Automatic control method and system of SNCR system in waste power plant based on machine learning algorithm

Abstract

The present invention relates to a garbage power plant based on machine learning algorithm SNCR System automatic control method and system, wherein waste power plant based on machine learning algorithm SNCR The system automatic control method includes: obtaining the current flue gas outlet of the waste power plant NO X emission value; obtain the current opening of the ammonia regulating valve and ammonia flow rate of the waste power plant; set the main PID The module's set value and the current flue gas outlet NO X The emission value is obtained from the main PID The module’s tracking value; obtains the main PID The output upper and lower limits of the module are set according to the master PID Module setpoints and current flue gas outlets NO X The emission value is corrected for deviation; PID The module output value is used as a PID The module setting value is obtained according to the current ammonia flow rate. PID The module's tracking value; PID The output value of the module adjusts the opening of the ammonia regulating valve, achieving the emission of nitrogen oxides in compliance with the standards, reducing ammonia escape, improving the level of automation and reducing SNCR Use the production cost of the unit.

Description

Technical Field

[0001] The present invention belongs to the technical field of waste incineration tail gas treatment, and in particular relates to an automatic control method and system for an SNCR system of a waste power plant based on a machine learning algorithm. Background Art

[0002] In view of the current situation that the SNCR system has a low automatic control investment rate, lags in tracking the outlet nitrogen oxides, and cannot effectively control ammonia slip, the SNCR control system is optimized.

[0003] Since the SNCR process has a certain reaction time after adding the reducing agent and is a lagging system, and the signal measurement is also relatively lagging relative to the changes in the actual operating conditions, the currently observed operating conditions are not the current, real-time operating conditions, but the operating conditions at a certain moment before. It has the characteristics of strong coupling, nonlinearity, variability, and large lag. Through the changes in signals such as nitrogen oxide content and flue gas oxygen content, the machine learning algorithm of exponential weighted average and deviation correction is adopted to overcome the problems of control nonlinearity, large dead zone and insensitive response in the adjustment process, and simulate manual operation to accurately control the amount of reducing agent injection under various operating conditions.

[0004] Therefore, based on the above technical problems, it is necessary to design a new automatic control method and system for the SNCR system of a waste power plant based on machine learning algorithm. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for automatic control of a SNCR system in a waste power plant based on a machine learning algorithm.

[0006] In order to solve the above technical problems, the present invention provides an automatic control method of the SNCR system of a waste power plant based on a machine learning algorithm, comprising:

[0007] Get the current flue gas outlet NO of the waste power plant X Emission value;

[0008] Obtain the current opening degree of the ammonia regulating valve and ammonia flow rate of the waste power plant;

[0009] Set the set value of the main PID module and adjust the current flue gas outlet NO X The emission value obtains the tracking value of the main PID module;

[0010] The output upper and lower limits of the main PID module are obtained according to the current opening of the ammonia regulating valve and the ammonia flow rate, and the output upper and lower limits of the main PID module are obtained according to the set value of the main PID module and the current flue gas outlet NO X To make deviation correction for the emission values;

[0011] The output value of the main PID module is used as the setting value of the auxiliary PID module, and the tracking value of the auxiliary PID module is obtained according to the current ammonia water flow rate;

[0012] The opening of the ammonia regulating valve is adjusted according to the output value of the auxiliary PID module.

[0013] Further, the setting value of the main PID module is set, and the current flue gas outlet NO X The method of obtaining the tracking value of the main PID module includes:

[0014] The setting value of the main PID module is the flue gas outlet NO set according to relevant standards. X Emission value;

[0015] The tracking value of the main PID module is the current flue gas outlet NO X The emission value is the output value of the first-order inertia algorithm.

[0016] Further, the current flue gas outlet NO X The method for obtaining the output value of the emission value through the first-order inertia algorithm includes:

[0017] The filtering time is t f , the calculation period is t c , that is, Δt=t c ; Δt is the time difference of the calculation cycle

[0018] Low-pass filtered first-order inertial link transfer function

[0019] but y(s)·s·t f +y(s)=x(s);

[0020] Through the reverse pull transformation, we can obtain:

[0021]

[0022] Obtained by the forward Euler method:

[0023] dt is the sampling period of the discrete system, y k is the output of the first-order inertia link in the kth sampling period; then

[0024]

[0025] therefore:

[0026] Current flue gas outlet NO X The output value of the emission value after the first-order inertia algorithm is:

[0027]

[0028] Among them, Y(K) 1 is the output value of the current algorithm, that is, the current flue gas outlet NO X The emission value is the output value of the first-order inertia algorithm; Y(K-1) 1 is the output value of the previous period; KG is the first-order inertia time constant; TC is the first-order inertia proportional gain; TS is the filter time; NO X (K) is the current flue gas outlet NO X emission value.

[0029] Further, the output upper and lower limits of the main PID module are obtained according to the current opening of the ammonia regulating valve and the ammonia flow rate, and the output upper and lower limits of the main PID module are obtained according to the set value of the main PID module and the current flue gas outlet NO X Methods for correcting the deviation of emission values ​​include:

[0030] F i =β*F i-1 +(1-β)*θ i ;

[0031] Where i is the current opening of the ammonia regulating valve; θ i is the actual flow rate of ammonia water when the opening of the ammonia water regulating valve is i; F i is the weighted average value of the ammonia flow rate when the ammonia regulating valve is opened at degree i; β is an adjustable parameter, 0<β<1;

[0032] When β=0.9, 0.9 10 ≈0.35≈1 / e, with 1 / e as the dividing point. When the weight decays below this value, it is ignored.

[0033] For any β, Then the exponentially weighted average is the average of the most recent I openings:

[0034]

[0035] When β = 0.9, then at 100% opening, F 100 =0.9*F 99 +(1-0.9)*θ 100 ;

[0036] The average value of the opening of the 100% ammonia regulating valve is obtained by weighting the opening data of the first 100%, and the weight coefficient conforms to an exponential decay function: when I = 10,

[0037] Correct the deviation: when i is very small, the denominator amplifies the current value; when i is very large, the value of the denominator tends to 1:

[0038]

[0039] Among them, DP is the deviation correction;

[0040] The output upper limit of the main PID module is:

[0041] The output lower limit of the main PID module is:

[0042] According to the current flue gas outlet NO X The emission value is obtained by the output value of the first-order inertia algorithm and the set value of the main PID module to obtain the output value of the main PID module.

[0043] Further, the method of using the output value of the main PID module as the set value of the auxiliary PID module and obtaining the tracking value of the auxiliary PID module according to the current ammonia water flow rate includes:

[0044] The tracking value is the output value of the current ammonia flow rate after the first-order inertia algorithm:

[0045]

[0046] Among them, Y(K) 2 is the output value of the current algorithm, that is, the output value of the current ammonia flow through the first-order inertia algorithm; Y(K-1) 2 is the output value of the previous period; FIT(K) is the current ammonia flow rate;

[0047] The output value of the slave PID module is obtained according to the setting value and tracking value of the slave PID module.

[0048] On the other hand, the present invention also provides an automatic control system for a SNCR system of a waste power plant based on a machine learning algorithm, comprising:

[0049] The detection module is suitable for detecting the current flue gas outlet NO of the waste power plant. X The emission value, the opening of the current ammonia regulating valve and the ammonia flow rate of the waste power plant;

[0050] The algorithm module is suitable for obtaining the output upper and lower limits of the main PID module according to the current opening of the ammonia regulating valve and the ammonia flow rate, and the output of the main PID module according to the current flue gas outlet NO X The tracking value of the main PID module is obtained according to the emission value, and the tracking value of the auxiliary PID module is obtained according to the current ammonia flow rate;

[0051] The main PID module is configured to generate a signal according to the set value of the main PID module and the current flue gas outlet NO. XThe emission value is corrected for deviation, and the output value of the main PID module is obtained according to the tracking value of the main PID module and the set value of the main PID module;

[0052] A sub-PID module, wherein the sub-PID module is adapted to obtain an output value of the sub-PID module according to a set value and a tracking value of the sub-PID module, so as to adjust the opening of the ammonia water regulating valve according to the output value of the sub-PID module.

[0053] Furthermore, the automatic control system of the SNCR system of a waste power plant based on a machine learning algorithm is suitable for adopting the above-mentioned automatic control method of the SNCR system of a waste power plant based on a machine learning algorithm to adjust the opening of the ammonia regulating valve according to the output value of the sub-PID module.

[0054] The beneficial effect of the present invention is that the present invention obtains NO of the current flue gas outlet of the waste power plant X emission value; obtain the current opening of the ammonia regulating valve and the ammonia flow rate of the waste power plant; set the setting value of the main PID module, and X The emission value of the main PID module is obtained; the output upper and lower limits of the main PID module are obtained according to the current opening of the ammonia regulating valve and the ammonia flow rate, and the setting value of the main PID module and the current flue gas outlet NO are obtained. X The emission value is corrected for deviation; the output value of the main PID module is used as the set value of the sub-PID module, and the tracking value of the sub-PID module is obtained according to the current ammonia flow rate; the opening of the ammonia regulating valve is adjusted according to the output value of the sub-PID module, thereby achieving nitrogen oxide emission standards, reducing ammonia escape, improving the level of automation, and reducing the production cost of the SNCR user unit.

[0055] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description and the drawings.

[0056] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0058] Figure 1 It is a flow chart of the automatic control method of the SNCR system of a waste power plant based on a machine learning algorithm involved in the present invention;

[0059] Figure 2 It is a logic diagram of the automatic control method of the SNCR system of the garbage power plant based on the machine learning algorithm involved in the present invention;

[0060] Figure 3 It is a principle block diagram of the automatic control system of the SNCR system of a waste power plant based on a machine learning algorithm involved in the present invention. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are 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.

[0062] Example 1

[0063] Figure 1 It is a flow chart of the automatic control method of the SNCR system of a waste power plant based on a machine learning algorithm involved in the present invention.

[0064] like Figure 1 As shown, this embodiment 1 provides an automatic control method for the SNCR system of a waste power plant based on a machine learning algorithm, including: obtaining the current flue gas outlet NO X emission value; obtain the current opening of the ammonia regulating valve and the ammonia flow rate of the waste power plant; set the setting value of the main PID module, and X The emission value of the main PID module is obtained; the output upper and lower limits of the main PID module are obtained according to the current opening of the ammonia regulating valve and the ammonia flow rate, and the setting value of the main PID module and the current flue gas outlet NO are obtained. X The emission value is corrected for deviation; the output value of the main PID module is used as the setting value of the auxiliary PID module, and the tracking value of the auxiliary PID module is obtained according to the current ammonia flow rate; the opening of the ammonia regulating valve is adjusted according to the output value of the auxiliary PID module, and the ammonia flow rate is adjusted to adjust NO X The emission value is achieved, nitrogen oxide emissions meet the standards, ammonia escape is reduced, the level of automation is improved, and the production cost of SNCR users is reduced; according to different operating conditions, through machine learning algorithms, it can achieve automatic tracking operation control, automatic accumulation, learning and analysis of data, automatic adjustment of ammonia water opening, and automatic selection of injection layer.

[0065] Figure 2 It is a logic diagram of the automatic control method of the SNCR system of a waste power plant based on a machine learning algorithm involved in the present invention.

[0066] like Figure 2 As shown, in this embodiment, the setting value of the main PID module is set, and the current flue gas outlet NO X The method for obtaining the tracking value of the main PID module from the emission value includes: the setting value of the main PID module is the flue gas outlet NO set according to the relevant standards X For example, the flue gas outlet NO X The emission value is used as the set value (SP1) of the main PID module; the tracking value (PV 1 ) is the current flue gas outlet NO X The emission value is the output value of the first-order inertia algorithm (Y 1 ).

[0067] In this embodiment, the current flue gas outlet NO X The method for obtaining the output value of the emission value through the first-order inertia algorithm includes:

[0068] The filtering time is t f , the calculation period is t c , that is, Δt=t c ; Δt is the time difference of the calculation cycle;

[0069] Low-pass filtered first-order inertial link transfer function

[0070] but y(s)·s·t f +y(s)=x(s);

[0071] Through the reverse pull transformation, we can obtain:

[0072] dt is the sampling period of the discrete system, y k is the output of the first-order inertia link in the kth sampling period;

[0073] Obtained by the forward Euler method (when the interval between two adjacent discrete points is small, the first-order difference quotient is used instead of the first-order derivative):

[0074]

[0075] The joint and To obtain:

[0076]

[0077] therefore:

[0078] Current flue gas outlet NO X The output value of the emission value after the first-order inertia algorithm is:

[0079]

[0080] Among them, Y(K) 1 is the output value of the current algorithm, that is, the current flue gas outlet NO X The emission value is the output value of the first-order inertia algorithm; Y(K-1) 1 is the output value of the previous period; KG is the first-order inertia time constant, the default is 1; TC is the first-order inertia proportional gain, the default is 10; TS is the filter time; NO X (K) is the current flue gas outlet NO X emission value.

[0081] In this embodiment, the output upper and lower limits of the main PID module are obtained according to the current opening of the ammonia regulating valve and the ammonia flow rate, and the output upper and lower limits of the main PID module are obtained according to the set value of the main PID module and the current flue gas outlet NO X The method for correcting the deviation of the emission value includes: the opening of the ammonia flow control valve and the ammonia flow rate are used as exponentially weighted averages (Exponentially Weighted Averages) and two parameters of the learning algorithm to control the output upper and lower limits of the main PID module;

[0082] F i =β*F i-1 +(1-β)*θ i ;

[0083] Where i is the current opening of the ammonia regulating valve; θ i is the actual flow rate of ammonia water when the opening of the ammonia water regulating valve is i; F i is the weighted average value of the ammonia flow rate when the ammonia regulating valve is opened at degree i; β is an adjustable parameter, 0<β<1;

[0084] When β=0.9, 0.9 10 ≈0.35≈1 / e, with 1 / e as the dividing point. When the weight decays below this value, it is ignored.

[0085] For any β, Then the exponentially weighted average is the average of the most recent I openings:

[0086]

[0087] When β = 0.9, then at 100% opening, F 100=0.9*F 99 +(1-0.9)*θ 100 ;

[0088] The average value of the opening of the 100% ammonia regulating valve is obtained by weighting the opening data of the first 100%, and the weight coefficient conforms to an exponential decay function: when I = 10,

[0089] Correct the deviation: when i is very small, the denominator amplifies the current value (the denominator can amplify the current value very well); when i is very large, the value of the denominator tends to 1, which has almost no effect on the current value:

[0090]

[0091] Among them, DP is the deviation correction;

[0092] The output upper limit of the main PID module is:

[0093] The output lower limit of the main PID module is:

[0094] According to the current flue gas outlet NO X The emission value is obtained by the output value of the first-order inertia algorithm and the set value of the main PID module to obtain the output value of the main PID module.

[0095] In this embodiment, the output value of the main PID module is used as the setting value of the sub-PID module (SP 2 ), obtain the tracking value (PV 2 ) method includes: the tracking value is the output value (Y 2 ):

[0096]

[0097] Among them, Y(K) 2 is the output value of the current algorithm, that is, the output value of the current ammonia flow through the first-order inertia algorithm; Y(K-1) 2 is the output value of the previous period; FIT(K) is the current ammonia flow rate;

[0098] Obtain the output value of the sub-PID module according to the setting value and tracking value of the sub-PID module;

[0099] The process input value of the main PID module (that is, the tracking value PV of the main PID module) 1 ) is taken from the flue gas outlet NO XThe emission value is the output value of the first-order inertia algorithm. The tracking point (TP) of the main PID module is the output value of the sub-PID module, and the process input value of the sub-PID module (that is, the tracking value PV of the sub-PID module) 2 ) is taken from the output value of the current ammonia flow through the first-order inertia algorithm. The tracking point (TP) of the sub-PID module is the current opening of the ammonia regulating valve. The opening of the ammonia regulating valve is adjusted according to the output value of the sub-PID module.

[0100] Example 2

[0101] Figure 3 It is a principle block diagram of the automatic control system of the SNCR system of a waste power plant based on a machine learning algorithm involved in the present invention.

[0102] like Figure 3 As shown, based on Example 1, this Example 2 also provides an automatic control system for the SNCR system of a waste power plant based on a machine learning algorithm, including: a detection module, the detection module is suitable for detecting the current flue gas outlet NO X emission value, the opening of the current ammonia regulating valve and the ammonia flow rate of the waste power plant; an algorithm module, which is suitable for obtaining the output upper and lower limits of the main PID module according to the opening of the current ammonia regulating valve and the ammonia flow rate, and according to the current flue gas outlet NO X The emission value obtains the tracking value of the main PID module, and the tracking value of the auxiliary PID module is obtained according to the current ammonia flow rate; the main PID module (main PID adjustment), the main PID module is based on the set value of the main PID module and the current flue gas outlet NO X The emission value is corrected for deviation, and the output value of the main PID module is obtained according to the tracking value of the main PID module and the setting value of the main PID module; the auxiliary PID module (auxiliary PID adjustment), the auxiliary PID module is suitable for obtaining the output value of the auxiliary PID module according to the setting value and tracking value of the auxiliary PID module, so as to adjust the opening of the ammonia regulating valve according to the output value of the auxiliary PID module; the ammonia regulating valve adopts a cascade PID adjustment method, and the main PID module is adjusted with NO X The indicator is the controlled parameter. The auxiliary PID module takes the ammonia flow rate as the controlled parameter. The main PID adjustment output is used as the given value of the auxiliary PID adjustment. The auxiliary PID adjustment output controls the opening of the ammonia regulating valve. The machine learning module with exponential weighted average and deviation correction is used to set the upper and lower limits of the main regulating valve ammonia, which improves the problem that the traditional control method cannot accurately adjust the output amplitude, thereby achieving optimal control of NO X Emissions, to reduce ammonia slip.

[0103] In this embodiment, the automatic control system of the SNCR system of a waste power plant based on a machine learning algorithm is suitable for adopting the automatic control method of the SNCR system of a waste power plant based on a machine learning algorithm in Example 1 to adjust the opening of the ammonia regulating valve according to the output value of the sub-PID module.

[0104] In summary, the present invention obtains the NO of the current flue gas outlet of the waste power plant. X emission value; obtain the current opening of the ammonia regulating valve and the ammonia flow rate of the waste power plant; set the setting value of the main PID module, and X The emission value of the main PID module is obtained; the output upper and lower limits of the main PID module are obtained according to the current opening of the ammonia regulating valve and the ammonia flow rate, and the setting value of the main PID module and the current flue gas outlet NO are obtained. X The emission value is corrected for deviation; the output value of the main PID module is used as the set value of the sub-PID module, and the tracking value of the sub-PID module is obtained according to the current ammonia flow rate; the opening of the ammonia regulating valve is adjusted according to the output value of the sub-PID module, thereby achieving nitrogen oxide emission standards, reducing ammonia escape, improving the level of automation, and reducing the production cost of the SNCR user unit.

[0105] In several embodiments provided in the present application, it should be understood that the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the modules, methods and computer program products according to multiple embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a part of a module, a program segment or a code, and a part of the module, program segment or code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0106] In addition, the functional modules in the various embodiments of the present invention may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.

[0107] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0108] Based on the above ideal embodiments of the present invention, the relevant staff can make various changes and modifications without departing from the technical concept of the present invention through the above description. The technical scope of the present invention is not limited to the contents of the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A method for automatic control of SNCR system in waste power plant based on machine learning algorithm, characterized in that: include: Get the current flue gas outlet NO of the waste power plant X Emission value; Obtain the current opening degree of the ammonia regulating valve and ammonia flow rate of the waste power plant; Set the set value of the main PID module and adjust the current flue gas outlet NO X The emission value obtains the tracking value of the main PID module; The output upper and lower limits of the main PID module are obtained according to the current opening of the ammonia regulating valve and the ammonia flow rate, and the output upper and lower limits of the main PID module are obtained according to the set value of the main PID module and the current flue gas outlet NO X To make deviation correction for the emission values; The output value of the main PID module is used as the setting value of the auxiliary PID module, and the tracking value of the auxiliary PID module is obtained according to the current ammonia water flow rate; Adjust the opening of the ammonia regulating valve according to the output value of the auxiliary PID module; The setting value of the main PID module is set according to the current flue gas outlet NO X The method of obtaining the tracking value of the main PID module includes: The setting value of the main PID module is the flue gas outlet NO set according to relevant standards. X Emission value; The tracking value of the main PID module is the current flue gas outlet NO X The emission value is the output value of the first-order inertia algorithm; The current flue gas outlet NO X The method for obtaining the output value of the emission value through the first-order inertia algorithm includes: The filtering time is t f , the calculation period is t c , that is, Δt=t c ; Δt is the time difference of the calculation cycle; Low pass filtered first order inertial link transfer function; but y(s)·s·t f +y(s)=x(s); Through the reverse pull transformation, we can obtain: Obtained by the forward Euler method: dt is the sampling period of the discrete system, y k is the output of the first-order inertia link in the kth sampling period; but therefore: Current flue gas outlet NO X The output value of the emission value after the first-order inertia algorithm is: Among them, Y(K)1 is the output value of the current algorithm, that is, the current flue gas outlet NO X The emission value is the output value of the first-order inertia algorithm; Y(K-1)1 is the output value of the previous period; KG is the first-order inertia time constant; TC is the first-order inertia proportional gain; TS is the filtering time; NO X (K) is the current flue gas outlet NO X Emission value; The output upper and lower limits of the main PID module are obtained according to the current opening of the ammonia regulating valve and the ammonia flow rate, and the output upper and lower limits of the main PID module are obtained according to the set value of the main PID module and the current flue gas outlet NO X Methods for correcting the deviation of emission values ​​include: F i =β*F i-1 +(1-β)*θ i ; Where i is the current opening of the ammonia regulating valve; θ i is the actual flow rate of ammonia water when the opening of the ammonia water regulating valve is i; F i is the weighted average value of the ammonia flow rate when the ammonia regulating valve is opened at degree i; β is an adjustable parameter, 0<β<1; When β=0.9, 0.9 10 ≈0.35≈1 / e, with 1 / e as the dividing point. When the weight decays below this value, it is ignored. For any β, Then the exponentially weighted average is the average of the most recent I openings: When β = 0.9, then at 100% opening, F 100 =0.9*F 99 +(1-0.9)*θ 100 ; The average value of the opening of the 100% ammonia regulating valve is obtained by weighting the opening data of the first 100%, and the weight coefficient conforms to an exponential decay function: when I = 10, Correct the deviation: when i is very small, the denominator amplifies the current value; when i is very large, the value of the denominator tends to 1: Among them, DP is the deviation correction; The output upper limit of the main PID module is: The output lower limit of the main PID module is: According to the current flue gas outlet NO X The emission value is obtained by the output value of the first-order inertia algorithm and the set value of the main PID module to obtain the output value of the main PID module.

2. The automatic control method of the SNCR system of a waste power plant based on a machine learning algorithm as claimed in claim 1, characterized in that: The method of using the output value of the main PID module as the setting value of the auxiliary PID module and obtaining the tracking value of the auxiliary PID module according to the current ammonia water flow rate includes: The tracking value is the output value of the current ammonia flow rate after the first-order inertia algorithm: Among them, Y(K)2 is the output value of the current algorithm, that is, the output value of the current ammonia flow rate after the first-order inertia algorithm; Y(K-1)2 is the output value of the previous period; FIT(K) is the current ammonia flow rate; The output value of the sub-PID module is obtained according to the setting value and tracking value of the sub-PID module.

3. An automatic control system for the SNCR system of a waste-to-energy plant based on machine learning algorithms, Its features include: The detection module is suitable for detecting the current flue gas outlet NO of the waste power plant. X The emission value, the opening of the current ammonia regulating valve and the ammonia flow rate of the waste power plant; The algorithm module is suitable for obtaining the output upper and lower limits of the main PID module according to the current opening of the ammonia regulating valve and the ammonia flow rate, and the output of the main PID module according to the current flue gas outlet NO X The tracking value of the main PID module is obtained according to the emission value, and the tracking value of the auxiliary PID module is obtained according to the current ammonia flow rate; The main PID module is configured to generate a signal according to the set value of the main PID module and the current flue gas outlet NO. X The emission value is corrected for deviation, and the output value of the main PID module is obtained according to the tracking value of the main PID module and the set value of the main PID module; A sub-PID module, wherein the sub-PID module is adapted to obtain an output value of the sub-PID module according to a set value and a tracking value of the sub-PID module, so as to adjust the opening of the ammonia regulating valve according to the output value of the sub-PID module; The automatic control system of the SNCR system of a waste power plant based on a machine learning algorithm is suitable for adopting the automatic control method of the SNCR system of a waste power plant based on a machine learning algorithm as described in claim 1 or 2, so as to adjust the opening of the ammonia regulating valve according to the output value of the sub-PID module.

Images