Thermal power plant air supply control loop optimization method, system, device and storage medium

Abstract

The present application discloses a method, system, device and computer-readable storage medium for optimizing a thermal power air supply control loop, including: using sensors to obtain the operating parameters of the air supply loop of the thermal power unit; judging whether the operating parameters meet the preset evaluation criteria; if the evaluation criteria are not met, optimizing the control parameters in the optimization model to obtain the optimization results; sending the optimization results to the control model; wherein the optimization model is a pre-established model for parameter optimization corresponding to the current control model of the air supply loop. After judging that the control model of the air supply loop does not meet the preset evaluation criteria, the present application uses a pre-established optimization model for parameter optimization corresponding to the current control model of the air supply loop to replace the control model for optimization adjustment, without the need to take the control model offline for parameter adjustment, so that the optimization process does not require the control model to be offline, ensuring that the thermal power unit can continue to operate stably at any time without interfering with production.

Description

Technical Field

[0001] The present invention relates to the field of distributed storage, and in particular to a method, system, device and computer-readable storage medium for optimizing a thermal power plant air supply control loop. Background Art

[0002] Thermal power units have hundreds of control loops, and the quality of the control loops can directly affect the safety and economy of the thermal power unit. On the one hand, the quality of automatic control can directly affect the stable and reliable operation of the thermal power unit. On the other hand, with the deterioration of the equipment of the operating unit, the regulation characteristics of the control loops of the thermal power unit will also change accordingly. Moreover, due to the lack of an online evaluation mechanism, it becomes more difficult for engineers to maintain the control loops and keep them in the optimal regulation state, and maintenance is not timely.

[0003] The air supply control loop of a thermal power plant is an important guarantee for ensuring the economic and reliability of boiler combustion in a thermal power plant. Therefore, the automatic adjustment of the air supply loop plays a vital role in the coordinated control of the entire thermal power plant. At present, the automatic control of the air supply loop mostly adopts a feedforward-feedback control architecture. However, during the operation of the thermal power unit, the air supply control blades often deteriorate, causing the adjustment characteristics to change, resulting in poor air supply control quality. However, the daily maintenance of engineers lacks theoretical and data guidance, which makes it impossible to optimize and maintain the automatic control of the air supply loop in a timely manner, and to adjust the PID control parameters in a timely manner to adapt to the new equipment characteristics.

[0004] Therefore, a thermal power air supply control loop optimization method is needed which can optimize control parameters online without affecting the continuous and stable operation of thermal power units. Summary of the invention

[0005] In view of this, the purpose of the present invention is to provide a method, system, device and computer-readable storage medium for optimizing the air supply control loop of a thermal power plant, which has high timeliness and can optimize the control parameters online without affecting the continuous and stable operation of the thermal power unit. The specific scheme is as follows:

[0006] A method for optimizing a thermal power plant air supply control loop, comprising:

[0007] Use sensors to obtain the operating parameters of the air supply circuit of the thermal power unit;

[0008] Determining whether the operating parameters meet preset evaluation criteria;

[0009] If the evaluation criteria are not met, the control parameters in the optimization model are optimized to obtain optimization results;

[0010] Sending the optimization result to the control model;

[0011] The optimization model is a pre-established model for parameter optimization corresponding to the current control model of the air supply circuit.

[0012] Optionally, the process of obtaining the operating parameters of the air supply circuit of the thermal power unit by using a sensor includes:

[0013] Using the sensor to obtain original operating parameters of the air supply circuit of the thermal power unit;

[0014] The original operating parameters are cleaned to obtain the operating parameters.

[0015] Optionally, the process of performing data cleaning on the original operating parameters to obtain the operating parameters includes:

[0016] Performing a stability check on the original operating parameters to obtain verification data that satisfies the stability check;

[0017] Eliminate singular data from the verification data to obtain normal data whose sample deviation meets the preset distribution conditions;

[0018] The Laida criterion is used to eliminate the error data that does not meet the conditions in the normal data to obtain the operating parameters.

[0019] Optionally, the process of optimizing the control parameters in the optimization model to obtain the optimization result includes:

[0020] The control parameters in the optimization model are optimized using a particle swarm optimization algorithm to obtain the optimization result.

[0021] Optionally, the process of determining whether the operating parameter meets a preset evaluation criterion includes:

[0022] Using the evaluation calculation formula and the operating parameters, an evaluation parameter is obtained;

[0023] Determining whether the evaluation parameters meet the evaluation criteria;

[0024] Wherein, the evaluation calculation formula includes:

[0025] a=actual investment time / theoretical investment time;

[0026] b = l1 / l2;

[0027] c = s1 / s2;

[0028] The evaluation criteria include: within the time period t,

[0029] When a<70%, the evaluation score of the loop y=0;

[0030] When a≥70%, the loop evaluation score y=a-x1b+x2c;

[0031] In the formula, a represents the automatic input rate of the air supply circuit during period t, b represents the maximum deviation ratio of the air supply circuit during period t, c represents the excess area ratio of the air supply circuit during period t, l1 represents the maximum difference between the regulated quantity and the set value during period t, l2 represents the rated value of the air supply circuit, s1 represents the excess area between the regulated quantity and the maximum allowable deviation during period t, s2 represents the integrated area of ​​the rated value during period t, and x is the weight of indicators b and c.

[0032] Optionally, the process of building the optimization model includes:

[0033] Building an optimization model framework for parameter optimization of the air supply circuit according to the model framework of the current control model of the air supply circuit;

[0034] Using the operating parameters, identifying and obtaining a model structure and model parameters of the control model;

[0035] The optimization model is obtained by using the model structure, the model parameters and the optimization model framework.

[0036] The present invention also discloses a thermal power air supply control loop optimization system, comprising:

[0037] A parameter acquisition module, used to acquire the operating parameters of the air supply circuit of the thermal power unit by using sensors;

[0038] An evaluation and judgment module, used to judge whether the operating parameters meet the preset evaluation criteria;

[0039] A model optimization module, used for optimizing the control parameters in the optimization model to obtain optimization results if the evaluation criteria are not met;

[0040] A result sending module, used for sending the optimization result to the control model;

[0041] The optimization model is a pre-established model for parameter optimization corresponding to the current control model of the air supply circuit.

[0042] Optionally, also include:

[0043] An architecture building module, used to build an optimization model architecture for parameter optimization of the air supply circuit according to the model architecture of the current control model of the air supply circuit;

[0044] A model identification module, used to identify and obtain the model structure and model parameters of the control model using the operating parameters;

[0045] The model building module is used to obtain the optimization model by using the model structure, the model parameters and the optimization model architecture.

[0046] The present invention also discloses a thermal power air supply control loop optimization device, comprising:

[0047] Memory for storing computer programs;

[0048] A processor is used to execute the computer program to implement the thermal power air supply control loop optimization method as described above.

[0049] The present invention also discloses a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the thermal power air supply control loop optimization method as described above is implemented.

[0050] In the present invention, a method for optimizing a thermal power plant air supply control loop comprises: obtaining operating parameters of an air supply loop of a thermal power plant unit by using sensors; determining whether the operating parameters meet preset evaluation criteria; if the evaluation criteria are not met, optimizing the control parameters in an optimization model to obtain optimization results; and sending the optimization results to a control model; wherein the optimization model is a pre-established model for parameter optimization corresponding to a current control model of the air supply loop.

[0051] After determining that the control model of the air supply circuit does not meet the preset evaluation criteria, the present invention utilizes a pre-established optimization model for parameter optimization corresponding to the current control model of the air supply circuit to replace the control model for optimization adjustment without taking the control model offline for parameter adjustment, so that the optimization process does not require the control model to be offline, thereby ensuring that the thermal power unit can continue to operate stably at any time without interfering with production. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0053] Figure 1 A schematic flow chart of a thermal power plant air supply control loop optimization method disclosed in an embodiment of the present invention;

[0054] Figure 2 A schematic flow chart of another thermal power plant air supply control loop optimization method disclosed in an embodiment of the present invention;

[0055] Figure 3 An evaluation analysis curve diagram disclosed in an embodiment of the present invention;

[0056] Figure 4 A schematic diagram of an optimization model and a comparison model architecture disclosed in an embodiment of the present invention;

[0057] Figure 5 A schematic diagram of a set value calculation model disclosed in an embodiment of the present invention;

[0058] Figure 6 An input and output schematic diagram disclosed in an embodiment of the present invention;

[0059] Figure 7 The present invention is a schematic diagram of the structure of a thermal power plant air supply control loop optimization system disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] The embodiment of the present invention discloses a method for optimizing a thermal power plant air supply control loop, see Figure 1 As shown, the method includes:

[0062] S11: Using sensors to obtain operating parameters of the air supply circuit of the thermal power unit.

[0063] Specifically, the sensor data at the site is collected, and the operating parameters of the air supply circuit in the thermal power unit obtained by the sensor are received through OPC or other communication methods. The sensor can transmit the data to the plant-level supervisory information system (Supervisory Information System, referred to as SIS).

[0064] S12: Determine whether the operating parameters meet the preset evaluation criteria.

[0065] It is understandable that the control model of the air supply circuit that does not meet the preset production target needs to be optimized, and the control model of the air supply circuit that can achieve the preset production target does not need to be optimized temporarily. Therefore, after collecting the operating parameters of the thermal power unit, the operating parameters of the thermal power unit are used to determine whether the control model of the air supply circuit meets the preset evaluation criteria. If it meets the criteria, it does not need to be optimized temporarily. If it does not meet the criteria, subsequent optimization is required. For example, the evaluation criterion is that the operating efficiency of the thermal power unit is not less than 90%. If the current operating efficiency of the thermal power unit is lower than 90%, the control model of the air supply circuit needs to be optimized. If it is higher, no optimization is required. The evaluation criteria can also be the start-up time, stabilization time and overshoot time of the air supply circuit.

[0066] S13: If the evaluation criteria are not met, the control parameters in the optimization model are optimized to obtain optimization results.

[0067] Specifically, if the operating parameters do not meet the evaluation criteria, the air supply loop control model needs to be optimized. However, since the control model of the air supply loop needs to run continuously to ensure the continuous, stable and normal operation of the thermal power unit, it cannot be easily interrupted and taken offline. Therefore, the current control model of the air supply loop is not directly debugged, but an optimization model for parameter optimization corresponding to the current control model of the air supply loop is pre-established. The optimization model can be constructed based on the known architecture of the control model, and the specific parameters of the control model can be deduced based on the operating parameters generated during the operation of the control model, thereby obtaining an optimization model that is highly similar to or even exactly the same as the control model. Therefore, debugging the optimization model is equivalent to indirect debugging and optimization of the control model, thereby achieving online closed-loop optimization.

[0068] Specifically, when the air supply circuit is evaluated and optimized for the first time, an optimization model needs to be established on site. If the air supply circuit that needs to be optimized currently has an optimization model established, then there is no need to re-establish a new optimization model, and the previously established optimization model can be directly used for optimization. In other words, it is only necessary to establish an optimization model for the air supply circuit that does not have a corresponding optimization model.

[0069] S14: Send the optimization result to the control model.

[0070] Specifically, after obtaining the optimization result, the optimization result can be sent to the control model, and the original control parameters in the control model can be directly replaced according to the optimized control parameters in the optimization result, thereby completing the control parameter optimization of the control model and realizing online closed-loop optimization.

[0071] It can be seen that after determining that the control model of the air supply circuit does not meet the preset evaluation criteria, the embodiment of the present invention uses a pre-established optimization model for parameter optimization corresponding to the current control model of the air supply circuit to replace the control model for optimization adjustment, without the need to take the control model offline for parameter adjustment, so that the optimization process does not require the control model to be offline, thereby ensuring that the thermal power unit can continue to operate stably at any time without interfering with production.

[0072] The embodiment of the present invention discloses a specific method for optimizing a thermal power plant air supply control loop. Compared with the previous embodiment, this embodiment further illustrates and optimizes the technical solution. Figure 2 As shown, specifically:

[0073] S21: using sensors to obtain original operating parameters of the air supply circuit of the thermal power unit;

[0074] S22: Perform data cleaning on the original operating parameters to obtain operating parameters.

[0075] Specifically, the original operating parameters of the air supply circuit collected by the sensor will inevitably contain erroneous data or abnormal data caused by errors in sensor collection. Such data needs to be eliminated to avoid affecting the subsequent optimization results. For this reason, the original operating parameters are cleaned to obtain reliable operating parameters.

[0076] Furthermore, S22 performs data cleaning on the original operating parameters to obtain the specific process of the operating parameters, which may include S221 to S223; wherein,

[0077] S221: Perform a stability check on the original operating parameters to obtain verification data that meets the stability check.

[0078] Specifically, since the online closed-loop identification data requires continuous and stable data, a stability test is required before entering the next link. If the data is not stable, the next step of calculation will not be entered.

[0079] Specifically, for continuously changing sensor time series data, its stability has a crucial impact on what kind of analysis method we adopt and what kind of model we choose. Stability is divided into strong stability and weak stability. The requirements for strong stability are very strict, and it is difficult to verify in the actual application of thermal processes, and there is no application scenario. Therefore, the embodiment of the present invention mainly uses weak stability, which has three main requirements:

[0080] 1. For any time series Xt, the mean of the series is a constant, E(Xt) = μ;

[0081] 2. For any period of time series Xt, its variance exists;

[0082] 3. For any time series Xt, any integer h, and any order l, Yl(Xt)=Yl(Xt+h);

[0083] After the distance h between the two points, the l-order autocovariance of the two sets of data will not fluctuate over time and is only related to the order. A set of time series data, whose mean is constant, variance always exists, and autocovariance does not fluctuate over time, is considered a weakly balanced sequence and is used as verification data.

[0084] S222: Remove singular data from the verification data to obtain normal data whose sample deviation meets the preset distribution conditions.

[0085] Specifically, the sensor data in the thermal process changes continuously and smoothly. During the entire thermal process operation, the time series data sample size is large, and under normal conditions, the data distribution satisfies or basically satisfies the normal distribution data sample. In addition to artificial disturbances in the thermal process, abnormal operation conditions will cause the data sample deviation to be particularly large, and the reasons for this type of data cannot or do not need to be accurately obtained in the subsequent in-depth analysis process. This type of data becomes suspicious data in the analysis process, and the data state should be marked as a singular condition. The data within this time range can be eliminated in subsequent analysis. Therefore, on the basis of verification data, singular data is eliminated to obtain cleaner normal data whose sample deviation meets the preset distribution conditions.

[0086] S223: Using the Laida criterion, the error data that does not meet the conditions in the normal data is eliminated to obtain the operating parameters.

[0087] Specifically, since the sample size of the real-time data platform is very large, it meets the prerequisite for the application of the Laida rule. The Laida rule means that a set of test data is assumed to contain only random errors, and the standard deviation is calculated and processed to determine an interval according to a certain probability. It is believed that any error exceeding this interval is not a random error but a gross error, and the data containing this error should be eliminated. This discriminant processing principle and method is limited to the processing of sample data with normal or approximately normal distribution, and it is based on the premise that the number of measurements is sufficiently large.

[0088] Specifically, assume that the measured data are measured with equal precision, and x1, x2..., xn are obtained independently. The arithmetic mean x and residual error vi=xi-x (i=1,2,...,n) are calculated, and the standard deviation σ is calculated according to the Bessel formula. If the residual error vb (1<=b<=n) of a certain measurement value xb satisfies the following formula:

[0089] |vb|=|xb-x|>3σ;

[0090] It is considered that xb is a bad value containing a large error value and should be eliminated to obtain the operating parameters.

[0091] S23: Obtain evaluation parameters using the evaluation calculation formula and the operating parameters.

[0092] Specifically, in the air supply circuit evaluation system, the deviation between the statistically relevant set value and the actual value is adopted. On the basis of identifying the dynamic process and steady-state process of the unit operation, the positive standard deviation and negative standard deviation are evaluated to evaluate the quality of its regulation.

[0093] Specifically, the state of the air supply circuit includes steady state and dynamic state. In the actual evaluation process, statistical calculations are performed separately for the two states. In order to facilitate the explanation of the calculation method, the embodiment of the present invention considers that the circuit is in the same state within the time period t; in the actual evaluation process, the statistical period is 1 month and the calculation period is 1 hour. In order to facilitate the explanation of the calculation method, the time period t is considered to be a calculation cycle. The allowable deviation of the dynamic quality indicators of the load change test and AGC load following test specified in the industry standard is ±100t / h for the air supply circuit in the embodiment of the present invention.

[0094] Specifically, the evaluation calculation formula includes:

[0095] a=actual investment time / theoretical investment time;

[0096] b = l1 / l2;

[0097] c = s1 / s2;

[0098] Where a represents the automatic input rate of the air supply circuit in period t, b represents the maximum deviation ratio of the air supply circuit in period t, and c represents the excess area ratio of the air supply circuit in period t. Figure 3 As shown, l1 represents the maximum difference between the regulated quantity and the set value during the period t, l2 represents the rated value of the air supply circuit, s1 represents the excess area between the regulated quantity and the maximum allowable deviation during the period t, and s2 represents the integral area of ​​the rated value during the period t.

[0099] S24: Determine whether the evaluation parameters meet the evaluation criteria.

[0100] The evaluation criteria include: within the t period,

[0101] When a<70%, the evaluation score of the loop y=0;

[0102] When a≥70%, the loop evaluation score y=a-x1b+x2c;

[0103] In the formula, x1 and x2 are the weights of indicators b and c, respectively, which can be obtained by calculating the minimum score and deviation percentage set by the evaluation.

[0104] Specifically, it can be set that when the control model score of the air supply circuit is lower than 80 points, it does not meet the evaluation criteria and is optimized.

[0105] S25: If the evaluation criteria are not met, the particle swarm optimization algorithm is used to optimize the control parameters in the optimization model to obtain an optimization result.

[0106] Specifically, the particle swarm optimization algorithm is used to repeatedly optimize and adjust the control parameters summarized by the optimization model, so as to obtain the expected optimization results.

[0107] S26: Send the optimization result to the control model;

[0108] The optimization model is a pre-established model for parameter optimization corresponding to the current control model of the air supply circuit.

[0109] Furthermore, the process of constructing the above optimization model may include S31 to S33; wherein,

[0110] S31: Building an optimization model framework for parameter optimization of the air supply circuit according to the model framework of the current control model of the air supply circuit.

[0111] Specifically, the automatic control of the air supply circuit is modeled through the configuration software. At the same time, the parameters of its PID adjustment function module are derived, and the optimized PID parameters can be dynamically accepted for verification and calibration.

[0112] For details, see Figure 4 As shown, two Simulink models can be built, one as an optimization model and the other as a comparison model, to compare the optimization effect of the optimization model after optimization. For loops with unfixed set values, a set value calculation model can be added to provide set values ​​for optimization comparison, such as Figure 5 shown.

[0113] Each model can include three parts: controller, object and performance monitoring. Controller: build the automatic operation part according to the air supply circuit sama diagram or configuration diagram; object: import the results of model identification; performance monitoring: has no effect on operation, but is used to facilitate viewing of the result curve.

[0114] S32: Using the operating parameters, identify and obtain the model structure and model parameters of the control model.

[0115] Specifically, the on-site historical data is the data record collected during the normal operation of the thermal power unit. Since the feedback loop is not disconnected, it is closed-loop data. During the normal operation of the unit, the load changes frequently and the unit operating conditions are constantly adjusted. At this time, the closed-loop operation data is very rich. Therefore, based on the closed-loop identification principle and the on-site historical operation data, the closed-loop identification of the air supply circuit of the thermal power unit is carried out to obtain the model structure and model parameters that correctly reflect the dynamic characteristics of the actual object.

[0116] Specifically, for most system model identification problems, the structure of the system model to be identified is unknown, and it is necessary to determine the structural form of the system mathematical model based on the known information of the system to be identified. Taking the ARX model of formula 1 as an example, it is necessary to determine the model order n and the process delay d.

[0117] A(z -1 )y(k)=z -d B(z -1 )u(k)+e(k) (1)

[0118] When the singularity is equal to the model order r, a sudden drop will occur. However, when the measurement signal is interfered by noise, the number of non-zero singular values ​​will be much larger than the model order r. At this time, the non-zero singular values ​​greater than the model order are all generated by noise, and their values ​​are generally small and the amplitude of change is small. The influence of noise will make it difficult to detect the sudden change of the singular value of the diagonal matrix S in the model order. In general, when observing the changes in the singular value curve, when its rate of change is small and the curve is close to the horizontal asymptote, it can be considered as the order of the model.

[0119] The model parameters are further identified, and the specific process is as follows:

[0120] Transform equation (1) to obtain the following equation:

[0121] y(k)=f[y(k-1),…y(0),u(k-1),…,u(1),θ]+e(k) (2)

[0122] Among them, y(k) is the q-dimensional output vector, u(k) is the p-dimensional input vector, and θ is the parameter vector of the model.

[0123] set up:

[0124]

[0125] Then formula (2) can be written as:

[0126] y(k)=f(y (k-1) ,u (k-1) ,θ)+e(k) (4)

[0127] Formula (4) can be used to calculate the output at time k using data from the past. (k-1) and u (k -1) After that, the “best” prediction for the output y(k) can be taken as its conditional mathematical expectation, i.e.

[0128]

[0129] Formula (5) satisfies

[0130]

[0131] At this time, this "best" output forecast is the output of the "best" model. This can be achieved by minimizing the forecast error criterion. Commonly used forecast error criteria are the following two in formula (7):

[0132]

[0133] The weight matrix Λ is a positive definite matrix.

[0134]

[0135] When the data length N→∞, D(θ) will converge to the covariance matrix of e(k). The parameter estimate obtained by minimizing J1(θ) or J2(θ) is the prediction error estimate of the object model.

[0136] Specifically, specific inputs and outputs are determined for different objects; the inputs include: the controlled quantity, the main disturbance, and the output is the controlled quantity, for example Figure 6 shown.

[0137] S33: Utilize the model structure, model parameters and optimized model architecture to obtain an optimized model.

[0138] Correspondingly, the embodiment of the present invention also discloses a thermal power air supply control loop optimization system, see Figure 6 As shown, the system includes:

[0139] A parameter acquisition module, used to acquire the operating parameters of the air supply circuit of the thermal power unit by using sensors;

[0140] An evaluation and judgment module is used to judge whether the operating parameters meet the preset evaluation criteria;

[0141] The model optimization module is used to optimize the control parameters in the optimization model to obtain the optimization results if the evaluation criteria are not met;

[0142] A result sending module, used for sending the optimization result to the control model;

[0143] The optimization model is a pre-established model for parameter optimization corresponding to the current control model of the air supply circuit.

[0144] It can be seen that after determining that the control model of the air supply circuit does not meet the preset evaluation criteria, the embodiment of the present invention uses a pre-established optimization model for parameter optimization corresponding to the current control model of the air supply circuit to replace the control model for optimization adjustment, without the need to take the control model offline for parameter adjustment, so that the optimization process does not require the control model to be offline, thereby ensuring that the thermal power unit can continue to operate stably at any time without interfering with production.

[0145] Specifically, it may also include an architecture building module, a model identification module and a model building module; wherein,

[0146] An architecture building module, used to build an optimization model architecture for parameter optimization of the air supply circuit according to the model architecture of the current control model of the air supply circuit;

[0147] A model identification module is used to identify and obtain the model structure and model parameters of the control model using the operating parameters;

[0148] The model building module is used to obtain an optimized model by utilizing the model structure, model parameters and optimized model architecture.

[0149] Specifically, the evaluation and judgment module may include a parameter calculation unit and an evaluation and judgment unit; wherein,

[0150] A parameter calculation unit, used to obtain evaluation parameters using evaluation calculation formulas and operating parameters;

[0151] An evaluation judgment unit, used to judge whether the evaluation parameters meet the evaluation criteria;

[0152] The evaluation calculation formula includes:

[0153] a=actual investment time / theoretical investment time;

[0154] b = l1 / l2;

[0155] c = s1 / s2;

[0156] The evaluation criteria include: within the t period,

[0157] When a<70%, the evaluation score of the loop y=0;

[0158] When a≥70%, the loop evaluation score y=a-x1b+x2c;

[0159] In the formula, a represents the automatic start-up rate of the air supply circuit during period t, b represents the maximum deviation ratio of the air supply circuit during period t, c represents the excess area ratio of the air supply circuit during period t, l1 represents the maximum difference between the regulated quantity and the set value during period t, l2 represents the rated value of the air supply circuit, s1 represents the excess area between the regulated quantity and the maximum allowable deviation during period t, s2 represents the integrated area of ​​the rated value during period t, and x is the weight of indicators b and c.

[0160] Specifically, the parameter acquisition module includes a parameter acquisition unit and a parameter cleaning unit; wherein,

[0161] A parameter acquisition unit, used for acquiring original operating parameters of the air supply circuit of the thermal power unit by using sensors;

[0162] The parameter cleaning unit is used to perform data cleaning on the original operating parameters to obtain the operating parameters.

[0163] Specifically, the parameter cleaning unit may include a stationarity check subunit, a singular data elimination subunit and a Laida criterion elimination subunit; wherein,

[0164] The stability check subunit is used to check the stability of the original operating parameters and obtain the check data that meets the stability check;

[0165] The singular data elimination subunit is used to eliminate the singular data of the verification data to obtain normal data whose sample deviation meets the preset distribution conditions;

[0166] The Laida criterion elimination subunit is used to eliminate the error data that does not meet the conditions in the normal data using the Laida criterion to obtain the operating parameters.

[0167] Specifically, the model optimization module is specifically used to optimize the control parameters in the optimization model using the particle swarm optimization algorithm to obtain the optimization results.

[0168] In addition, the embodiment of the present invention also discloses a thermal power air supply control loop optimization device, including:

[0169] Memory for storing computer programs;

[0170] A processor is used to execute a computer program to implement the aforementioned thermal power plant air supply control loop optimization method.

[0171] In addition, an embodiment of the present invention further discloses a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the aforementioned thermal power air supply control loop optimization method is implemented.

[0172] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0173] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0174] The technical content provided by the present invention is introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for general technical personnel in the field, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for optimizing a thermal power plant air supply control loop, characterized in that: include: Use sensors to obtain the operating parameters of the air supply circuit of the thermal power unit; Determining whether the operating parameters meet preset evaluation criteria; If the evaluation criteria are not met, the control parameters in the optimization model are optimized to obtain optimization results; Wherein, the optimization model is a pre-established model for parameter optimization corresponding to the current control model of the air supply circuit; Sending the optimization result to the control model; The process of using sensors to obtain the operating parameters of the air supply circuit of the thermal power unit includes: Using the sensor to obtain original operating parameters of the air supply circuit of the thermal power unit; Performing data cleaning on the original operating parameters to obtain the operating parameters; The process of performing data cleaning on the original operating parameters to obtain the operating parameters includes: Performing a stability check on the original operating parameters to obtain verification data that satisfies the stability check; The weakly balanced sequence is used as the verification data. The weakly stable sequence includes: For any period of sensor time series Xt, the mean of the series is constant, E(Xt) = μ, for any period of sensor time series Xt, its variance exists, and for any period of sensor time series Xt, any integer h, any order l, there is Yl(Xt) = Yl(Xt+h); Wherein, E(Xt) is the mean of the sensor time series, Yl(Xt) is the autocovariance of the Xt sensor time series, Yl(Xt+h) is the autocovariance of the Xt+h sensor time series, and the sensor time series is the original operating parameters collected by the sensor; Eliminate singular data from the verification data to obtain normal data whose sample deviation meets the preset distribution conditions; The Laida criterion is used to eliminate the error data that does not meet the conditions in the normal data to obtain the operating parameters.

2. The thermal power air supply control loop optimization method according to claim 1, characterized in that: The process of optimizing the control parameters in the optimization model to obtain the optimization result includes: The control parameters in the optimization model are optimized using a particle swarm optimization algorithm to obtain the optimization result.

3. The thermal power plant air supply control loop optimization method according to claim 1, characterized in that: The process of determining whether the operating parameters meet the preset evaluation criteria includes: Using the evaluation calculation formula and the operating parameters, an evaluation parameter is obtained; Determining whether the evaluation parameters meet the evaluation criteria; Wherein, the evaluation calculation formula includes: a=actual investment time / theoretical investment time; b＝l1 / l2； c = s1 / s2; The evaluation criteria include: within the time period t, When a<70%, the evaluation score of the loop y=0; When a≥70%, the loop evaluation score y=a-x1b+x2c; In the formula, a represents the automatic input rate of the air supply circuit during period t, b represents the maximum deviation ratio of the air supply circuit during period t, c represents the excess area ratio of the air supply circuit during period t, l1 represents the maximum difference between the regulated quantity and the set value during period t, l2 represents the rated value of the air supply circuit, s1 represents the excess area between the regulated quantity and the maximum allowable deviation during period t, s2 represents the integrated area of ​​the rated value during period t, and x is the weight of indicators b and c.

4. The method for optimizing the thermal power plant air supply control loop according to any one of claims 1 to 3, characterized in that: The process of building the optimization model includes: Building an optimization model framework for parameter optimization of the air supply circuit according to the model framework of the current control model of the air supply circuit; Using the operating parameters, identifying and obtaining a model structure and model parameters of the control model; The optimization model is obtained by using the model structure, the model parameters and the optimization model framework.

5. A thermal power plant air supply control loop optimization system, characterized in that: include: A parameter acquisition module, used to acquire the operating parameters of the air supply circuit of the thermal power unit by using sensors; An evaluation and judgment module, used to judge whether the operating parameters meet the preset evaluation criteria; A model optimization module, used for optimizing the control parameters in the optimization model to obtain optimization results if the evaluation criteria are not met; Wherein, the optimization model is a pre-established model for parameter optimization corresponding to the current control model of the air supply circuit; A result sending module, used for sending the optimization result to the control model; The parameter acquisition module includes a parameter acquisition unit and a parameter cleaning unit; wherein, The parameter acquisition unit is used to acquire the original operating parameters of the air supply circuit of the thermal power unit by using sensors; The parameter cleaning unit is used to perform data cleaning on the original operating parameters to obtain the operating parameters; The parameter cleaning unit includes a stability check subunit, a singular data elimination subunit and a Laida criterion elimination subunit; wherein, The stability check subunit is used to perform stability check on the original operating parameters to obtain verification data that meets the stability check; The weakly balanced sequence is used as the verification data. The weakly stable sequence includes: For any period of sensor time series Xt, the mean of the series is constant, E(Xt) = μ, for any period of sensor time series Xt, its variance exists, and for any period of sensor time series Xt, any integer h, any order l, there is Yl(Xt) = Yl(Xt+h); Wherein, E(Xt) is the mean of the sensor time series, Yl(Xt) is the autocovariance of the Xt sensor time series, Yl(Xt+h) is the autocovariance of the Xt+h sensor time series, and the sensor time series is the original operating parameters collected by the sensor; The singular data elimination subunit is used to eliminate the singular data of the verification data to obtain normal data whose sample deviation meets the preset distribution condition; The Laida criterion elimination subunit is used to eliminate error data that does not meet the conditions in normal data using the Laida criterion to obtain operating parameters.

6. The thermal power air supply control loop optimization system according to claim 5, characterized in that: Also includes: An architecture building module, used to build an optimization model architecture for parameter optimization of the air supply circuit according to the model architecture of the current control model of the air supply circuit; A model identification module, used to identify and obtain the model structure and model parameters of the control model using the operating parameters; The model building module is used to obtain the optimization model by using the model structure, the model parameters and the optimization model architecture.

7. A thermal power plant air supply control loop optimization device, characterized in that: include: Memory for storing computer programs; A processor, used to execute the computer program to implement the thermal power air supply control loop optimization method as described in any one of claims 1 to 4.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for optimizing the thermal power plant air supply control loop according to any one of claims 1 to 4 is implemented.

Images