A pollutant control quantity distribution and control system and method for multiple furnaces operating in parallel

Abstract

The present invention relates to a pollutant control quantity allocation and control system and method for multiple furnaces in parallel operation, the system includes: a data acquisition module for collecting flue gas temperature, oxygen content and pollutant inlet concentration; a pollutant concentration prediction model establishment module for establishing a pollutant concentration prediction model for each furnace outlet; a pollutant concentration prediction module for predicting pollutant concentrations to obtain pollutant concentration prediction values; a total control quantity calculation module caused by the change value of each pollutant concentration, for determining the total control quantity caused by the change value of each pollutant concentration of each furnace based on the predicted value of the pollutant concentration, the target value of the pollutant concentration and the real-time feedback value of the pollutant concentration by using a genetic-neural network algorithm; a distribution control quantity calculation module for each furnace, for determining the distribution control quantity of each furnace based on the total control quantity caused by the change value of each pollutant concentration. The above method in the present invention can accurately distribute the control quantity of pollutants in each furnace.

Description

Technical Field

[0001] The invention relates to the field of boiler control, and in particular to a pollutant control quantity distribution and control system and method for multiple boilers operating in parallel. Background Art

[0002] Due to production needs, multiple furnaces (power plant boilers, lime kilns, etc.) are often built. In order to reduce the cost of flue gas treatment, the flue gases from multiple furnaces are often converged and unified into the flue gas treatment system, and then discharged from the chimney after the flue gas treatment. Because the monitoring platform is set between the chimney and the treatment system, the total exhaust of the flue gas is often collected. When the operating conditions of one or more furnaces change, it is difficult to accurately allocate the control amount of pollutants in each furnace through monitoring data. Summary of the invention

[0003] The object of the present invention is to provide a pollutant control quantity allocation and control system and method for multiple furnaces operating in parallel, so as to accurately allocate the control quantity of pollutants in each furnace.

[0004] To achieve the above object, the present invention provides the following solutions:

[0005] A pollutant control quantity distribution and control system for multiple furnaces operating in parallel, the control system comprising:

[0006] Data acquisition module, used to collect historical flue gas temperature, oxygen content and pollutant inlet concentration;

[0007] A pollutant concentration prediction model establishment module is used to establish a pollutant concentration prediction model for each furnace outlet; the pollutant concentration prediction model is a neural network model;

[0008] A pollutant concentration prediction module, used to predict the pollutant concentration using the pollutant concentration prediction model to obtain a pollutant concentration prediction value;

[0009] A total control amount calculation module caused by the change value of each pollutant concentration is used to determine the total control amount caused by the change value of each pollutant concentration of each furnace by using a genetic-neural network algorithm based on the predicted value of the pollutant concentration, the target value of the pollutant concentration and the real-time feedback value of the pollutant concentration; the target value of the pollutant concentration is the pollutant emission limit required by the user, and the real-time feedback value is the online measured value;

[0010] The distribution control amount calculation module of each furnace is used to determine the distribution control amount of each furnace based on the total control amount caused by the change value of the concentration of each pollutant.

[0011] Optionally, the system further includes:

[0012] A correction module is used to correct the predicted value of the pollutant concentration.

[0013] Optionally, the correction module specifically includes:

[0014] A pollutant concentration actual value acquisition unit, used to acquire the actual value of the pollutant concentration at time k;

[0015] A pollutant concentration prediction value acquisition unit, used to acquire the pollutant concentration prediction value at time k-1;

[0016] A deviation calculation unit, used to calculate the deviation between the actual value of the pollutant concentration at the time k and the predicted value of the pollutant concentration at the time k-1;

[0017] The correction unit is used to use the deviation as an estimated value of the prediction error at time k, and use the estimated value as a feedback correction signal to compensate the pollutant concentration prediction model at time k to obtain a corrected pollutant concentration.

[0018] Optionally, the distribution control amount calculation module of each furnace specifically adopts the following formula:

[0019] , where i=1,2,3....n, Ui represents the control quantity allocated to the i-th boiler, Qi represents the flue gas volume of the i-th boiler, the unit is m3 / h, It represents the change of pollutants in the i-th furnace, in mg / m3. Represents the total control amount caused by changes in pollutants.

[0020] Based on the above system in the present invention, the present invention further provides a method for controlling the distribution of pollutant control amounts for multiple furnaces operating in parallel, the method comprising the following steps:

[0021] Collect historical flue gas temperature, oxygen content and pollutant inlet concentration;

[0022] Establishing a pollutant concentration prediction model for each furnace outlet; the pollutant concentration prediction model is a neural network model;

[0023] Using the pollutant concentration prediction model to predict the pollutant concentration to obtain a pollutant concentration prediction value;

[0024] Based on the predicted value of the pollutant concentration, the target value of the pollutant concentration and the real-time feedback value of the pollutant concentration, a genetic-neural network algorithm is used to determine the total control amount caused by the change value of the concentration of each pollutant in each furnace; the target value of the pollutant concentration is the pollutant emission limit required by the user, and the real-time feedback value is the online measured value;

[0025] The distributed control amount of each furnace is determined based on the total control amount caused by the change value of the concentration of each pollutant.

[0026] Optionally, before using the pollutant concentration prediction model to predict the pollutant concentration and obtaining the predicted pollutant concentration value, the method further includes:

[0027] The predicted value of the pollutant concentration is corrected.

[0028] Optionally, correcting the predicted value of the pollutant concentration specifically includes the following steps:

[0029] Get the actual value of pollutant concentration at time k;

[0030] Get the predicted value of pollutant concentration at time k-1;

[0031] Calculate the deviation between the actual value of the pollutant concentration at the time k and the predicted value of the pollutant concentration at the time k-1;

[0032] The deviation is used as an estimated value of the prediction error at time k, and the estimated value is used as a feedback correction signal to compensate the pollutant concentration prediction model at time k to obtain a corrected pollutant concentration.

[0033] Optionally, the total control amount caused by the change in the concentration of each pollutant is used to determine the distribution control amount of each furnace using the following formula:

[0034] , where i=1,2,3....n, Ui represents the control quantity allocated to the i-th boiler, Qi represents the flue gas volume of the i-th boiler, the unit is m3 / h, It represents the change of pollutants in the i-th furnace, in mg / m3. Represents the total control amount caused by changes in pollutants.

[0035] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0036] The above-mentioned system and method in the present invention collects flue gas temperature, oxygen content and pollutant inlet concentration by setting a data acquisition module; establishes a pollutant concentration prediction model for each furnace outlet by setting a pollutant concentration prediction model; predicts the pollutant concentration by using the pollutant concentration prediction model by setting the pollutant concentration prediction module to obtain a pollutant concentration prediction value; and calculates the total control amount caused by the change value of each pollutant concentration by setting a total control amount calculation module for determining the total control amount caused by the change value of each pollutant concentration of each furnace based on the predicted value of the pollutant concentration, the target value of the pollutant concentration and the real-time feedback value of the pollutant concentration by using a genetic-neural network algorithm. ; By setting up an allocated control quantity calculation module for each furnace, it is used to determine the allocated control quantity of each furnace based on the total control quantity caused by the change value of the concentration of each pollutant. The above method and system can quantitatively calculate the change of the control quantity of each furnace due to the operating condition fluctuation, and independently allocate the control quantity of each furnace. The genetic-neural network combined model is used to more accurately analyze and calculate the total control quantity change caused by the load change of the furnace. When the operating condition of each furnace changes, the control quantity can be accurately allocated, which greatly saves the consumption of the control quantity, thereby reducing the operation and investment costs; the present invention can solve the problem that multiple furnaces share the same pollutant monitoring equipment, which makes it difficult to effectively control and adjust them separately. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0038] Figure 1 This is a flowchart of intelligent allocation of multiple furnaces according to an embodiment of the present invention;

[0039] Figure 2 This is a pollutant and control amount prediction diagram of an embodiment of the present invention;

[0040] Figure 3 The measured value of the outlet pollutants in the embodiment of the present invention;

[0041] Figure 4 This is a flow chart of a genetic-neural network algorithm according to an embodiment of the present invention;

[0042] Figure 5 Schematic diagram of the pollutant concentration prediction model structure according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] The object of the present invention is to provide a pollutant control quantity allocation and control system and method for multiple furnaces operating in parallel, so as to accurately allocate the control quantity of pollutants in each furnace.

[0045] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0046] The above-mentioned system and method in the present invention collects data based on flue gas parameters such as flue gas temperature, oxygen content, pollutant inlet concentration, and equipment parameters, and uses online monitoring of outlet pollutant concentration for feedback correction to establish a pollutant concentration prediction model for each furnace outlet, and obtains the pollutant change concentration. The pollutant concentration prediction value, pollutant concentration target value, and pollutant real-time feedback value are used as parameters, and the total control amount caused by the change value of the concentration of each pollutant (NOx, SO2, dust, etc.) of each furnace is analyzed and calculated through the genetic-neural network algorithm combined model. The contribution of each furnace is determined by the proportion of the change in pollutant concentration of each furnace to the change in the total pollution concentration. According to the control amount distribution calculation formula, the distribution control amount of each furnace is obtained for accurate distribution.

[0047] Combination Figure 1-Figure 3 The system of the present invention includes: a data acquisition module, a pollutant concentration prediction model establishment module, a pollutant concentration prediction module, a total control quantity calculation module caused by the change value of each pollutant concentration, and a distribution control quantity calculation module for each furnace.

[0048] Among them, the data acquisition module is used to collect flue gas temperature, oxygen content and pollutant inlet concentration.

[0049] The pollutant concentration prediction model establishment module is used to establish the pollutant concentration prediction model for each furnace outlet.

[0050] The pollutant concentration prediction module is used to predict the pollutant concentration using the pollutant concentration prediction model to obtain a pollutant concentration prediction value.

[0051] The total control amount calculation module caused by the change value of each pollutant concentration is used to determine the total control amount caused by the change value of each pollutant concentration of each furnace using a genetic-neural network algorithm based on the predicted value of the pollutant concentration, the target value of the pollutant concentration and the real-time feedback value of the pollutant concentration.

[0052] The distribution control amount calculation module for each furnace is used to determine the distribution control amount for each furnace based on the total control amount caused by the change value of the concentration of each pollutant.

[0053] The distribution control quantity calculation module of each furnace specifically adopts the following formula:

[0054] , where i=1,2,3....n, Ui represents the control quantity allocated to the i-th boiler, Qi represents the flue gas volume of the i-th boiler, the unit is m3 / h, It represents the change of pollutants in the i-th furnace, in mg / m3. Represents the total control amount caused by changes in pollutants.

[0055] In order to improve the measurement accuracy, the above system in the present invention also includes:

[0056] A correction module is used to correct the predicted value of the pollutant concentration.

[0057] Taking the pollutant control system as the research object, the pollutant map m is the target value of the outlet pollutant concentration, y is the predicted value of the outlet pollutant concentration, and y c is the predicted value of the pollutant concentration at the outlet of the prediction model, y f is the predicted value of the outlet pollutant concentration after correction. By running the algorithm, we can get the outlet pollutant concentration y(k-1) and the predicted model outlet pollutant concentration y c (k). Correspondingly, we can also get y(k) and y c (k+1). Compare the actual output y(k) at time k with the model output y at time k-1 c The deviation between (k) and time is regarded as the estimated value of the prediction error at time k, and is used as a feedback correction signal to compensate the prediction model output y at time k c (k+1), that is, the predicted value after feedback correction is:

[0058] y f (k+1)=y f (k+1)+y(k)-y f (k)

[0059] The feedback correction link takes into account the model prediction error at the previous moment, which improves the model prediction accuracy to a certain extent. f (k+1) and y m (k+1) is input into the objective function and continuous rolling optimization is performed to obtain a more accurate prediction of the outlet pollutant concentration.

[0060] Specifically, the correction module mainly includes:

[0061] A pollutant concentration actual value acquisition unit, used to acquire the actual value of the pollutant concentration at time k;

[0062] A pollutant concentration prediction value acquisition unit, used to acquire the pollutant concentration prediction value at time k-1;

[0063] A deviation calculation unit, used to calculate the deviation between the actual value of the pollutant concentration at the time k and the predicted value of the pollutant concentration at the time k-1;

[0064] The correction unit is used to use the deviation as an estimated value of the prediction error at time k, and use the estimated value as a feedback correction signal to compensate the pollutant concentration prediction model at time k to obtain a corrected pollutant concentration.

[0065] Based on the above system in the present invention, the present invention further provides a method for allocating and regulating pollutant control amounts for multiple furnaces operating in parallel, the method comprising:

[0066] Step 101: Collect historical flue gas temperature, oxygen content and pollutant inlet concentration. The historical flue gas temperature, oxygen content and pollutant inlet concentration are used to train the subsequent pollutant concentration prediction model. Where x = (x1, x2, ..., xn), y = f (x1, x2, ..., xn) {(x, y)} x is the input, y is the ideal output corresponding to x}.

[0067] Step 102: Establish a pollutant concentration prediction model for each furnace outlet; the pollutant concentration prediction model is a neural network model. Figure 5 FIG. 1 is a schematic diagram of the pollutant concentration prediction model structure of an embodiment of the present invention. Figure 5 As shown, the pollutant concentration prediction model is a neural network model, which specifically includes: an input layer, a hidden layer and an output layer.

[0068] Step 103: Use the pollutant concentration prediction model to predict the pollutant concentration to obtain a predicted pollutant concentration value.

[0069] Step 104: Based on the predicted value of the pollutant concentration, the target value of the pollutant concentration and the real-time feedback value of the pollutant concentration, a genetic-neural network algorithm is used to determine the total control amount caused by the change value of each pollutant concentration of each furnace. The target value of the pollutant concentration is the pollutant emission limit required by the user, and the real-time feedback value is the online measured value.

[0070] The genetic algorithm variable selection is optimized by genetic algorithm. The coding length is designed to be VI, the population size is P, the maximum evolutionary number is set to n, and the chromosomes correspond to the input variables one by one. Each gene can only take two cases, "0" and "1": if it is "1", its corresponding input variable participates in the final optimization; if it is "0", it does not participate in the final optimization. The inverse of the test sample set and the variance is taken as the fitness function of the genetic algorithm. The selection operation adopts the proportional selection operator, the crossover operation adopts the multi-point crossover operator, and the mutation operation adopts the multi-point mutation operator. After continuous iterative evolution of a large amount of data. When the iteration termination condition (maximum evolutionary number n) is met, the optimal variables {u1, u2, u3, ..., ui} are screened out, and {u1, u2, u3, ..., ui} screened out by the genetic algorithm are used as network input variables to construct a neural network model. A large amount of data is used for data mining, and the neural network model is trained by cross-validation method. The cycle training method is used to find the optimal value, so as to achieve the best prediction effect. Specifically, Figure 4 shown.

[0071] Step 105: Determine the distribution control amount of each furnace based on the total control amount caused by the change value of each pollutant concentration.

[0072] The specific formula is as follows:

[0073] , where i=1,2,3....n, Ui represents the control quantity allocated to the i-th boiler, Qi represents the flue gas volume of the i-th boiler, the unit is m3 / h, It represents the change of pollutants in the i-th furnace, in mg / m3. Represents the total control amount caused by changes in pollutants.

[0074] The above system and method in the present invention have the following effects:

[0075] By adopting the above method in the present invention, the change of the control quantity of each furnace caused by the fluctuation of the working condition can be quantitatively calculated, and the control quantity of each furnace can be autonomously allocated.

[0076] The genetic-neural network combined model is used to more accurately analyze and calculate the changes in the total control quantity caused by the load changes of the furnace.

[0077] When the operating conditions of each furnace change, the control quantity can be accurately allocated, which greatly saves the consumption of control quantity and thus reduces operating and investment costs.

[0078] The invention can solve the problem that multiple furnaces share the same pollutant monitoring equipment, which makes it difficult to effectively control and adjust them individually.

[0079] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0080] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A pollutant control quantity distribution and control system for multiple furnaces operating in parallel, characterized in that: The control system comprises: Data acquisition module, used to collect historical flue gas temperature, oxygen content and pollutant inlet concentration; A pollutant concentration prediction model establishment module is used to establish a pollutant concentration prediction model for each furnace outlet; the pollutant concentration prediction model is a neural network model; A pollutant concentration prediction module, used to predict the pollutant concentration using the pollutant concentration prediction model to obtain a pollutant concentration prediction value; A correction module is used to correct the predicted value of the pollutant concentration; the correction module specifically includes: A pollutant concentration actual value acquisition unit, used to acquire the actual value of the pollutant concentration at time k; A pollutant concentration prediction value acquisition unit, used to acquire the pollutant concentration prediction value at time k-1; A deviation calculation unit, used to calculate the deviation between the actual value of the pollutant concentration at the time k and the predicted value of the pollutant concentration at the time k-1; A correction unit, used to use the deviation as an estimated value of the prediction error at time k, and use the estimated value as a feedback correction signal to compensate the pollutant concentration prediction model at time k to obtain a corrected pollutant concentration; A total control amount calculation module caused by the change value of each pollutant concentration is used to determine the total control amount caused by the change value of each pollutant concentration of each furnace by using a genetic-neural network algorithm based on the predicted value of the pollutant concentration, the target value of the pollutant concentration and the real-time feedback value of the pollutant concentration; the target value of the pollutant concentration is the pollutant emission limit required by the user, and the real-time feedback value is the online measured value; The distribution control amount calculation module of each furnace is used to determine the distribution control amount of each furnace based on the total control amount caused by the change value of each pollutant concentration, and specifically adopts the following formula: , where, i=1,2,3....n, U i分 represents the control quantity allocated to the ith boiler, Q i Indicates the flue gas volume of the i-th furnace, in m 3 / h, Indicates the change in pollutants in the i-th furnace, in mg / m 3 , Represents the total control amount caused by changes in pollutants.

2. A method for allocating and regulating pollutant control amounts for multiple furnaces operating in parallel, characterized in that: The control method comprises the following steps: Collect historical flue gas temperature, oxygen content and pollutant inlet concentration; Establishing a pollutant concentration prediction model for each furnace outlet; the pollutant concentration prediction model is a neural network model; Using the pollutant concentration prediction model to predict the pollutant concentration to obtain a pollutant concentration prediction value; Correcting the predicted value of the pollutant concentration includes the following steps: Get the actual value of pollutant concentration at time k; Get the predicted value of pollutant concentration at time k-1; Calculate the deviation between the actual value of the pollutant concentration at the time k and the predicted value of the pollutant concentration at the time k-1; The deviation is used as an estimated value of the prediction error at time k, and the estimated value is used as a feedback correction signal to compensate the pollutant concentration prediction model at time k to obtain a corrected pollutant concentration; Based on the predicted value of the pollutant concentration, the target value of the pollutant concentration and the real-time feedback value of the pollutant concentration, a genetic-neural network algorithm is used to determine the total control amount caused by the change value of the concentration of each pollutant in each furnace; the target value of the pollutant concentration is the pollutant emission limit required by the user, and the real-time feedback value is the online measured value; The distribution control amount of each furnace is determined based on the total control amount caused by the change value of each pollutant concentration, and the following formula is specifically used: , where, i=1,2,3....n, U i分 represents the control quantity allocated to the i-th boiler, Q i Indicates the flue gas volume of the i-th furnace, in m 3 / h, Indicates the change in pollutants in the i-th furnace, in mg / m 3 , Represents the total control amount caused by changes in pollutants.

Images