A flue gas desulfurization automatic control method and system

Abstract

The present application relates to the technical field of flue gas desulfurization, and specifically to a flue gas desulfurization automation control method and system, the method comprising: collecting the SO2 concentration of flue gas at each moment, and the slurry pH value after the flue gas reacts with limestone slurry at each moment; obtaining the candidate PH mutation point corresponding to each concentration mutation point; determining the matching degree of each candidate PH mutation point corresponding to each concentration mutation point, determining the initial PH mutation point corresponding to each concentration mutation point; determining the optimized matching degree of each candidate PH mutation point corresponding to each concentration mutation point, obtaining the corresponding mutation point pair; determining the inertia degree of each corresponding mutation point pair; obtaining the abnormal coefficient of SO2 concentration at each moment, using a time series model to obtain the predicted slurry pH value, and adjusting the slurry flow rate. Thereby improving the accuracy of the automatic control of flue gas desulfurization, and further improving the desulfurization efficiency of the flue gas desulfurization system.

Description

Technical Field

[0001] The present application relates to the technical field of flue gas desulfurization, and in particular to a flue gas desulfurization automation control method and system. Background Art

[0002] Environmental pollution is one of the most pressing challenges we are facing. Sulfur dioxide is one of the air pollutants, and its emission needs to be restricted in industrial production. Therefore, exploring desulfurization control that can achieve ultra-low emissions and low energy consumption is an important environmental protection topic.

[0003] At present, limestone-gypsum wet method is usually used to complete flue gas desulfurization. In the flue gas desulfurization system, the pH value of the desulfurization liquid and the SO2 concentration of the flue gas are the key factors that directly affect the desulfurization efficiency of the desulfurization system. The pH value of the desulfurization liquid and the SO2 concentration of the flue gas need to be monitored. During the monitoring process, the SO2 concentration in the flue gas will fluctuate due to the influence of the combustion efficiency of the coal, etc. If the pH value of the desulfurization liquid is not adjusted in real time, the flue gas desulfurization efficiency will decrease. Therefore, it is necessary to control the flow of the desulfurization liquid. However, when the SO2 concentration of the flue gas at the inlet changes, the pH value of the slurry changes with hysteresis, resulting in the prediction of the pH value of the slurry by the SO2 concentration and the pH value of the desulfurization liquid at each moment. The accuracy of the model prediction is low, thereby reducing the control accuracy of the slurry flow, which in turn leads to low desulfurization efficiency of the flue gas desulfurization system. Summary of the invention

[0004] In order to solve the above technical problems, a flue gas desulfurization automation control method and system are provided to solve the existing problems.

[0005] The solution to the technical problem of this application is to provide a flue gas desulfurization automation control method and system, including the following steps:

[0006] In a first aspect, an embodiment of the present application provides a flue gas desulfurization automation control method, the method comprising the following steps:

[0007] Collect the SO2 concentration of flue gas at each moment, and the pH value of the slurry after the flue gas reacts with the limestone slurry at each moment;

[0008] According to the change trends of the slurry pH value and SO2 concentration at each moment, the candidate pH mutation points corresponding to each concentration mutation point are obtained;

[0009] According to the difference and time interval between the SO2 concentration and the slurry pH value in the local neighborhood of each concentration mutation point and its corresponding candidate pH mutation point, the matching degree of each candidate pH mutation point corresponding to each concentration mutation point is determined to determine the initial pH mutation point corresponding to each concentration mutation point;

[0010] According to the difference between the time intervals of different concentration mutation points and the time intervals of their corresponding initial pH mutation points, combined with the matching degree, the optimized matching degree of each candidate pH mutation point corresponding to each concentration mutation point is determined to obtain a corresponding mutation point pair;

[0011] According to the difference in the time interval between the concentration mutation point and its corresponding pH mutation point in different corresponding mutation point pairs, the inertia degree of each corresponding mutation point pair is determined;

[0012] Based on the inertia, the optimal matching degree of the slurry pH value corresponding to the SO2 concentration at each moment is determined, the abnormal coefficient of the SO2 concentration at each moment is obtained, the time series model is used to obtain the predicted slurry pH value, and the slurry flow is adjusted.

[0013] Preferably, the step of obtaining the candidate pH mutation points corresponding to the concentration mutation points includes:

[0014] Analyze the changes in slurry pH value and SO2 concentration at all times to obtain each pH mutation point and concentration mutation point;

[0015] Calculate the difference between each concentration mutation point and the SO2 concentration at the next moment. If the calculated result is greater than or equal to 0, the first difference is assigned a value of 1, otherwise, it is assigned a value of -1;

[0016] Calculate the difference between each pH mutation point and the pH value of the slurry at the next moment. If the calculated result is greater than or equal to 0, the second difference is assigned a value of 1, otherwise, it is assigned a value of -1;

[0017] The pH mutation points at the moments corresponding to all the second differences that are not equal to the first difference of each concentration mutation point are selected as the candidate pH mutation points corresponding to each concentration mutation point.

[0018] Preferably, the step of obtaining each pH mutation point and concentration mutation point comprises:

[0019] The Z-score algorithm is used for the slurry pH value at all times to obtain the Z score of the slurry pH value at each time;

[0020] Calculate the normalized mean of the Z scores of the slurry pH values ​​at all times; record the slurry pH value with the Z score greater than or equal to the mean as the pH mutation point;

[0021] For the SO2 concentration at all times, the same method as the pH mutation point at each time is used to obtain each concentration mutation point.

[0022] Preferably, the step of determining the matching degree of each candidate pH mutation point corresponding to each concentration mutation point to determine the initial pH mutation point corresponding to each concentration mutation point includes:

[0023] Analyze the corresponding relationship between SO2 concentration and slurry pH value in the local neighborhood of each concentration mutation point and its corresponding candidate pH mutation point, as well as the sign difference of Z score, to obtain neighborhood point pairs and neighborhood point pairs with the same sign;

[0024] Calculate the difference in Z scores between SO2 concentration and slurry pH value in each pair of neighboring points with the same sign, and record it as the first difference;

[0025] The time interval between the SO2 concentration and its concentration mutation point in each neighboring point pair with the same sign is taken as the difference weight of each neighboring point pair with the same sign;

[0026] Calculate the ratio of the first difference to the difference weight, and take the sum of the ratios of all neighboring point pairs with the same sign as the matching difference coefficient of each candidate PH mutation point;

[0027] Calculate the reciprocal of the product of the number of different signs of the Z scores of the SO2 concentration and the slurry pH value in all neighborhood point pairs and the matching difference coefficient as the matching degree of each candidate pH mutation point corresponding to each concentration mutation point;

[0028] The selected pH mutation point corresponding to the maximum value of all the matching degrees of each concentration mutation point is used as the initial pH mutation point corresponding to each concentration mutation point.

[0029] Preferably, the obtaining of neighboring point pairs and neighboring point pairs with the same number includes:

[0030] Taking the concentration mutation point at each moment as the center, select the SO2 concentrations at multiple moments in its neighborhood to form the local concentration sequence of the concentration mutation point at each moment;

[0031] Taking each candidate PH mutation point corresponding to the concentration mutation point at each moment as the center, select the slurry pH values ​​at multiple moments in its neighborhood to form a local PH sequence of each candidate PH mutation point corresponding to the concentration mutation point at each moment;

[0032] Elements at the same position in the local concentration sequence and the local PH sequence are recorded as neighborhood point pairs;

[0033] Among all the neighborhood point pairs, the neighborhood point pairs with the same sign of the Z score of SO2 concentration and slurry pH value are selected and recorded as the neighborhood point pairs with the same sign.

[0034] Preferably, the step of determining the optimal matching degree of each candidate pH mutation point corresponding to each concentration mutation point to obtain a corresponding mutation point pair includes:

[0035] Calculate the time interval between each concentration mutation point and the next concentration mutation point, and record it as the first time series distance;

[0036] Calculate the time interval between the initial pH mutation point corresponding to each concentration mutation point and the initial pH mutation point corresponding to the next concentration mutation point, and record it as the second time series distance;

[0037] The difference between the first time series distance and the second time series distance is used as the time series difference of each concentration mutation point;

[0038] The reciprocal of the sum of the time series differences of all concentration mutation points is used as a matching degree optimization factor;

[0039] The normalized result of the product of the matching degree optimization factor and the matching degree of each candidate PH mutation point corresponding to each concentration mutation point is used as the optimized matching degree of each candidate PH mutation point corresponding to each concentration mutation point;

[0040] The selected PH mutation point corresponding to the maximum value of all the optimized matching degrees of each concentration mutation point is taken as the PH mutation point corresponding to each concentration mutation point, and is recorded as a corresponding mutation point pair.

[0041] Preferably, determining the degree of inertia of each corresponding mutation point pair includes:

[0042] The time interval between the concentration mutation point and its corresponding pH mutation point in each corresponding mutation point pair is used as the candidate time length for each corresponding mutation point pair;

[0043] The sum of the differences between each corresponding mutation point pair and all other corresponding mutation point pairs as the relative duration difference of each corresponding mutation point pair;

[0044] The number of all corresponding mutation point pairs that are equal to the candidate duration of each corresponding mutation point pair is calculated; and the ratio between the number and the relative duration difference is used as the inertia degree of each corresponding mutation point pair.

[0045] Preferably, the step of determining the optimal matching degree of the slurry pH value corresponding to the SO2 concentration at each moment and obtaining the abnormal coefficient of the SO2 concentration at each moment includes:

[0046] The selected duration of the corresponding mutation point pair corresponding to the maximum value of the inertia degree is used as the inertia duration;

[0047] The slurry pH value at the moment corresponding to the inertia time interval after each moment is used as the slurry pH value corresponding to the SO2 concentration at each moment;

[0048] For the SO2 concentration at each moment and its corresponding slurry pH value, the same method as the optimization matching degree of each selected pH mutation point is used to obtain the optimization matching degree of the slurry pH value corresponding to the SO2 concentration at each moment;

[0049] The abnormal value of SO2 concentration at each moment is obtained by using an abnormal detection algorithm;

[0050] The ratio of the abnormal value to the optimized matching degree of the slurry pH value corresponding to the SO2 concentration at each moment is used as the abnormal coefficient of the SO2 concentration at each moment.

[0051] Preferably, the slurry flow rate is adjusted, and the calculation formula is: ,in, is the adjusted slurry flow rate at time t, L is the preset initial slurry flow rate, is the Z score of SO2 concentration at each moment, is the Z score of the predicted slurry pH value corresponding to the SO2 concentration at each moment, is the preset control coefficient.

[0052] In the second aspect, an embodiment of the present application also provides a flue gas desulfurization automation control system, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of any one of the above-mentioned flue gas desulfurization automation control methods are implemented.

[0053] This application has at least the following beneficial effects:

[0054] The present application obtains the candidate PH mutation points corresponding to each concentration mutation point according to the changing trends of the slurry PH value and SO2 concentration at each moment, and its beneficial effect is that when the SO2 concentration changes, the corresponding moment that causes the slurry PH value to change is preliminarily screened out; the matching degree of each candidate PH mutation point corresponding to each concentration mutation point is determined to determine the initial PH mutation point corresponding to each concentration mutation point; its beneficial effect is that the similarity of the local changes of the mutation point is taken into account, so as to preliminarily determine the initial PH mutation point corresponding to each concentration mutation point; the optimized matching degree of each candidate PH mutation point corresponding to each concentration mutation point is determined to obtain the corresponding mutation point pair, and its beneficial effect is that the initial PH mutation point is further screened for further analysis and consideration of the time interval differences between different mutation points to find the final corresponding PH mutation point pair and avoid the mutation point pair caused by noise interference; determine each corresponding mutation point The inertia degree of the pair is beneficially analyzed, and the time intervals between different corresponding mutation point pairs are analyzed to determine the inertial characteristics of the changes between the slurry pH value and the SO2 concentration, so as to facilitate the subsequent acquisition of the time interval when the SO2 concentration changes at each moment, and the time interval that will cause the slurry pH value to change; determine the optimal matching degree of the slurry pH value corresponding to the SO2 concentration at each moment, obtain the abnormal coefficient of the SO2 concentration at each moment, and use the time series model to obtain the predicted slurry pH value, and adjust the slurry flow rate. Its beneficial effect is that based on the inertia duration, the matching of the slurry pH corresponding to the SO2 concentration is analyzed, and the abnormality of the SO2 concentration is analyzed to obtain the abnormal coefficient, and the abnormal coefficient is used as the prediction weight of the prediction model to predict the predicted slurry pH corresponding to the SO2 concentration at each moment, thereby improving the prediction accuracy of the model, thereby controlling the slurry flow rate, improving the accuracy of the automatic control of the flue gas desulfurization, and thereby improving the desulfurization efficiency of the flue gas desulfurization system. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The following is a detailed description of an automatic control method for flue gas desulfurization of the present application in conjunction with the accompanying drawings.

[0056] Figure 1 A flow chart of the steps of a flue gas desulfurization automation control method provided in an embodiment of the present application;

[0057] Figure 2 A flowchart of the steps of a method for obtaining the matching degree of each candidate pH mutation point corresponding to each concentration mutation point provided in an embodiment of the present application;

[0058] Figure 3 A flowchart of the steps of a method for obtaining the optimized matching degree of each candidate pH mutation point corresponding to each concentration mutation point provided in an embodiment of the present application. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of this application more clear, the following is a further detailed description of a flue gas desulfurization automation control method and system proposed in this application in combination with the accompanying drawings and implementation examples. It should be understood that the specific embodiments described here are only used to explain this application and are not used to limit this application.

[0060] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0061] See also Figure 1 , which shows a flow chart of a flue gas desulfurization automation control method provided by an embodiment of the present application, the method comprising the following steps:

[0062] Step 1, collecting the SO2 concentration of the flue gas at each moment, and the pH value of the slurry after the flue gas reacts with the limestone slurry at each moment.

[0063] The flue gas desulfurization process is a continuous process. By installing a pH detection element on the conveying pipeline of the circulating slurry circulation pump or at the outlet of the gypsum discharge pump, the pH value of the slurry after the flue gas reacts with the limestone slurry is obtained in real time. Secondly, the SO2 monitoring sensor is installed at the flue gas inlet to collect the SO2 concentration in the flue gas in real time, so that the flue gas enters the absorption tower to react with the limestone slurry. The SO2 concentration of the flue gas at each moment and the pH value of the slurry after the flue gas reacts with the limestone slurry at each moment are obtained.

[0064] At this point, the SO2 concentration of the flue gas at each moment and the pH value of the slurry after the flue gas reacts with the limestone slurry at each moment are obtained.

[0065] Step 2: According to the changing trends of the slurry pH value and SO2 concentration at each moment, the candidate pH mutation points corresponding to each concentration mutation point are obtained.

[0066] In the flue gas desulfurization system, the pH value of the slurry directly affects the desulfurization efficiency of the desulfurization system. Limestone needs to be dissolved before it can effectively react with sulfur dioxide to form gypsum. If the pH value of the slurry is too high, the dissolution efficiency of limestone will decrease, which will lead to a decrease in desulfurization efficiency. In flue gas desulfurization, SO2 needs to react with water to form sulfite ions first, and then react with the dissolved limestone. In this process, if the pH value of the slurry is too low, the reaction efficiency of SO2 reacting with water to form sulfite ions will decrease, which will lead to a decrease in flue gas desulfurization efficiency.

[0067] In the flue gas desulfurization system, the flow rate of fresh limestone slurry flowing into the absorption tower and the change of SO2 concentration in the flue gas may affect the pH value of the slurry. Since the flow rate of limestone slurry is controlled by the valve opening of the slurry valve, it is controllable. However, the SO2 concentration in the flue gas is generated by the combustion of coal, and there are too many factors affecting the quality of coal and the combustion reaction. Therefore, the change of SO2 concentration is uncontrollable.

[0068] Based on the above analysis, the changes in slurry pH value and SO2 concentration are analyzed to obtain the mutation points in the slurry pH value and SO2 concentration change process, specifically:

[0069] The Z-score algorithm is used for the slurry pH value at all times to obtain the Z score of the slurry pH value at each time;

[0070] Calculate the normalized mean of the Z scores of the slurry pH values ​​at all times; record the slurry pH value with the Z score greater than or equal to the mean as the pH mutation point;

[0071] For the SO2 concentration at all times, the same method as the pH mutation point at each time is used to obtain each concentration mutation point;

[0072] Furthermore, since the chemical reaction in the flue gas desulfurization system is usually in a state of equilibrium, when the SO2 concentration in the flue gas suddenly changes, there must be a corresponding mutation point in the slurry pH value. It should be noted that the mutation point here is not a data point with a large change in the local data, but a data point that deviates from the normal value. At the same time, since the pH value is also affected by other factors such as the flow rate of limestone slurry, the concentration mutation point and the pH mutation point are not one-to-one corresponding; due to the increase in SO2 concentration at the flue gas inlet, more SO2 in the flue gas is dissolved in water, causing the pH value of the limestone slurry to decrease, and when the SO2 concentration at the flue gas inlet decreases, it will cause the pH value of the limestone slurry to increase. Therefore, the concentration mutation point and the pH mutation point are further screened, and the moment corresponding to the mutation of the slurry pH value due to the change in the concentration mutation point is initially screened out, specifically:

[0073] Calculate the difference between each concentration mutation point and the SO2 concentration at the next moment. If the calculated result is greater than or equal to 0, the first difference is assigned a value of 1, otherwise, it is assigned a value of -1;

[0074] Calculate the difference between each pH mutation point and the pH value of the slurry at the next moment. If the calculated result is greater than or equal to 0, the second difference is assigned a value of 1, otherwise, it is assigned a value of -1;

[0075] Selecting the pH mutation point at the time corresponding to the second difference value that is not equal to the first difference value of each concentration mutation point as the candidate pH mutation point corresponding to each concentration mutation point;

[0076] It should be noted that if the first difference is not equal to the second difference, it means that the corresponding pH mutation point is the candidate pH mutation point. If the first difference is equal to the second difference, it means that the corresponding slurry mutation pH value is a mutation point caused by interference from other factors rather than a change in SO2 concentration.

[0077] Step 3, according to the difference and time interval between the SO2 concentration and the slurry pH value in the local neighborhood of each concentration mutation point and its corresponding candidate pH mutation point, determine the matching degree of each candidate pH mutation point corresponding to each concentration mutation point to determine the initial pH mutation point corresponding to each concentration mutation point.

[0078] Furthermore, among all the candidate PH mutation points, it is necessary to screen out the PH mutation point corresponding to one of the concentration mutation points. Therefore, the local changes of the PH mutation point and the concentration mutation point are analyzed to construct the mutation matching degree, which is specifically:

[0079] Taking the concentration mutation point at each moment as the center, select the SO2 concentrations at multiple moments in its neighborhood to form the local concentration sequence of the concentration mutation point at each moment;

[0080] Taking each candidate PH mutation point corresponding to the concentration mutation point at each moment as the center, select the slurry pH values ​​at multiple moments in its neighborhood to form a local PH sequence of each candidate PH mutation point corresponding to the concentration mutation point at each moment;

[0081] Preferably, in this embodiment, the SO2 concentration and slurry pH value at 10 moments in its neighborhood are selected as other implementation methods, and the implementer can set them according to actual conditions.

[0082] Elements at the same position in the local concentration sequence and the local PH sequence are recorded as neighborhood point pairs;

[0083] Select the neighboring point pairs with the same sign of the Z score of SO2 concentration and slurry pH value among all the neighboring point pairs, and record them as the neighboring point pairs with the same sign;

[0084] Calculate the difference in Z scores between SO2 concentration and slurry pH value in each pair of neighboring points with the same sign, and record it as the first difference;

[0085] It should be noted that, assuming that a certain concentration mutation point is recorded as concentration mutation point, then A candidate pH mutation point corresponding to the concentration mutation point is recorded as There are PH mutation points to be selected. If The local concentration sequence of the concentration mutation point is , No. The local PH sequence of the candidate PH mutation points is: ,but and is the same position in the sequence, recorded as a pair of neighboring points, and is the same position in the sequence, also recorded as a pair of neighboring points, then and The difference is recorded as the first difference, and The difference is also recorded as the first difference.

[0086] The time interval between the SO2 concentration and its concentration mutation point in each neighboring point pair with the same sign is taken as the difference weight of each neighboring point pair with the same sign;

[0087] Calculate the ratio of the first difference to the difference weight, and take the sum of the ratios of all neighboring point pairs with the same sign as the matching difference coefficient of each candidate PH mutation point;

[0088] Calculate the reciprocal of the product of the number of different signs of the Z scores of the SO2 concentration and the slurry pH value in all neighborhood point pairs and the matching difference coefficient as the matching degree of each candidate pH mutation point corresponding to each concentration mutation point;

[0089] The selected pH mutation point corresponding to the maximum value of all the matching degrees of each concentration mutation point is used as the initial pH mutation point corresponding to each concentration mutation point;

[0090] Preferably, in this embodiment, ,in, For the The concentration mutation point corresponds to the The matching degree of the candidate PH mutation points, For the The Z score of SO2 concentration in the same-sign neighboring point pairs, For the The Z score of the slurry pH value in the same-sign neighboring point pairs, For the The time corresponding to the SO2 concentration in the same-sign neighboring point pairs, For the The corresponding moment of the concentration mutation point is and Respectively The concentration mutation point and The number of all neighboring point pairs with the same sign and the number of all neighboring point pairs of the selected PH mutation point, To preset a value greater than 0 to avoid the denominator being 0, in this embodiment, The value is 0.001. represents the first difference, represents the difference weight, is the first ratio, is the matching difference coefficient, It is the number of different signs of SO2 concentration and slurry pH value in all the neighborhood point pairs.

[0091] Further, the flowchart of the steps of the method for obtaining the matching degree of each candidate pH mutation point corresponding to each concentration mutation point provided in the embodiment of the present application is as follows: Figure 2 shown.

[0092] It should be noted that the larger the first difference is, the greater the difference between the SO2 concentration and the pH value with the same trend change in the neighborhood is, and the change situation in the neighborhood of the concentration mutation point and the candidate pH mutation point is quite different, indicating that the greater the influence of the change in SO2 concentration on the pH value, the smaller the match between the concentration mutation point and the corresponding candidate pH mutation point.

[0093] At this point, the matching degree of each candidate PH mutation point corresponding to each concentration mutation point is obtained, and the initial PH mutation point corresponding to each concentration mutation point is obtained.

[0094] Step 4, according to the difference between the time intervals of different concentration mutation points and their corresponding initial pH mutation points, combined with the matching degree, determine the optimized matching degree of each candidate pH mutation point corresponding to each concentration mutation point to obtain the corresponding mutation point pair.

[0095] The matching degree reflects the similarity of the local changes of the two mutation points. However, since the collection of SO2 concentration and slurry pH value is easily interfered by noise, the corresponding mutation point pairs caused by noise interference cannot be effectively screened out by screening through matching degree. Secondly, the volume of the absorption tower in the flue gas desulfurization system is very large, and the liquid holding volume in the slurry tank is large, which leads to the slurry pH value changing very slowly compared with the change of flue gas load. When the inlet SO2 concentration changes, the slurry pH value usually does not change immediately, that is, the slurry pH value change has inertial characteristics.

[0096] Based on the above analysis, since the change of pH value in the flue gas desulfurization process has an inertial characteristic, that is, the change of slurry pH value usually occurs only after the SO2 concentration changes for a period of time, when the slurry flow rate does not change, the time interval between the two concentration mutation points and the time interval between the corresponding pH mutation points should be close to the same. Based on this feature, the initial pH mutation points are further distributed, and the matching optimization factor is constructed through the time distribution characteristics between the concentration mutation point and the selected pH mutation point.

[0097] First, based on the matching degree, the pH mutation point corresponding to each concentration mutation point is determined, specifically:

[0098] Calculate the time interval between each concentration mutation point and the next concentration mutation point, and record it as the first time series distance;

[0099] Calculate the time interval between the initial pH mutation point corresponding to each concentration mutation point and the initial pH mutation point corresponding to the next concentration mutation point, and record it as the second time series distance;

[0100] The difference between the first time series distance and the second time series distance is used as the time series difference of each concentration mutation point;

[0101] The reciprocal of the sum of the time series differences of all concentration mutation points is used as a matching degree optimization factor;

[0102] The normalized result of the product of the matching degree optimization factor and the matching degree of each candidate PH mutation point corresponding to each concentration mutation point is used as the optimized matching degree of each candidate PH mutation point corresponding to each concentration mutation point;

[0103] The selected pH mutation point corresponding to the maximum value of all the optimized matching degrees of each concentration mutation point is used as the pH mutation point corresponding to each concentration mutation point, and recorded as a corresponding mutation point pair;

[0104] Preferably, in this embodiment, the absolute value of the difference between the first time series distance and the second time series distance is used as the time series difference of each concentration mutation point.

[0105] In this embodiment, the calculation method of the matching optimization factor is: ,in, is the matching optimization factor, For the The concentration mutation point and The time interval between the concentration mutation points is the first temporal distance. For the The initial pH mutation point corresponding to the concentration mutation point is The time interval between the initial pH mutation points corresponding to the concentration mutation points, that is, the second temporal distance, is the number of all concentration mutation points, To preset a value greater than 0 to avoid the denominator being 0, in this embodiment, The value is 0.001.

[0106] Further, the flowchart of the steps of the method for obtaining the optimized matching degree of each candidate pH mutation point corresponding to each concentration mutation point provided in the embodiment of the present application is as follows: Figure 3 shown.

[0107] It should be noted that if the time series difference is larger, it means that the concentration mutation point and its corresponding initial PH mutation point are caused by noise interference and are not a truly corresponding mutation point pair. In this case, the matching degree between the concentration mutation point and the corresponding PH mutation point is lower, and the obtained optimized matching degree is smaller.

[0108] At this point, the optimized matching degree of each candidate pH mutation point corresponding to each concentration mutation point is obtained, and the corresponding mutation point pair is obtained.

[0109] Step 5, according to the difference in the time interval between the concentration mutation point and its corresponding pH mutation point in different corresponding mutation point pairs, determine the inertia degree of each corresponding mutation point pair.

[0110] Furthermore, based on the optimized matching degree, the pH mutation points corresponding to each concentration mutation point are screened out, the inertial characteristics of the changes between each concentration mutation point and its corresponding pH mutation point are analyzed, and the inertial degree is constructed to reflect the inertial characteristics of the pH value change caused by the sudden change in SO2 concentration, specifically:

[0111] The time interval between the concentration mutation point and its corresponding pH mutation point in each corresponding mutation point pair is used as the candidate time length for each corresponding mutation point pair;

[0112] The sum of the differences between each corresponding mutation point pair and all other corresponding mutation point pairs as the relative duration difference of each corresponding mutation point pair;

[0113] Calculate the number of all corresponding mutation point pairs that are equal to the selected duration of each corresponding mutation point pair; and use the ratio between the number and the relative duration difference as the inertia degree of each corresponding mutation point pair;

[0114] Preferably, in this embodiment, the sum of the absolute values ​​of the differences between the candidate durations of each corresponding mutation point pair and all other corresponding mutation point pairs is taken as the relative duration difference of each corresponding mutation point pair.

[0115] In this embodiment, the calculation method of the inertia degree of each corresponding mutation point pair is: ,in, For the The inertia of the corresponding mutation point pairs, For the The number of all corresponding mutation point pairs with equal to-be-selected durations, For the The waiting time for the corresponding mutation point pair is For the The waiting time for the corresponding mutation point pair is To preset a value greater than 0 to avoid the denominator being 0, in this embodiment, The value is 0.001. For the The relative duration difference of the corresponding mutation point pairs, is the number of all corresponding mutation point pairs, that is, the number of all concentration mutation points.

[0116] It should be noted that if the relative time difference is smaller, it means that the time interval between the change of SO2 concentration and slurry pH value in each corresponding mutation point pair is smaller, which means that when the SO2 concentration changes and the slurry pH value changes, the time interval is roughly the same, which has a certain inertial change characteristic; if The larger the value is, the more likely it is that the slurry pH value will change after the corresponding waiting time when the SO2 concentration changes. The larger the obtained inertia degree is, the more likely it is that the inertial change time of the SO2 concentration and the slurry pH value will be the corresponding waiting time.

[0117] At this point, the inertia degree of each corresponding mutation point pair is obtained.

[0118] Step 6, based on the inertia, determine the optimal matching degree of the slurry pH value corresponding to the SO2 concentration at each moment, obtain the abnormal coefficient of the SO2 concentration at each moment, use the time series model to obtain the predicted slurry pH value, and adjust the slurry flow rate.

[0119] Further, based on the inertia, the abnormal coefficient of SO2 concentration at each moment is determined to reflect the possibility of abnormal change in SO2 concentration, specifically:

[0120] The selected duration of the corresponding mutation point pair corresponding to the maximum value of the inertia degree is used as the inertia duration;

[0121] The slurry pH value at the moment corresponding to the inertia time interval after each moment is used as the slurry pH value corresponding to the SO2 concentration at each moment;

[0122] It should be noted that, assuming the inertia duration is , select The SO2 concentration and slurry pH value at the time The pH value of the slurry at the time The slurry pH value corresponding to the SO2 concentration at the moment.

[0123] For the SO2 concentration at each moment and its corresponding slurry pH value, the same method as the optimization matching degree of each selected PH mutation point is adopted to obtain the optimization matching degree between the SO2 concentration at each moment and its corresponding slurry pH value;

[0124] An abnormality detection algorithm is used for the SO2 concentration at all times to obtain the abnormal value of the SO2 concentration at each time;

[0125] The ratio of the abnormal value to the optimized matching degree of the slurry pH value corresponding to the SO2 concentration at each moment is used as the abnormal coefficient of the SO2 concentration at each moment;

[0126] Preferably, in this embodiment, the LOF anomaly detection algorithm is used to obtain the abnormal value of SO2 concentration at each moment, wherein the LOF anomaly detection algorithm is a well-known technology and will not be described in detail here. As other implementation methods, the implementer can adopt other methods of the prior art, such as isolated forest, etc., and this embodiment does not impose any special restrictions on this.

[0127] Further, based on the abnormal coefficient, the predicted pH value of the slurry is obtained, and the slurry flow rate is adjusted based on the predicted pH value, specifically:

[0128] The SO2 concentration and slurry pH value at all times before each moment are trained using a time series model, wherein the abnormal coefficient is used as the prediction weight of the model, and the SO2 concentration at each moment is collected in real time, and the predicted slurry pH value corresponding to the SO2 concentration at each moment is predicted by the time series model;

[0129] Preferably, in this embodiment, an autoregressive difference moving average model (ARIMA model) is used for prediction, and the abnormal coefficient is used as the corresponding weight for moving average in the model, wherein the ARIMA model is a well-known technology and will not be described in detail here. As other implementation methods, implementers may adopt other methods of the prior art, such as the autoregressive moving average model (ARMA model), etc., and this embodiment does not impose any special restrictions on this.

[0130] The Z-score algorithm is used to obtain the Z score of the predicted slurry pH value corresponding to the SO2 concentration at each moment;

[0131] It should be noted that due to the existence of inertia time, the time of predicting the slurry pH value will occur after the SO2 concentration at each moment, that is, the SO2 concentration at each moment is inconsistent with the corresponding predicted slurry pH value time, separated by an inertia time.

[0132] Based on the Z score of the predicted slurry pH value, the slurry flow rate is adjusted as follows: ,in, is the adjusted slurry flow rate at time t, L is the preset initial slurry flow rate. In this embodiment, the preset initial slurry flow rate L is 50m 3 / h, as other implementation methods, the implementer can set it according to the actual situation. is the Z score of SO2 concentration at each moment, is the Z score of the predicted slurry pH value corresponding to the SO2 concentration at each moment, is the preset control coefficient;

[0133] If the Z score of SO2 concentration at each moment is greater than 0, the preset control coefficient Assign a value of 1. If the Z score of the SO2 concentration at each moment is less than or equal to 0, the preset control coefficient Assign a value of -1.

[0134] The flue gas desulfurization system controls the pH value of the slurry in real time according to the slurry flow rate adjusted at each moment, thereby ensuring that the pH value of the slurry in the absorption tower of the desulfurization system is in a stable state, so that the SO2 in the flue gas can fully react with the limestone slurry to absorb the SO2 in the flue gas and desulfurize the flue gas, thereby improving the desulfurization efficiency of the flue gas desulfurization system and realizing automatic control of the slurry flow rate in the flue gas desulfurization system.

[0135] Based on the same inventive concept as the above method, an embodiment of the present application also provides a flue gas desulfurization automation control system, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of any one of the above-mentioned flue gas desulfurization automation control methods are implemented.

[0136] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0137] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0138] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the present application. It should be pointed out that, for those of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application without departing from the content of the technical solution of the present application, shall fall within the protection scope of the technical solution of the present application.

Claims

1. A flue gas desulfurization automation control method, characterized in that: The method comprises the following steps: Collect the SO2 concentration of flue gas at each moment, and the pH value of the slurry after the flue gas reacts with limestone slurry at each moment; According to the change trends of the slurry pH value and SO2 concentration at each moment, the candidate pH mutation points corresponding to each concentration mutation point are obtained; According to the difference and time interval between the SO2 concentration and the slurry pH value in the local neighborhood of each concentration mutation point and its corresponding candidate pH mutation point, the matching degree of each candidate pH mutation point corresponding to each concentration mutation point is determined to determine the initial pH mutation point corresponding to each concentration mutation point; According to the difference between the time intervals of different concentration mutation points and the time intervals of their corresponding initial pH mutation points, combined with the matching degree, the optimized matching degree of each candidate pH mutation point corresponding to each concentration mutation point is determined to obtain a corresponding mutation point pair; According to the difference in the time interval between the concentration mutation point and its corresponding pH mutation point in different corresponding mutation point pairs, the inertia degree of each corresponding mutation point pair is determined; Based on the inertia, the optimal matching degree of the slurry pH value corresponding to the SO2 concentration at each moment is determined, the abnormal coefficient of the SO2 concentration at each moment is obtained, the predicted slurry pH value is obtained by using a time series model, and the slurry flow rate is adjusted; Analyze the changes in slurry pH value and SO2 concentration at all times to obtain each pH mutation point and concentration mutation point; Calculate the difference between each concentration mutation point and the SO2 concentration at the next moment. If the calculated result is greater than or equal to 0, the first difference is assigned a value of 1, otherwise, it is assigned a value of -1; Calculate the difference between each pH mutation point and the pH value of the slurry at the next moment. If the calculated result is greater than or equal to 0, the second difference is assigned a value of 1, otherwise, it is assigned a value of -1; Selecting pH mutation points at the moments corresponding to all the second differences that are not equal to the first difference of each concentration mutation point as the pH mutation points to be selected corresponding to each concentration mutation point; The step of obtaining each pH mutation point and concentration mutation point comprises: The Z-score algorithm was used for the slurry pH values ​​at all times to obtain the Z score of the slurry pH value at each time; Calculate the normalized mean of the Z scores of the slurry pH values ​​at all times; record the slurry pH value with the Z score greater than or equal to the mean as the pH mutation point; For the SO2 concentration at all times, the same method as the pH mutation point at each time is used to obtain each concentration mutation point; The step of determining the matching degree of each candidate pH mutation point corresponding to each concentration mutation point to determine the initial pH mutation point corresponding to each concentration mutation point includes: Analyze the corresponding relationship between SO2 concentration and slurry pH value in the local neighborhood of each concentration mutation point and its corresponding candidate pH mutation point, as well as the sign difference of Z score, to obtain neighborhood point pairs and neighborhood point pairs with the same sign; Calculate the difference in Z scores between SO2 concentration and slurry pH value in each pair of neighboring points with the same sign, and record it as the first difference; The time interval between the SO2 concentration and its concentration mutation point in each neighboring point pair with the same sign is taken as the difference weight of each neighboring point pair with the same sign; Calculating the ratio of the first difference to the difference weight, and taking the sum of the ratios of all neighboring point pairs with the same sign as the matching difference coefficient of each pH mutation point to be selected; Calculate the reciprocal of the product of the number of Z-scores of SO2 concentration and slurry pH value with different signs in all neighboring point pairs and the matching difference coefficient as the matching degree of each candidate pH mutation point corresponding to each concentration mutation point; The selected pH mutation point corresponding to the maximum value of all the matching degrees of each concentration mutation point is used as the initial pH mutation point corresponding to each concentration mutation point; The obtaining of neighboring point pairs and neighboring point pairs with the same number includes: Taking the concentration mutation point at each moment as the center, select the SO2 concentrations at multiple moments in its neighborhood to form the local concentration sequence of the concentration mutation point at each moment; Taking each candidate pH mutation point corresponding to the concentration mutation point at each moment as the center, select the slurry pH values ​​at multiple moments in its neighborhood to form a local pH sequence of each candidate pH mutation point corresponding to the concentration mutation point at each moment; Elements at the same position in the local concentration sequence and the local pH sequence are recorded as neighborhood point pairs; Select the neighboring point pairs with the same sign of the Z score of SO2 concentration and slurry pH value among all the neighboring point pairs, and record them as the neighboring point pairs with the same sign; The step of determining the optimal matching degree of each candidate pH mutation point corresponding to each concentration mutation point to obtain a corresponding mutation point pair includes: Calculate the time interval between each concentration mutation point and the next concentration mutation point, and record it as the first time series distance; Calculate the time interval between the initial pH mutation point corresponding to each concentration mutation point and the initial pH mutation point corresponding to the next concentration mutation point, and record it as the second time series distance; The difference between the first time series distance and the second time series distance is used as the time series difference of each concentration mutation point; The reciprocal of the sum of the time series differences of all concentration mutation points is used as a matching degree optimization factor; The normalized result of the product of the matching degree optimization factor and the matching degree of each candidate pH mutation point corresponding to each concentration mutation point is used as the optimized matching degree of each candidate pH mutation point corresponding to each concentration mutation point; The selected pH mutation point corresponding to the maximum value of all the optimized matching degrees of each concentration mutation point is used as the pH mutation point corresponding to each concentration mutation point, and recorded as a corresponding mutation point pair; Determining the degree of inertia of each corresponding mutation point pair includes: The time interval between the concentration mutation point and its corresponding pH mutation point in each corresponding mutation point pair is used as the candidate time length for each corresponding mutation point pair; The sum of the differences between each corresponding mutation point pair and all other corresponding mutation point pairs as the relative duration difference of each corresponding mutation point pair; Calculate the number of all corresponding mutation point pairs that are equal to the selected duration of each corresponding mutation point pair; and use the ratio between the number and the relative duration difference as the inertia degree of each corresponding mutation point pair; The step of determining the optimal matching degree of the slurry pH value corresponding to the SO2 concentration at each moment and obtaining the abnormal coefficient of the SO2 concentration at each moment includes: The selected duration of the corresponding mutation point pair corresponding to the maximum value of the inertia degree is used as the inertia duration; The slurry pH value at the moment corresponding to the inertia time interval after each moment is used as the slurry pH value corresponding to the SO2 concentration at each moment; For the SO2 concentration at each moment and its corresponding slurry pH value, the same method as the optimization matching degree of each selected pH mutation point is used to obtain the optimization matching degree of the slurry pH value corresponding to the SO2 concentration at each moment; The abnormal value of SO2 concentration at each moment is obtained by using an abnormal detection algorithm; The ratio of the abnormal value to the optimized matching degree of the slurry pH value corresponding to the SO2 concentration at each moment is used as the abnormal coefficient of the SO2 concentration at each moment; The slurry flow rate is adjusted, and the calculation formula is: ,in, is the adjusted slurry flow rate at time t, L is the preset initial slurry flow rate, is the Z score of SO2 concentration at each moment, is the Z score of the predicted slurry pH value corresponding to the SO2 concentration at each moment, is the preset control coefficient; The expression of the matching degree of each candidate pH mutation point is: ,in, For the The concentration mutation point corresponds to the The matching degree of the candidate pH mutation points, For the The Z score of SO2 concentration in the same-sign neighboring point pairs, For the The Z score of the slurry pH value in the same-sign neighboring point pairs, For the The time corresponding to the SO2 concentration in the same-sign neighboring point pairs, For the The corresponding moment of the concentration mutation point is and Respectively The concentration mutation point and The number of all neighboring point pairs with the same sign and the number of all neighboring point pairs of the selected pH mutation point, The value is preset to be greater than 0; The expression of the matching optimization factor is: ,in, is the matching optimization factor, For the The concentration mutation point and The time interval between the concentration mutation points is the first temporal distance. For the The initial pH mutation point corresponding to the concentration mutation point is The time interval between the initial pH mutation points corresponding to the concentration mutation points, that is, the second temporal distance, is the number of all concentration mutation points; The expression of the inertia degree of each corresponding mutation point pair is: ,in, For the The inertia of the corresponding mutation point pairs, For the The number of all corresponding mutation point pairs with equal to-be-selected durations, For the The waiting time for the corresponding mutation point pair is For the The waiting time for the corresponding mutation point pairs.

2. A flue gas desulfurization automation control system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the flue gas desulfurization automation control method as claimed in claim 1 are implemented.

Images