CN113807196A - A method for obtaining the thermoelectric coupling characteristics of a cogeneration unit - Google Patents

Abstract

The invention provides a method for obtaining the thermoelectric coupling characteristics of a cogeneration unit, including: obtaining relevant data of the unit under all operating conditions in the heating season and non-heating season that need to be calculated, and performing preprocessing; Perform steady-state identification of data parameters, select steady-state operating conditions, identify abnormal changes in non-steady state, and eliminate them; introduce extraction steam pressure and extraction steam flow, and perform data training regression with relevant data parameters to obtain relevant data parameters and active power. The relationship between power; set each pressure parameter according to the value of the target working condition; limit the reasonable range of each flow parameter according to data statistics, and obtain the relationship between the power and the main steam flow under different flow rates under the target working condition, so as to achieve arbitrary steam extraction Determination of thermoelectric coupling characteristics of units under working conditions. The method described in the present disclosure is simple and efficient, can improve the accuracy of data, and provides an analysis method for the comprehensive energy supply capability and flexibility in a cogeneration system.

Description

A method for obtaining the thermoelectric coupling characteristics of a cogeneration unit

technical field

The invention relates to the technical field of production and scheduling of cogeneration units, in particular to a method for obtaining the thermoelectric coupling characteristics of a cogeneration unit.

Background technique

The information disclosed in this Background section is only for enhancement of understanding of the general background of the invention and should not necessarily be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person of ordinary skill in the art.

In recent years, the adjustment of my country's energy structure has been continuously advanced, and the proportion of new energy sources such as wind power and photovoltaic power generation has been continuously increased. In the face of the randomness and uncontrollability of new energy sources such as wind power and photovoltaic power generation, the increase in the proportion of their installed capacity will undoubtedly increase the disturbance on the power supply side. Traditional thermal power has played an important role in ensuring stable power supply, power auxiliary services and comprehensive energy supply. In the northern cities of my country, cogeneration units account for a large proportion of thermal power units. In the heating season, there is a constraint of determining electricity by heat, which restricts the peak shaving capacity of thermal power units. The introduction of industrial extraction steam has caused various heat loads and power. The coupling relationship of the unit is more complicated; after the flexibility transformation of the unit, the adjustable space of the unit's electric load in the heating season is increased to a certain extent, and the thermoelectric coupling characteristics of the unit will also deviate from the original design conditions. For the power grid dispatching department, if the thermoelectric coupling characteristics of the unit can be more accurately grasped, the mediation plan can be more reasonably arranged to promote the consumption of new energy.

The thermoelectric coupling characteristics of the cogeneration unit objectively reflect the peak shaving capability of the unit under a certain heat load. The common methods for determining the thermoelectric coupling characteristics mainly include working condition diagram analysis method, test method, and thermal calculation method. The test method has a wide range of applications and accurate results. However, it requires on-site tests and requires a lot of work. During the test, the electric and thermal loads are constantly adjusted, which has a certain impact on the power grid and heat network users. At the same time, with the aging of the equipment during the operation of the unit And related transformations, it is necessary to re-test and test to reflect the actual situation of the unit. The working condition diagram method is fast and convenient, but this method is obtained based on the design working conditions. With the installation, aging, and equipment renovation of the equipment, there is a certain deviation between the design value and the actual operating value of the unit, and it is obtained from the ordinary pure condensing unit. For heating units, this method has limitations. The thermodynamic calculation method has strong versatility and is suitable for various types of units, but the model is complex and the amount of calculation is large, and with the increase of the operating time of the unit, there is a deviation between the actual operating state of the unit and the model calculation, resulting in inaccurate results.

In addition, the thermal system of the cogeneration unit under the integrated energy supply is complex, and there is a coupling relationship between the systems. At present, most of the research is carried out on the typical design conditions of the unit, and different technical solutions are proposed for the heating units with different capacities and characteristics. The thermoelectric coupling characteristics of the unit depend on the matching degree of electricity, heat load and their load demand parameters, and power generation, industrial steam extraction and heating are the comprehensive energy supply forms of cogeneration units. Flexibility in load adjustment.

Therefore, it is of great significance to study a method that can obtain the thermoelectric coupling characteristics of cogeneration units with high applicability and accuracy.

SUMMARY OF THE INVENTION

In order to overcome the above problems, the present invention designs a method for obtaining the thermoelectric coupling characteristics of a cogeneration unit. For units with existing industrial steam extraction and heating steam extraction, data preprocessing is performed on historical operation data to eliminate data interference factors. Improve the accuracy of identification and prediction, use machine learning to perform regression analysis and modeling for complex relationships between variables, and establish active power and industrial extraction steam flow under various industrial extraction steam pressure conditions. The relationship between the heating extraction steam volume and the main steam flow under each main steam pressure condition is obtained, taking into account the thermoelectric coupling characteristics of industrial extraction steam, heating extraction steam and active power, overcoming the problem of deviation between design data and actual operation data, The available space for safe operation of thermal and electrical loads under the current operating conditions of the unit can be objectively given. The present disclosure is based on historical operation data analysis, not limited to typical operating conditions, and reflects the thermoelectric coupling relationship of the unit under any operating conditions.

Based on the above research results, the present disclosure provides the following technical solutions:

A first aspect of the present disclosure provides a method for obtaining the thermoelectric coupling characteristics of a cogeneration unit, comprising the following steps:

Step 1: Obtain the relevant data of the unit under the operating conditions of the unit in the heating season and non-heating season that need to be calculated;

Step 2: Preprocess the relevant data obtained in Step 1;

Step 3: Perform steady-state identification on the preprocessed relevant data parameters in step 2, select steady-state operating conditions, identify abnormal changes in non-steady-state conditions, and eliminate them;

Step 4: Introduce extraction steam pressure and extraction steam flow, perform data training regression with the relevant data parameters in step 3, and obtain the relationship between relevant extraction steam data parameters and active power;

Step 5: Set each pressure parameter according to the value of the target operating condition;

Step 6: Limit the reasonable range of each flow parameter according to data statistics;

Step 7: Through the reasonable range of Step 6, obtain the variation relationship between the active power and the main steam flow under different flow rates of the target operating conditions, so as to realize the determination of the thermoelectric coupling characteristics of the unit under any steam extraction conditions.

The second aspect of the present disclosure provides a software program that can be installed on a computer equipment terminal or a storage device to form an independent analysis platform, or embedded in the SIS system of a power plant in the form of an analysis module, so as to realize the above-mentioned method of obtaining thermoelectric power. The specific application of the method of the thermoelectric coupling characteristics of the cogeneration unit.

One or more specific embodiments of the present disclosure have achieved at least the following technical effects:

(1) The present disclosure starts from the massive operation data of the unit, uses the Regression Learner tool of MATLAB to select the regression model for training the preprocessed data, and establishes the mapping between the active power of the unit and the main steam, heating extraction steam, and industrial extraction steam. After setting the boundary conditions according to the historical data, the upper and lower limits of the adjustable space of the active power of the unit under any operating conditions can be obtained, reflecting the thermoelectric coupling characteristics of the unit.

(2) The present disclosure performs data preprocessing on the operating data of the unit, uses machine learning, and performs data regression modeling to obtain a comprehensive thermoelectric coupling characteristic model that takes into account industrial extraction steam, heating extraction steam and active power, which is more in line with the current operating characteristics of the unit . At the same time, by regularly updating the operating data of the unit used for processing training, the latest thermoelectric coupling characteristics obtained will be more in line with the characteristics of the actual operating unit, and can more accurately determine the peak shaving capability of the unit under various operating conditions, which is conducive to improving The accounting and supervision of the power grid dispatching department, the reasonable arrangement of mediation plans, and the promotion of new energy consumption have a wide range of application value.

(3) The present disclosure is applicable to the determination of the thermoelectric coupling characteristics of the unit under any steam extraction conditions, and to a certain extent, it can replace the thermodynamic test to determine the thermoelectric coupling characteristics of a part of the unit, or can be used as a supplement to the test to determine the thermoelectric coupling characteristics.

(4) The method described in the present disclosure is simple and efficient, can improve the accuracy of the data, and provides a feasible idea for the analysis and quantification of the comprehensive energy supply capacity of the cogeneration system.

Description of drawings

The accompanying drawings that constitute a part of the present disclosure are used to provide further understanding of the present disclosure, and the exemplary embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation of the present disclosure.

FIG. 1 is a smoothing denoising effect diagram of the active power of a unit in Embodiment 1 of the present invention;

Fig. 2 is the change of active power before and after the steady-state identification of the unit in Embodiment 1 of the present invention;

Fig. 3 is the comparative effect of active power prediction and real value in Embodiment 1 of the present invention;

Fig. 4 is the relation diagram of the maximum heating extraction steam flow rate and the industrial extraction steam flow rate in Embodiment 1 of the present invention;

Fig. 5 is the boundary diagram of the heating working condition of the unit with industrial extraction steam in Embodiment 1 of the present invention;

FIG. 6 is a characteristic diagram of thermoelectric coupling under the maximum industrial extraction steam working condition in Example 1 of the present invention.

Detailed ways

It should be noted that the following detailed description is exemplary and intended to provide further explanation of the present disclosure. Unless otherwise defined, 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 disclosure belongs.

It should be noted that the terminology used herein is for the purpose of describing specific embodiments only, and is not intended to limit the exemplary embodiments according to the present disclosure. As used herein, unless the context clearly dictates otherwise, the singular is intended to include the plural as well, furthermore, it is to be understood that when the terms "comprising" and/or "including" are used in this specification, it indicates that There are features, steps, operations, devices, components and/or combinations thereof.

As mentioned in the background, the thermoelectric coupling characteristics of the cogeneration unit obtained by the existing method have deviations from the actual operating state of the unit, resulting in the problem of inaccurate results, and at the same time, poor flexibility and adaptability. Therefore, the present disclosure provides a method for obtaining the thermoelectric coupling characteristics of a cogeneration unit.

A first aspect of the present disclosure provides a method for obtaining the thermoelectric coupling characteristics of a cogeneration unit, comprising the following steps:

Step 1: Obtain the relevant data of the unit under the operating conditions of the unit in the heating season and non-heating season that need to be calculated;

Step 2: Preprocess the relevant data obtained in Step 1;

Step 3: Perform steady-state identification on the preprocessed relevant data in step 2, select steady-state operating conditions, identify abnormal changes in non-steady state, and eliminate them;

Step 4: Introduce extraction steam pressure and extraction steam flow, perform data training regression with the relevant data parameters in step 3, and obtain the relationship between relevant data parameters and active power;

Step 5: Set each pressure parameter according to the value of the target operating condition;

Step 6: Limit the reasonable range of each flow parameter according to data statistics;

Step 7: Through the reasonable range of Step 6, obtain the variation relationship between the active power and the main steam flow under different flow rates of the target operating conditions, so as to realize the determination of the thermoelectric coupling characteristics of the unit under any steam extraction conditions.

In a typical implementation, in step 1, the relevant data includes active power, main steam pressure, main steam flow, main steam temperature, reheat steam temperature, heating extraction steam flow, heating extraction steam pressure, industrial extraction steam Flow rate and industrial extraction steam pressure, the present disclosure has obtained the above parameters as the research object through correlation analysis and continuous screening and optimization, which can have high accuracy and scientificity;

当满足|v_i|>3σ时，将该时刻的数据作为野值，进行剔除；式中，方差

n表示测量值的个数。现有技术中的野值剔除方法众多，包括均方值法、肖维涅法、一阶或二阶差分方法、时域微分方法等，各个方法均有自身的特点和适应对象，需要根据参数的自身特点选择相适应的野点剔除方法，才能够保证数据的完整性和处理的准确性。上述野值剔除方法适用性高，用于所取参数能够达到很好的野值剔除效果。进一步，所述不利因素包括电磁、噪声等，其对数据产生干扰，需要利用合适的方法进行去除，尽可能将不利影响降至最低，发明人发现MATLAB中medfilt1函数的处理方法能够很好的降低不良因素的干扰作用。In a typical implementation, in step 2, the preprocessing-related data parameters include: active power, main steam pressure, main steam flow, heating extraction steam flow, heating extraction steam pressure, industrial extraction steam flow, industrial extraction steam pressure; further, the preprocessing process includes: identifying and eliminating outliers, and using the medfiltl function in MATLAB to perform data filtering, smoothing and denoising, and eliminating the interference of unfavorable factors on the data; the reason why the above data parameters are outliers Identification and elimination are due to the complexity of the power generation system, and the sensors and data acquisition systems inevitably contain uncertain random errors and disturbances. The specific method for identifying and eliminating outliers is as follows: traversing 9 variables in turn in each group of data, and the residual corresponding to the i-th measurement value is

When |v _i |>3σ is satisfied, the data at this moment is regarded as an outlier and eliminated; in the formula, the variance

n represents the number of measurement values. There are many outlier removal methods in the prior art, including the mean square method, the Chauvigne method, the first or second order difference method, the time domain differentiation method, etc. Each method has its own characteristics and adaptation objects, and needs to be adjusted according to the parameters. Only by selecting the appropriate outlier elimination method according to its own characteristics can ensure the integrity of the data and the accuracy of the processing. The above outlier elimination method has high applicability, and can achieve a good outlier elimination effect when used for the selected parameters. Further, the unfavorable factors include electromagnetic, noise, etc., which interfere with the data and need to be removed by a suitable method to minimize the unfavorable impact as much as possible. The inventor found that the processing method of the medfilt1 function in MATLAB can be well reduced. Interfering effects of adverse factors.

In a typical implementation, in step 3, according to the changes of the four key parameters of active power, main steam temperature, main steam pressure, and reheat steam temperature, the abnormal changes in the non-steady state are identified and eliminated; The changes of the four key parameters are the important basis for judging the operating status of the unit, and the abnormal changes of the unit in the unsteady state are excluded through the changes of the four parameters;

超过阈值的时刻进行排除。其中，

为[t-d,d]时间段内的均值，ξ表示设定的阈值。Further, the specific steps are: traversing the active power, main steam temperature, main steam pressure, and reheat steam temperature in turn, and setting the threshold value of the unit parameter change within ten hours

Exclusions are made when the threshold is exceeded. in,

is the mean value in the [td,d] time period, and ξ represents the set threshold.

In a typical embodiment, in step 4, the extraction steam pressure and extraction steam flow are introduced, and the active power, main steam flow, main steam pressure, heating extraction steam flow, heating extraction steam pressure, industrial extraction steam in the above steps Flow, industrial extraction steam pressure, and 7 variables are subjected to data training regression to obtain the relationship between main steam pressure, heating extraction steam flow, heating extraction steam pressure, industrial extraction steam flow, industrial extraction steam pressure and active power;

The specific steps are: using the Regression Learner toolbox in MATLAB, take the main steam flow, main steam pressure, heating extraction steam flow, heating extraction steam pressure, industrial extraction steam flow, and industrial extraction steam pressure as the predictor variables, and the active power as the response. For the complex relationship between variables, the machine learning regression model selects the Gaussian process regression model for training. The Gaussian process regression does not find a single function, but the distribution of the optimal function found. By sampling this distribution, a series of The optimal function makes the thermoelectric coupling characteristic model tend to be accurate.

In a typical implementation, in step 5, the main steam pressure, heating extraction steam pressure, and industrial extraction steam pressure are set according to the values of the target operating conditions; in step 6, the main steam flow, heating Reasonable range of extraction steam flow and industrial extraction steam flow; the reason why the above steps are carried out is that the flow parameters and pressure parameters are one of the main factors affecting the power error of the heating unit under different working conditions.

In a typical implementation, in step 7, the boundary ranges of main steam flow, heating extraction steam flow, and industrial extraction steam flow are given in step 6, and the active power under different industrial extraction steam flow and heating extraction steam flow under the target operating condition is obtained. The relationship between power and main steam flow; among them, each extraction steam flow corresponds to the maximum and minimum value of the active power of the unit, so as to realize the determination of the thermoelectric coupling characteristics of the unit under any extraction steam condition. It can reflect the coupling relationship between the heat load and its extraction parameters and the electric load, and provides an analysis method for the comprehensive energy supply capacity and flexibility of the cogeneration system.

In the second aspect of the present disclosure, a software program can be written, installed on a computer equipment terminal or a storage device to form an independent analysis platform, or embedded in the SIS system of a power plant in the form of an analysis module, so as to realize the above-mentioned method of obtaining a cogeneration unit. Specific applications of the method for thermoelectric coupling characterization.

In order to enable those skilled in the art to understand the technical solutions of the present disclosure more clearly, the technical solutions of the present disclosure will be described in detail below with reference to specific embodiments and comparative examples.

Example 1

The specific conditions of the reference unit are as follows: subcritical, intermediate reheat, extraction steam condensing unit, model N330-16.67/537/537, the industrial steam extraction position is the cold section and hot section of the reheat steam, and the heating extraction steam is medium pressure Cylinder exhaust.

Using the operation data analysis proposed by the present invention to determine the thermoelectric coupling characteristics of the unit, the method includes the following steps:

Step 1: Obtain the operating data of the unit for a period of time from the database, including active power, main steam pressure, main steam flow, main steam temperature, reheat steam temperature, heating extraction steam flow, heating extraction steam pressure, and industrial extraction steam flow , Industrial extraction steam pressure, the amount of data is based on the length of the time span of the data taken;

当满足|v_i|>3σ时将该时刻的数据作为野值，并进行剔除；式中，方差

n表示测量值的个数；再对有功功率、主汽压力、主汽流量、供暖抽汽流量、供暖抽汽压力、工业抽汽流量、工业抽汽压力等7个参数利用MATLAB中medfilt1函数进行中值滤波处理进行平滑与去噪。有功功率的平滑、去噪效果，如图1所示，从中可以看出经过处理后消除了一些电磁噪声等不利因素产生的波形偏差，从而消除因电磁、噪声等因素对数据的干扰；Step 2: In order to eliminate the errors generated in the data collection process, the active power, main steam pressure, main steam flow, heating extraction steam flow, heating extraction steam pressure, industrial extraction steam flow, and industrial extraction steam pressure data obtained in step 1 were analyzed. Identification and elimination of outliers; the specific steps are to traverse 7 variables in turn in each group of data, and the residual corresponding to the i-th measurement value is

When |v _i |>3σ is satisfied, the data at this moment is regarded as an outlier and eliminated;

n represents the number of measured values; then use the medfilt1 function in MATLAB for seven parameters such as active power, main steam pressure, main steam flow, heating extraction steam flow, heating extraction steam pressure, industrial extraction steam flow, and industrial extraction steam pressure. Median filter processing for smoothing and denoising. The smoothing and denoising effect of active power is shown in Figure 1. It can be seen that the waveform deviation caused by some unfavorable factors such as electromagnetic noise is eliminated after processing, thereby eliminating the interference of electromagnetic, noise and other factors on the data;

超过阈值的时刻进行排除。其中，

为[t-d,d]时间段内的均值，ξ表示设定的阈值，有功功率阈值10％，主汽压力阈值为2％，主汽温度阈值为10％，再热蒸汽温度阈值为10％。Step 3: Carry out steady-state identification of the processed data, identify the abnormal changes in non-steady state and eliminate them according to the changes of the four key parameters of active power, main steam temperature, main steam pressure, and reheat steam temperature; The specific steps are to traverse the active power, main steam temperature, main steam pressure, and reheat steam temperature in turn, and set the unit parameter change threshold within ten hours.

Exclusions are made when the threshold is exceeded. in,

is the average value in the [td,d] time period, ξ represents the set threshold, the active power threshold is 10%, the main steam pressure threshold is 2%, the main steam temperature threshold is 10%, and the reheat steam temperature threshold is 10%.

The change of active power before and after the above preprocessing step is shown in Figure 2; the waveform after steady-state identification becomes simplified, and the removed part is the waveform generated by the abnormal change of the unsteady state.

Step 4: Regress the active power, main steam flow, main steam pressure, heating extraction steam flow, heating extraction steam pressure, industrial extraction steam flow, and industrial extraction steam pressure under the above steps, the regression model selects Gaussian process regression, and the kernel function Choose an exponential function. The specific steps are to use the Regression Learner toolbox in MATLAB to take 6 variables of main steam flow, main steam pressure, heating extraction steam flow, heating extraction steam pressure, industrial extraction steam flow, and industrial extraction steam pressure as the predictor variables, and the active power as the response. . The machine learning regression model selects the Gaussian process regression model for training. The Gaussian process regression does not find a single function, but the distribution of the optimal functions found. By sampling this distribution, a series of optimal functions can be obtained, so that the thermoelectric The coupled characteristic model tends to be accurate. After training the data, input the actual prediction parameters to obtain the active power prediction comparison, and obtain the relationship between the main steam pressure, heating extraction steam flow, heating extraction steam pressure, industrial extraction steam flow, industrial extraction steam pressure and active power; as shown in Figure 3, The comparison between the real results and the predicted results can be seen from the figure, which can reflect the feasibility of using this training method and establish a basis for the determination of thermoelectric coupling characteristics.

Step 5: Set the main steam pressure, heating extraction steam pressure, and industrial extraction steam pressure according to the value of the target operating condition;

Step 6: Limit the reasonable range of main steam flow, heating extraction steam flow, and industrial extraction steam flow according to data statistics; the variation range of each parameter is shown in Table 1;

Table 1 Variation range of main parameters

Step 7: The variation range of main steam flow, heating extraction steam flow, and industrial extraction steam flow is given in step 6, and the variation relationship between active power and main steam flow under different extraction steam flow rates is obtained. The lower part corresponds to the maximum and minimum active power, so the thermoelectric coupling characteristics of the unit are obtained;

Existing thermal tests have adjusted the industrial steam extraction for the unit under the condition of no heating steam extraction. The test data recorded the maximum adjustable range of active power under different industrial steam extraction conditions, reflecting the true non-heating steam extraction of the unit. Thermoelectric coupling characteristics under industrial extraction steam, some test data of this unit under corresponding extraction steam conditions are shown in Table 2 below:

Table 2 Unit data under various industrial extraction steam conditions

Obtain the predicted thermoelectric coupling characteristics through the above steps, input the same extraction parameters as the test data and the corresponding flow boundary above, and the obtained results are shown in Table 3 below:

Table 3. Comparison table of unit operation data and test data effect

It can be seen from Table 3 that the calculation using the model obtained by the method proposed in the present invention can accurately obtain the thermoelectric coupling characteristics of the unit without heating extraction and with industrial extraction.

When the industrial extraction steam and the heating extraction steam exist at the same time, the maximum heating extraction steam flow will be limited by the industrial extraction steam flow, and the heat load becomes complicated. As shown in Figure 4, it is the situation corresponding to the maximum heating extraction steam flow of the unit when the industrial extraction steam is changed. Through the model, it can be calculated that when the main steam pressure is 16.3MPa, the heating extraction steam pressure is 0.45Mpa, and the industrial extraction steam pressure is 2.25MPa, the industrial extraction steam flow is from 0t/h to a maximum of 150t/h, and the heating extraction steam flow is 0t/h. h to the thermoelectric coupling characteristics under the corresponding maximum flow rate, the predicted range of thermoelectric coupling characteristics, that is, the boundary of the heating condition with industrial extraction steam, is shown in Figure 5. The thermoelectric coupling characteristics of the unit under the maximum industrial extraction steam condition are as follows As shown in Figure 6, it is the leftmost part of the boundary of the heating condition with industrial steam extraction in Figure 5. It can be seen that the increase of industrial flow will reduce the power of the unit, and the power adjustable space will become smaller under the same main steam flow .

Through the method proposed in the present invention, the thermoelectric coupling characteristics under various operating conditions can be obtained, and the calculation results can reflect that the thermoelectric coupling characteristics of the units under various operating conditions are significantly different.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still understand the foregoing embodiments. The technical solutions described are modified, or some technical features thereof are equivalently replaced. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims (10)

1. a method for obtaining cogeneration unit thermoelectric coupling characteristics, is characterized in that: comprise the following steps: Step 1: Obtain the relevant data under the operating conditions of the unit in heating season and non-heating season that need to be calculated; Step 2: Preprocess the relevant data obtained in Step 1; Step 3: Perform steady-state identification on the preprocessed relevant data parameters in step 2, select steady-state operating conditions, identify abnormal changes in non-steady-state conditions, and eliminate them; Step 4: Introduce extraction steam pressure and extraction steam flow, perform data training regression with the relevant data parameters in step 3, and obtain the relationship between relevant data parameters and active power; Step 5: Set each pressure parameter according to the value of the target operating condition; Step 6: Limit the reasonable range of each flow parameter according to data statistics; Step 7: Through the reasonable range of Step 6, obtain the variation relationship between the active power and the main steam flow under different flow rates of the target operating conditions, so as to realize the determination of the thermoelectric coupling characteristics of the unit under any steam extraction conditions. 2. The method according to claim 1, wherein: in step 1, the relevant data parameters include active power, main steam pressure, main steam flow, main steam temperature, reheat steam temperature, heating extraction steam flow, Heating extraction steam pressure, industrial extraction steam flow, industrial extraction steam pressure: Or, in step 2, the preprocessing-related data parameters include: active power, main steam pressure, main steam flow, heating extraction steam flow, heating extraction steam pressure, industrial extraction steam flow, and industrial extraction steam pressure; Or, in step 2, the preprocessing process includes: identifying and eliminating outliers, and using the medfiltl function in MATLAB to perform data filtering, smoothing and denoising, and eliminating the interference of unfavorable factors on the data. 3. method according to claim 2 is characterized in that: in step 2, the concrete method of described carrying out outlier identification and elimination is: in each group of data, 7 variables are traversed successively, the i-th measurement value The corresponding residual is

When |v _i |>3σ is satisfied, the data at this moment is regarded as an outlier and eliminated; in the formula, the variance

n represents the number of measurement values. 4. method according to claim 1 is characterized in that: in step 3, according to active power, the change situation of four key parameters of main steam temperature, main steam pressure, reheat steam temperature, identify unsteady abnormality Changes are excluded. 5. The method according to claim 4, characterized in that: the concrete steps are: traversing the active power, main steam temperature, main steam pressure, and reheat steam temperature in turn, and setting the unit parameter change threshold within ten hours

Exclude the moment when the threshold is exceeded; among them,

is the mean value in the [td, d] time period, and ξ represents the set threshold. 6. method according to claim 1, is characterized in that: in step 4, introduce extraction steam pressure, extraction steam flow, the active power under above-mentioned steps, main steam flow, main steam pressure, heating extraction steam flow, heating Extraction steam pressure, industrial extraction steam flow, and industrial extraction steam pressure are used for data training regression to obtain the relationship between main steam pressure, heating extraction steam flow, heating extraction steam pressure, industrial extraction steam flow, industrial extraction steam pressure and active power. 7. method according to claim 6 is characterized in that: concrete steps are: utilize RegressionLearner toolbox in MATLAB, by main steam flow, main steam pressure, heating extraction steam flow, heating extraction steam pressure, industrial extraction steam flow, Seven variables of industrial extraction steam pressure are used as predictors, and active power is used as the response. For the complex relationship between variables, the machine learning regression model selects the Gaussian process regression model for training. 8. The method according to claim 1, characterized in that: in step 5, the main steam pressure, the heating extraction steam pressure, and the industrial extraction steam pressure are set according to the values of the target operating conditions. 9. The method according to claim 1, characterized in that: in step 6, a reasonable range of main steam flow, heating extraction steam flow, and industrial extraction steam flow is defined according to data statistics. 10. method according to claim 1 is characterized in that: in step 7, by step 6 given main steam flow, heating extraction steam flow, industrial extraction steam flow boundary range, obtain the target operating condition different industrial extraction steam flow and The relationship between the active power and the main steam flow under the heating extraction steam flow; among them, each extraction steam flow corresponds to the maximum and minimum value of the active power of the unit, so as to realize the determination of the thermoelectric coupling characteristics of the unit under any extraction steam condition.

Images