarly, random samples other than a group of training samples of low load, heavy load, high load and full load of the same desulfurization tower object under different working conditions are selected as test samples, and taking the desulfurization efficiency of one of the output parameters as an example, the predicted value of the output parameter and the measured value of the test sample after calculation by the prediction model are shown in fig. 3 as a comparison graph of the predicted value of the output parameter and the measured value of the test sample. The relative error parameters are all less than 2%, the average value of the relative errors is 1.39%, and the precision is good.
The prediction accuracy of the optimized operation model of the desulfurizing tower can meet the requirement of the actual optimized operation process, and can be combined with the existing closed-loop control system of the flue gas desulfurizing device to optimize, control and regulate the process.
And obtaining an available optimal operation neural network model of the desulfurizing tower through the four steps.
The example is as follows:
1. the model structure is as follows: input vector { Q (flue gas volume), Tin(original flue gas temperature), F (limestone slurry flow), Qx (total flow of circulating pump in transit), M (gypsum slurry density), and output vector { eta (desulfurization efficiency), PH (slurry pH value), Tout(clean flue gas temperature). The model is a BP neural network structure with a single input layer, a single output layer and a 1-layer hidden layer, the number of input layer nodes is 5 corresponding to the number of input vectors, the number of output layer nodes is 3 corresponding to the number of output vectors, and the number of hidden layer nodes is 7.
2. Aiming at the running data of a flue gas desulfurization system of a 600MW unit of a certain power plant, after screening, the number of samples of the model is set to 300 groups, wherein 50 groups of low-smoke load data samples (the smoke amount is 50% or less of the total designed flue gas amount), 100 groups of medium-smoke load data samples (the smoke amount is 50% -75% of the total designed flue gas amount), 100 groups of high-smoke load data samples (the smoke amount is 75% -100% of the total designed flue gas amount), and 50 groups of full-smoke load data samples (the smoke amount is the total designed flue gas amount). The 300 training samples after normalization are shown in table 1 below.
TABLE 1 training samples
Q (smoke volume) Tin (original smoke temperature) F (limestone slurry flow) Qx (total flow of circulating pump) M (Gypsum slurry density) Eta (desulfurization efficiency) PH (slurry pH value) Tout (Smoke temperature)
0.45 0.76 0.51 0.50 0.95 0.93 0.94 0.92
0.47 0.73 0.49 0.50 0.93 0.94 0.95 0.93
0.43 0.72 0.44 0.50 0.92 0.93 0.93 0.92
0.39 0.68 0.41 0.50 0.91 0.92 0.93 0.90
0.45 0.75 0.50 0.50 0.94 0.93 0.93 0.92
0.38 0.68 0.40 0.50 0.92 0.92 0.96 0.91
0.41 0.69 0.44 0.50 0.92 0.91 0.94 0.93
0.42 0.71 0.44 0.50 0.93 0.92 0.92 0.94
0.45 0.76 0.51 0.50 0.95 0.93 0.94 0.92
0.47 0.73 0.49 0.50 0.93 0.94 0.95 0.93
0.45 0.75 0.50 0.50 0.94 0.93 0.93 0.92
0.49 0.78 0.51 0.50 0.95 0.94 0.96 0.94
0.43 0.72 0.44 0.50 0.92 0.93 0.93 0.92
0.39 0.68 0.41 0.50 0.91 0.92 0.93 0.90
0.45 0.74 0.49 0.50 0.94 0.93 0.94 0.92
0.38 0.68 0.40 0.50 0.92 0.92 0.96 0.91
0.38 0.67 0.41 0.50 0.91 0.91 0.95 0.92
0.41 0.69 0.44 0.50 0.92 0.91 0.94 0.93
0.42 0.71 0.45 0.50 0.93 0.92 0.92 0.93
0.45 0.77 0.52 0.50 0.95 0.93 0.94 0.92
0.49 0.78 0.55 0.50 0.97 0.95 0.96 0.94
0.43 0.72 0.44 0.50 0.92 0.93 0.93 0.92
0.39 0.68 0.41 0.50 0.91 0.92 0.93 0.90
0.45 0.75 0.50 0.50 0.94 0.93 0.93 0.92
0.38 0.68 0.40 0.50 0.92 0.92 0.96 0.91
0.41 0.69 0.44 0.50 0.92 0.91 0.94 0.93
0.42 0.71 0.44 0.50 0.93 0.92 0.92 0.94
0.49 0.78 0.51 0.50 0.95 0.94 0.96 0.94
0.43 0.72 0.44 0.50 0.92 0.93 0.93 0.92
0.39 0.68 0.41 0.50 0.91 0.92 0.93 0.90
0.50 0.72 0.51 0.50 0.96 0.95 0.96 0.95
0.50 0.71 0.52 0.75 0.95 0.94 0.97 0.94
0.45 0.76 0.50 0.50 0.93 0.94 0.92 0.93
0.47 0.73 0.49 0.50 0.93 0.94 0.95 0.93
0.43 0.72 0.44 0.50 0.92 0.93 0.93 0.92
0.39 0.68 0.41 0.50 0.91 0.92 0.93 0.90
0.45 0.73 0.50 0.50 0.93 0.94 0.95 0.93
0.41 0.69 0.44 0.50 0.92 0.91 0.94 0.93
0.42 0.71 0.45 0.50 0.93 0.92 0.92 0.93
0.45 0.77 0.52 0.50 0.95 0.93 0.94 0.92
0.47 0.76 0.51 0.50 0.94 0.94 0.95 0.93
0.42 0.71 0.46 0.50 0.93 0.93 0.96 0.93
0.49 0.79 0.55 0.75 0.97 0.98 0.99 0.96
0.43 0.71 0.45 0.50 0.95 0.94 0.95 0.93
0.45 0.69 0.45 0.50 0.94 0.94 0.93 0.95
0.43 0.72 0.44 0.50 0.92 0.93 0.93 0.92
0.39 0.68 0.41 0.50 0.91 0.92 0.93 0.90
0.45 0.75 0.50 0.50 0.94 0.93 0.93 0.92
0.38 0.68 0.40 0.50 0.92 0.92 0.96 0.91
0.66 0.79 0.70 0.75 0.96 0.97 0.99 0.97
0.58 0.75 0.66 0.75 0.95 0.96 0.97 0.96
0.63 0.73 0.71 0.75 0.94 0.94 0.95 0.95
0.67 0.70 0.69 0.75 0.96 0.96 0.96 0.98
0.70 0.78 0.73 0.75 0.95 0.95 0.97 0.97
0.71 0.80 0.75 0.75 0.94 0.97 0.95 0.96
0.65 0.70 0.71 0.75 0.96 0.96 0.95 0.95
0.72 0.77 0.70 0.75 0.94 0.95 0.97 0.99
0.55 0.64 0.67 0.50 0.93 0.93 0.96 0.98
0.62 0.68 0.70 0.75 0.96 0.96 0.98 0.96
0.64 0.67 0.69 0.50 0.95 0.92 0.95 0.94
0.69 0.72 0.71 0.75 0.95 0.95 0.93 0.97
0.65 0.69 0.68 0.50 0.91 0.91 0.94 0.95
0.74 0.79 0.77 0.75 0.98 0.98 0.97 0.99
0.66 0.72 0.70 0.75 0.96 0.97 0.96 0.96
0.58 0.62 0.66 0.75 0.95 0.96 0.97 0.96
0.63 0.69 0.71 0.75 0.94 0.94 0.95 0.95
0.70 0.78 0.72 0.75 0.95 0.95 0.97 0.97
0.71 0.80 0.73 0.75 0.94 0.97 0.95 0.96
0.65 0.71 0.71 0.75 0.95 0.96 0.94 0.95
0.72 0.75 0.70 0.75 0.94 0.95 0.97 0.96
0.52 0.61 0.65 0.50 0.93 0.91 0.95 0.93
0.54 0.67 0.69 0.75 0.95 0.94 0.96 0.93
0.65 0.69 0.67 0.50 0.93 0.92 0.94 0.95
0.63 0.68 0.70 0.75 0.96 0.96 0.98 0.96
0.64 0.67 0.69 0.50 0.95 0.92 0.95 0.94
0.67 0.72 0.71 0.75 0.95 0.95 0.93 0.97
0.66 0.69 0.68 0.50 0.91 0.91 0.94 0.95
0.62 0.71 0.71 0.75 0.95 0.96 0.94 0.95
0.58 0.75 0.66 0.75 0.95 0.96 0.97 0.96
0.63 0.73 0.70 0.75 0.94 0.94 0.95 0.95
0.67 0.70 0.68 0.75 0.96 0.96 0.96 0.98
0.70 0.75 0.73 0.75 0.95 0.95 0.97 0.97
0.71 0.80 0.75 0.75 0.94 0.96 0.95 0.96
0.65 0.70 0.71 0.75 0.96 0.96 0.95 0.95
0.72 0.77 0.70 0.75 0.94 0.95 0.97 0.99
0.55 0.62 0.67 0.50 0.93 0.93 0.96 0.98
0.62 0.64 0.69 0.75 0.96 0.96 0.98 0.96
0.64 0.67 0.71 0.50 0.95 0.92 0.95 0.94
0.52 0.61 0.65 0.50 0.93 0.91 0.95 0.93
0.54 0.67 0.69 0.75 0.95 0.94 0.96 0.93
0.65 0.69 0.67 0.50 0.93 0.92 0.94 0.95
0.63 0.68 0.70 0.75 0.96 0.96 0.98 0.96
0.64 0.67 0.69 0.50 0.95 0.92 0.95 0.94
0.65 0.70 0.71 0.75 0.96 0.96 0.95 0.95
0.72 0.77 0.70 0.75 0.94 0.95 0.97 0.99
0.55 0.64 0.67 0.50 0.93 0.93 0.96 0.98
0.62 0.68 0.70 0.75 0.96 0.96 0.98 0.96
0.71 0.80 0.75 0.75 0.94 0.96 0.95 0.96
0.65 0.70 0.71 0.75 0.96 0.96 0.95 0.95
0.72 0.77 0.70 0.75 0.94 0.95 0.97 0.99
0.64 0.67 0.69 0.50 0.95 0.92 0.95 0.94
0.69 0.72 0.71 0.75 0.95 0.95 0.93 0.97
0.65 0.69 0.68 0.50 0.91 0.91 0.94 0.95
0.74 0.79 0.77 0.75 0.98 0.98 0.97 0.99
0.66 0.72 0.70 0.75 0.96 0.97 0.96 0.96
0.52 0.56 0.65 0.50 0.93 0.91 0.95 0.93
0.54 0.57 0.69 0.75 0.95 0.94 0.96 0.95
0.65 0.71 0.67 0.50 0.93 0.92 0.94 0.94
0.63 0.64 0.69 0.75 0.96 0.96 0.98 0.95
0.64 0.67 0.69 0.50 0.95 0.92 0.95 0.96
0.67 0.72 0.71 0.75 0.95 0.95 0.93 0.94
0.66 0.69 0.68 0.50 0.91 0.91 0.94 0.96
0.62 0.66 0.71 0.75 0.95 0.96 0.94 0.95
0.66 0.65 0.66 0.50 0.91 0.91 0.94 0.95
0.62 0.71 0.72 0.75 0.94 0.96 0.94 0.95
0.58 0.75 0.65 0.75 0.95 0.95 0.97 0.96
0.63 0.73 0.70 0.75 0.94 0.94 0.95 0.95
0.67 0.70 0.68 0.75 0.96 0.96 0.96 0.98
0.70 0.75 0.73 0.75 0.95 0.95 0.97 0.97
0.71 0.80 0.75 0.75 0.94 0.96 0.95 0.96
0.65 0.70 0.71 0.75 0.96 0.96 0.95 0.95
0.72 0.77 0.70 0.75 0.94 0.95 0.97 0.99
0.55 0.62 0.67 0.50 0.93 0.93 0.96 0.98
0.62 0.64 0.69 0.75 0.96 0.96 0.98 0.96
0.64 0.67 0.71 0.50 0.95 0.92 0.95 0.94
0.52 0.61 0.65 0.50 0.93 0.91 0.95 0.93
0.54 0.67 0.69 0.75 0.95 0.94 0.96 0.93
0.65 0.69 0.67 0.50 0.93 0.92 0.94 0.95
0.63 0.68 0.70 0.75 0.96 0.96 0.98 0.96
0.64 0.67 0.69 0.50 0.95 0.92 0.95 0.94
0.65 0.70 0.71 0.75 0.96 0.96 0.95 0.95
0.72 0.77 0.70 0.75 0.94 0.95 0.97 0.99
0.55 0.64 0.67 0.50 0.93 0.93 0.96 0.98
0.73 0.78 0.80 1.00 0.97 0.99 0.98 0.99
0.74 0.81 0.82 1.00 0.98 0.99 0.97 0.98
0.72 0.79 0.80 1.00 0.97 0.98 0.99 0.97
0.70 0.75 0.76 0.75 0.95 0.95 0.96 0.96
0.70 0.75 0.73 0.75 0.95 0.95 0.97 0.97
0.71 0.80 0.75 0.75 0.94 0.96 0.95 0.96
0.65 0.70 0.71 0.75 0.96 0.96 0.95 0.95
0.72 0.77 0.70 0.75 0.94 0.95 0.97 0.99
0.57 0.62 0.67 0.50 0.93 0.93 0.96 0.98
0.74 0.79 0.77 0.75 0.98 0.98 0.97 0.99
0.66 0.72 0.70 0.75 0.96 0.97 0.96 0.96
0.58 0.62 0.66 0.75 0.95 0.96 0.97 0.96
0.63 0.69 0.71 0.75 0.94 0.94 0.95 0.95
0.70 0.78 0.72 0.75 0.95 0.95 0.97 0.97
0.71 0.80 0.73 0.75 0.94 0.97 0.95 0.96
0.65 0.71 0.71 0.75 0.95 0.96 0.94 0.95
0.72 0.75 0.70 0.75 0.94 0.95 0.97 0.96
0.76 0.80 0.81 1.00 0.97 0.98 0.96 0.98
0.78 0.82 0.84 1.00 0.98 0.99 0.98 0.98
0.83 0.85 0.88 1.00 0.96 0.97 0.96 0.97
0.82 0.86 0.87 1.00 0.95 0.98 0.97 0.96
0.85 0.89 0.88 1.00 0.97 0.98 0.99 0.99
0.90 0.92 0.93 1.00 0.96 0.97 0.97 0.98
0.87 0.86 0.89 1.00 0.96 0.98 0.97 0.96
0.93 0.95 0.96 1.00 0.97 0.97 0.98 0.96
0.99 1.00 1.00 1.00 1.00 1.00 1.00 1.02
0.95 1.02 1.03 1.00 1.00 1.01 1.00 1.00
0.94 0.98 1.01 1.00 1.01 1.00 1.02 1.00
0.96 1.02 1.03 1.00 1.00 1.01 1.00 1.00
0.88 0.95 0.98 1.00 1.00 1.00 0.98 1.00
0.89 0.97 0.98 1.00 0.99 1.00 1.00 1.00
0.92 0.98 1.00 1.00 0.99 1.00 1.01 0.98
0.98 1.00 1.03 1.00 1.01 1.00 1.02 1.00
0.99 1.05 1.10 1.00 1.03 1.00 1.00 1.00
0.96 1.02 1.05 1.00 1.00 1.02 1,01 0.97
0.82 0.90 0.89 1.00 0.95 0.97 0.94 0.96
0.77 0.88 0.86 1.00 0.94 0.95 0.97 0.94
0.76 0.80 0.81 1.00 0.97 0.98 0.96 0.98
0.78 0.82 0.84 1.00 0.98 0.99 0.98 0.98
0.83 0.85 0.88 1.00 0.96 0.97 0.96 0.97
0.82 0.86 0.87 1.00 0.95 0.98 0.97 0.96
0.90 0.92 0.93 1.00 0.96 0.97 0.97 0.98
0.87 0.86 0.89 1.00 0.96 0.98 0.97 0.96
0.93 0.95 0.96 1.00 0.97 0.97 0.98 0.96
0.99 1.00 1.00 1.00 1.01 1.00 1.02 1.02
0.95 1.02 1.03 1.00 1.00 1.01 0.99 1.00
0.94 0.98 1.01 1.00 1.01 1.00 1.02 1.00
0.96 1.02 1.03 1.00 0.98 1.01 0.99 1.00
0.98 1.00 1.03 1.00 1.01 1.00 1.02 1.00
0.99 1.05 1.10 1.00 1.03 1.00 1.00 1.00
0.96 1.02 1.05 1.00 1.00 1.02 1,01 0.97
0.82 0.90 0.89 1.00 0.95 0.97 0.94 0.96
0.77 0.88 0.86 1.00 0.94 0.95 0.97 0.94
0.76 0.80 0.81 1.00 0.97 0.98 0.96 0.98
0.78 0.82 0.84 1.00 0.98 0.99 0.98 0.98
0.83 0.85 0.88 1.00 0.96 0.97 0.96 0.97
0.82 0.86 0.87 1.00 0.95 0.98 0.97 0.96
0.82 0.84 0.92 1.00 0.97 0.95 0.96 0.96
0.90 0.92 0.93 1.00 0.96 0.97 0.94 0.98
0.87 0.86 0.89 1.00 0.96 0.98 0.96 0.96
0.93 0.95 0.96 1.00 0.97 0.97 0.98 0.96
0.99 1.00 1.00 1.00 1.00 1.00 1.02 1.02
0.95 1.02 1.03 1.00 1.00 1.01 0.99 1.00
0.94 0.98 1.01 1.00 1.01 1.00 1.02 1.00
0.96 1.02 1.03 1.00 0.98 1.01 0.99 1.00
0.98 1.00 1.03 1.00 1.01 1.00 1.02 1.00
0.99 1.05 1.08 1.00 1.02 1.00 1.00 1.00
0.96 1.02 1.05 1.00 0.99 1.02 1,01 0.97
0.82 0.89 0.86 1.00 0.95 0.97 0.94 0.96
0.75 0.81 0.88 1.00 0.97 0.95 0.98 1.00
0.99 1.00 1.00 1.00 1.01 1.00 1.02 1.02
0.95 1.02 1.03 1.00 1.00 1.01 0.99 0.99
0.94 0.98 1.01 1.00 1.01 1.00 1.02 1.01
0.96 1.02 1.03 1.00 0.98 1.01 0.99 0.98
0.82 0.86 0.92 1.00 0.97 0.95 0.96 0.96
0.90 0.92 0.93 1.00 0.96 0.97 0.94 0.98
0.87 0.89 0.89 1.00 0.96 0.98 0.96 1.00
0.93 0.95 0.96 1.00 0.97 0.97 0.98 0.99
0.78 0.82 0.84 1.00 0.98 0.99 0.98 0.98
0.83 0.85 0.88 1.00 0.96 0.97 0.96 0.97
0.82 0.86 0.87 1.00 0.95 0.98 0.97 0.96
0.90 0.92 0.93 1.00 0.96 0.97 0.97 0.98
0.87 0.86 0.89 1.00 0.96 0.98 0.97 0.96
0.93 0.95 0.96 1.00 0.97 0.97 0.98 0.96
0.99 1.00 1.00 1.00 1.01 1.00 1.02 1.02
0.95 1.02 1.03 1.00 1.00 1.01 0.99 1.00
0.94 0.98 1.01 1.00 1.01 1.00 1.02 1.00
0.96 1.02 1.03 1.00 0.98 1.01 0.99 1.00
0.98 1.00 1.03 1.00 1.01 1.00 1.02 1.00
0.99 1.05 1.10 1.00 1.03 1.00 1.00 1.00
0.96 1.02 1.05 1.00 1.00 1.02 1,01 0.97
0.94 0.98 1.01 1.00 1.01 1.00 1.02 1.01
0.96 1.02 1.03 1.00 0.98 1.01 0.99 0.98
0.82 0.86 0.92 1.00 0.97 0.95 0.96 0.96
0.90 0.92 0.93 1.00 0.96 0.97 0.94 0.98
0.87 0.89 0.89 1.00 0.96 0.98 0.96 1.00
0.93 0.95 0.96 1.00 0.97 0.97 0.98 0.99
0.78 0.82 0.84 1.00 0.98 0.99 0.98 0.98
0.83 0.85 0.88 1.00 0.96 0.97 0.96 0.97
0.87 0.86 0.89 1.00 0.96 0.98 0.96 0.96
0.93 0.95 0.96 1.00 0.97 0.97 0.98 0.96
0.99 1.00 1.00 1.00 1.00 1.00 1.02 1.02
0.95 1.02 1.03 1.00 1.00 1.01 0.99 1.00
0.94 0.98 1.01 1.00 1.01 1.00 1.02 1.00
0.96 1.02 1.03 1.00 0.98 1.01 0.99 1.00
0.98 1.00 1.03 1.00 1.01 1.00 1.02 1.00
0.96 1.02 1.05 1.00 1.00 1.02 1,01 0.97
0.82 0.90 0.89 1.00 0.95 0.97 0.94 0.96
0.77 0.88 0.86 1.00 0.94 0.95 0.97 0.94
0.76 0.80 0.81 1.00 0.97 0.98 0.96 0.98
0.78 0.82 0.84 1.00 0.98 0.99 0.98 0.98
0.83 0.85 0.88 1.00 0.96 0.97 0.96 0.97
0.82 0.86 0.87 1.00 0.95 0.98 0.97 0.96
0.85 0.89 0.88 1.00 0.97 0.98 0.99 0.99
0.90 0.92 0.93 1.00 0.96 0.97 0.97 0.98
0.94 0.95 0.96 1.00 0.97 0.97 0.98 0.96
0.99 1.02 1.12 1.00 1.00 1.05 1.02 1.01
1.00 1.02 1.05 1.00 1.00 1.02 1,01 0.97
1.00 1.01 1.03 1.00 1.00 1.02 1,01 0.98
1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
1.01 1.00 1.00 1.00 1.00 1.00 1.02 1.02
1.02 1.02 1.03 1.00 1.00 1.01 0.99 1.00
1.00 1.01 1.08 1.00 1.00 1.05 1.02 1.01
1.03 1.00 1.00 1.00 1.01 1.02 1.04 1.05
1.01 1.02 1.03 1.00 1.00 1.01 0.99 1.00
1.03 1.05 1.08 1.00 1.02 1.00 1.00 1.00
1.00 1.02 1.03 1.00 0.98 1.01 0.99 1.00
1.00 1.00 1.04 1.00 0.99 1.01 1.00 1.00
1.00 1.01 1.07 1.00 1.04 1.02 1.00 1.03
1.00 1.02 1.03 1.00 0.98 1.01 0.99 1.00
1.01 1.00 1.03 1.00 1.01 1.00 1.02 1.00
1.02 1.05 1.10 1.00 1.03 1.00 1.00 1.00
1.00 1.02 1.05 1.00 1.00 1.02 1,01 0.97
1.03 0.98 1.01 1.00 1.01 1.00 1.02 1.01
1.01 1.02 1.03 1.00 0.98 1.01 0.99 0.98
1.03 1.00 1.00 1.00 1.00 1.00 1.02 1.02
1.00 1.02 1.03 1.00 1.00 1.01 0.99 1.00
1.00 1.01 1.08 1.00 1.00 1.05 1.02 1.01
1.00 1.00 1.00 1.00 1.01 1.02 1.04 1.05
1.00 1.02 1.03 1.00 1.00 1.01 0.99 1.00
1.00 1.00 1.00 1.00 1.01 1.00 1.02 1.02
1.00 1.02 1.03 1.00 1.00 1.01 0.99 1.00
1.00 0.98 1.01 1.00 1.01 1.00 1.02 1.00
1.00 1.02 1.03 1.00 0.98 1.01 0.99 1.00
1.00 1.00 1.03 1.00 1.01 1.00 1.02 1.00
1.00 1.05 1.10 1.00 1.03 1.00 1.00 1.00
1.00 1.02 1.05 1.00 1.00 1.02 1,01 0.97
1.00 1.00 1.00 1.00 1.00 1.00 1.02 1.02
1.00 1.02 1.03 1.00 1.00 1.01 0.99 1.00
1.00 0.98 1.01 1.00 1.01 1.00 1.02 1.00
1.00 1.02 1.03 1.00 0.98 1.01 0.99 1.00
1.00 1.00 1.03 1.00 1.01 1.00 1.02 1.00
1.05 1.02 1.05 1.00 1.00 1.02 1,01 0.97
1.01 1.00 1.00 1.00 1.01 1.00 1.02 1.02
1.00 1.02 1.03 1.00 1.00 1.01 0.99 1.00
1.00 0.98 1.01 1.00 1.01 1.00 1.02 1.00
1.00 1.02 1.03 1.00 0.98 1.01 0.99 1.00
1.00 1.00 1.03 1.00 1.01 1.00 1.02 1.00
1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00
1.00 1.00 1.00 1.00 1.00 1.00 1.02 1.02
1.00 1.02 1.03 1.00 1.00 1.01 0.99 1.00
1.00 1.01 1.08 1.00 1.00 1.05 1.02 1.01
1.00 1.00 1.00 1.00 1.01 1.02 1.04 1.05
1.00 1.01 1.08 1.00 1.00 1.05 1.02 1.01
1.00 1.00 1.00 1.00 1.01 1.02 1.04 1.05
1.00 1.02 1.03 1.00 1.00 1.01 0.99 1.00
1.00 1.00 1.00 1.00 1.00 1.00 1.03 1.00
After training of the training samples, the final iteration of the model converges, and the error convergence curve is shown in fig. 4 below, so that the model construction is completed.
After the model of the embodiment is established, relative errors are controlled to be less than 2% according to the measured efficiency value (shown in the following table 2) of the plant desulfurization operation and the desulfurization efficiency prediction value output by the model, for example, as shown in fig. 3, and the model precision meets the application requirement.
TABLE 2 actual measurement data (normalized)
Q (smoke volume) Tin (original smoke temperature) F (limestone slurry flow) Qx (total flow of circulating pump) M (Gypsum slurry density) Eta (desulfurization efficiency) PH (slurry pH value) Tout (Smoke temperature)
0.45 0.72 0.49 0.50 0.95 0.93 0.97 0.92
0.67 0.73 0.71 0.75 0.93 0.95 0.95 0.94
0.81 0.87 0.90 1.00 0.96 0.97 0.98 0.96
1.00 1.01 1.08 1.00 0.99 0.98 0.94 0.98
According to the established optimization model of the desulfurization tower, under the condition that the structure and the reaction mechanism of the absorption tower do not need to be researched, through the judgment of input and output operation parameters with close correlation of the desulfurization tower, the input vectors of a neural network model are determined to be { Q (flue gas flow), Tin (original flue gas temperature), F (limestone slurry flow), Qx (total flow of a circulating pump in transportation) and M (gypsum slurry density) }. The output vector is { η (desulfurization efficiency), PH (slurry PH), Tout (clean flue gas temperature) }. The method comprises the steps of constructing a neural network structure of a single input layer, a single output layer and a single hidden layer, bringing a training sample set covering the complete operation condition of the desulfurizing tower into consideration, finally establishing a desulfurizing tower optimal operation model based on neural network discrimination, and providing guidance suggestions of operation control for the real-time operation process of the desulfurizing tower under different conditions, so that the aims of optimizing the operation parameters of the desulfurizing tower, ensuring stable desulfurizing efficiency and reducing operation consumption are fulfilled.
While the invention has been particularly shown and described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (9)

1. The desulfurizing tower optimization method based on neural network discrimination is characterized by comprising the following steps: the method comprises the following steps of,
step one, acquiring relevant operation parameters of a desulfurizing tower and processing data;
and step two, inputting the processed data serving as an input vector into the neural network model, and obtaining an output vector.
2. The method of claim 1 for desulfurizing tower optimization based on neural network discrimination, characterized by: the relevant operating parameters of the desulfurizing tower comprise the flue gas volume, the original flue gas temperature, the limestone slurry flow, the total flow of the circulating pump in the process of transportation and the gypsum slurry density.
3. The method of claim 1 for desulfurizing tower optimization based on neural network discrimination, characterized by: the output vector of the neural network model comprises desulfurization efficiency, slurry pH value and net flue gas temperature.
4. The method of claim 1 for desulfurizing tower optimization based on neural network discrimination, characterized by: in the first step, the data is normalized, and the normalization processing formula is as follows:
Figure 937139DEST_PATH_IMAGE001
wherein a and b are constants, xmax、xminActual maximum and minimum values for each set of factor variables; x is the number ofiAnd xi ’The values of the factor variables before and after normalization, respectively.
5. The method of claim 1 for desulfurizing tower optimization based on neural network discrimination, characterized by: the neural network model is a BP neural network.
6. The method of claim 1 for desulfurizing tower optimization based on neural network discrimination, characterized by: the structure of the neural network model is a BP neural network structure with a single input layer, a single output layer and a layer 1 hidden layer.
7. The method of claim 1 for desulfurizing tower optimization based on neural network discrimination, characterized by: the number of input layer nodes of the neural network model is 5 corresponding to the number of input vectors, the number of output layer nodes of the neural network model is 3 corresponding to the number of output vectors, and the number of initial hidden layer nodes of the neural network model is 7.
8. The method of claim 1 for desulfurizing tower optimization based on neural network discrimination, characterized by: the output transformation function of the hidden layer neuron of the neural network model adopts a nonlinear sigmoid function, and the function is specifically as follows:
Figure 407435DEST_PATH_IMAGE002
where x is the input to the neuron and y is the output of the neuron.
9. The method of claim 1 for desulfurizing tower optimization based on neural network discrimination, characterized by: the neural network model respectively corrects the weighted value and the threshold value of each network layer connecting line by adopting a first-order gradient method and a logarithmic reduction method until the error is smaller than a preset value.
CN202110386338.5A 2021-04-12 2021-04-12 Desulfurizing tower optimization method based on neural network discrimination Pending CN113094992A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110386338.5A CN113094992A (en) 2021-04-12 2021-04-12 Desulfurizing tower optimization method based on neural network discrimination

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110386338.5A CN113094992A (en) 2021-04-12 2021-04-12 Desulfurizing tower optimization method based on neural network discrimination

Publications (1)

Publication Number Publication Date
CN113094992A true CN113094992A (en) 2021-07-09

Family

ID=76676201

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110386338.5A Pending CN113094992A (en) 2021-04-12 2021-04-12 Desulfurizing tower optimization method based on neural network discrimination

Country Status (1)

Country Link
CN (1) CN113094992A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113723002A (en) * 2021-09-02 2021-11-30 大唐环境产业集团股份有限公司 Method and system for establishing slurry pH value prediction model of desulfurization system under all working conditions

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102693451A (en) * 2012-06-14 2012-09-26 东北电力大学 Method for predicting ammonia process flue gas desulphurization efficiency based on multiple parameters
US20160158701A1 (en) * 2013-01-14 2016-06-09 Babcock & Wilcox Power Generation Group, Inc. Controlling aqcs parameters in a combustion process
CN108509692A (en) * 2018-03-12 2018-09-07 重庆科技学院 A kind of high sulfur content natural gas desulfurization process modeling method based on MiUKFNN algorithms

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102693451A (en) * 2012-06-14 2012-09-26 东北电力大学 Method for predicting ammonia process flue gas desulphurization efficiency based on multiple parameters
US20160158701A1 (en) * 2013-01-14 2016-06-09 Babcock & Wilcox Power Generation Group, Inc. Controlling aqcs parameters in a combustion process
CN108509692A (en) * 2018-03-12 2018-09-07 重庆科技学院 A kind of high sulfur content natural gas desulfurization process modeling method based on MiUKFNN algorithms

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
程换新等: "BP神经网络预测技术在脱硫系统pH值中的应用", 《甘肃科学学报》 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113723002A (en) * 2021-09-02 2021-11-30 大唐环境产业集团股份有限公司 Method and system for establishing slurry pH value prediction model of desulfurization system under all working conditions

Similar Documents

Publication Publication Date Title
CN102693451B (en) Method for predicting ammonia process flue gas desulphurization efficiency based on multiple parameters
CN112163380B (en) Neural network hearth oxygen concentration prediction system and method based on numerical simulation
CN106777866B (en) Energy-saving consumption-reducing-oriented modeling and optimizing method for high-sulfur-content natural gas purification process
CN113433911B (en) Accurate control system and method for ammonia spraying of denitration device based on accurate concentration prediction
CN104656441B (en) Gas purifying process modeling optimization method based on Unscented kalman neutral net
US20090248175A1 (en) Plant control system and thermal power generation plant control system
CN113627071B (en) Coal-fired boiler NO based on whale algorithm optimization long-short-term memory network X Soft measurement method
CN111460726B (en) Optimization method for ammonia escape of coal slime fluidized bed boiler denitration system
CN112016754A (en) Power station boiler exhaust gas temperature advanced prediction system and method based on neural network
CN117031950A (en) Modeling method and device for deep peak-shaving thermal power generating unit control system
Dı´ az et al. Adaptive neurocontrol of heat exchangers
CN112742187A (en) Method and device for controlling pH value in desulfurization system
CN115344019A (en) Natural gas metering flow adjusting process based on composite intelligent algorithm
CN113094992A (en) Desulfurizing tower optimization method based on neural network discrimination
CN111401652A (en) Boiler optimization method and system based on CO online detection
CN105447256B (en) A kind of enhancing excitation emulation genetic optimization method
Xu et al. Design of type-2 fuzzy fractional-order proportional-integral-derivative controller and multi-objective parameter optimization under load reduction condition of the pumped storage unit
CN107621781A (en) Oil field heating furnace constant temperature heats and thermal efficiency Self-Optimizing Control Approach and system
CN107247994B (en) Fuzzy modeling method for desulfurization efficiency of tray tower desulfurization device
Fu et al. Prediction model of desulfurization efficiency of coal-fired power plants based on long short-term memory neural network
Hanafi et al. Heat exchanger's shell and tube modeling for intelligent control design
CN116755409A (en) Coal-fired power generation system coordination control method based on value distribution DDPG algorithm
CN112231978B (en) Boiler flue gas acid dew point testing method based on artificial neural network
CN115326237A (en) Wall temperature measuring method and system considering influence of boiler superheater oxide skin
CN113779858B (en) Combustion optimization method and system, storage medium and electronic equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20210709

RJ01 Rejection of invention patent application after publication