Thermal power plant production index analysis method based on unit load segmentation

Abstract

The invention relates to a thermal power plant production index analysis method based on unit load segmentation, and the method comprises the steps: obtaining data, and forming a fitting data set; performing fitting analysis on the fitting data set to obtain a fitting curve; segmenting the fitting curve according to the load of the unit; obtaining empirical value data of corresponding production indexes of each unit load section; forming a piecewise function of the reference value of the production index and the unit load according to the fitting curve and the empirical value; acquiring a real-time unit load; judging a section range where the real-time unit load is located; and comparing the real-time value of the production index with the reference value of the production index, and adjusting parameters of the generator set or performing maintenance according to a comparison result. According to the thermal power plant production index analysis method based on unit load segmentation, quantitative statistics of the production indexes is realized, and crew personnel can quickly and accurately respond to uneconomical operation, non-environment-friendly operation and abnormal operation of a generator set.

Description

Technical Field

[0001] The invention relates to the field of thermal power unit data analysis, and in particular to a method for analyzing production indicators of a thermal power plant based on unit load segmentation. Background Art

[0002] Thermal power plants involve many production indicators, including water flow, vacuum, circulating water temperature rise, terminal difference, feed water temperature, water makeup rate, main reheat steam temperature, exhaust temperature, exhaust oxygen content, fly ash combustibles, outlet CO content, condenser cleanliness, boiler efficiency, coal consumption, etc. Based on long-term experience accumulation or analysis of historical data under normal conditions, the reference value of each production indicator under normal conditions can be calculated. In the actual production process, if the real-time value of the production indicator of the generator set is less than the reference value, it is considered that the generator set is operating under normal conditions, otherwise it is abnormal. The crew refers to the reference value to determine whether the generator set is operating under normal conditions, so that it can detect abnormalities such as air leakage and water leakage of the generator set, or adjust the unit parameters to ensure the economy and environmental protection of the generator set operation.

[0003] In the prior art, a specific method for calculating reference values ​​through historical data is: measuring the specific value of each production indicator every day, and calculating the daily average value of each production indicator; taking the arithmetic average of the daily average values ​​of several daily production indicators to obtain the reference value of each production indicator.

[0004] With the establishment of the spot trading mechanism in the electricity market, the flexibility of grid load allocation has been significantly enhanced. The load of units entering the spot trading market has fluctuated significantly compared with the past, and the load variation rate (load variation rate = load variation time / total operating time*100%) has gradually increased from less than 10% to more than 30%.

[0005] However, the daily average values ​​of various production indicators calculated by the prior art are single, and do not take into account the impact of changes in unit load on various production indicators to change accordingly. It is not possible to accurately indicate whether an abnormal situation has occurred in the unit when the unit load changes rapidly, and it is difficult to effectively guide the operating personnel to adjust the parameters, inspect and maintain the unit in a timely manner, resulting in a variety of production indicators of the generator set deviating from the ideal environmental protection and economic operation range for a long time. Therefore, the production indicator reference values ​​of the prior art still have the problems of low general applicability and low accuracy. In addition, due to the wide variety of production indicators and the complex influence relationship between multiple production indicators, the analysis and calculation of each production indicator is complex, which not only increases the difficulty of operation optimization, but also affects the competitiveness and operational efficiency of the power plant in the electricity spot market. Summary of the invention

[0006] Based on this, the purpose of the present invention is to overcome the defects or shortcomings of the prior art and provide an analysis method for production indicators of a thermal power plant based on unit load segmentation, comprising the following steps:

[0007] Step S1: obtaining the measured value of each production indicator and its corresponding unit load to form a fitting data set; performing fitting analysis on the data of the fitting data set to obtain a fitting curve;

[0008] Step S2: segmenting the fitting curve according to the load size of the unit;

[0009] Step S3: Obtaining empirical value data of production indicators corresponding to the load sections of each unit;

[0010] Step S4: forming a piecewise function of the reference value of the production index and the unit load according to the fitting curve and the empirical value, wherein the piecewise function is divided into a fitting curve segment and an empirical value segment;

[0011] Step S5: obtaining the real-time unit load of the generator set on that day;

[0012] Step S6: determining the range of the real-time unit load;

[0013] Step S7: calculating the reference value of the production index according to the piecewise function of the reference value of the production index and the unit load;

[0014] Step S8: obtaining the real-time value of the production index of the generator set on that day; comparing the real-time value of the production index with the reference value of the production index, and adjusting the parameters of the generator set or performing maintenance according to the comparison result.

[0015] The method for analyzing production indicators of a thermal power plant based on unit load segmentation described in the present invention realizes quantitative statistics of production indicators. The statistical reference value of the production indicator can change with the change of unit load. It has the advantages of high universal applicability and high accuracy, and is conducive to the crew's rapid and accurate response to the uneconomical operation, environmentally unfriendly operation and abnormal operation of the generator set.

[0016] Furthermore, in step S2, after the fitting curve is segmented according to the unit load size, the following further includes: calculating the fitting degree R of each unit load segment respectively. 2 ;

[0017] The step S3 specifically includes: obtaining the fitting degree R 2 The empirical value data of production indicators corresponding to the unit load range of <0.8;

[0018] In step S4, the fitting curve segment is the fitting degree R 2 ≥0.8 unit load section; the empirical value section is the fitting degree R 2Unit load section of <0.8;

[0019] The piecewise function of the reference value of the production index and the unit load formed in step S4 is specifically:

[0020]

[0021] Among them, z 参考值 Indicates the reference value of the production index, R 2 Indicates the fitting degree of the fitting curve segment.

[0022] Furthermore, in step S1, the fitting data set also includes other influencing factors affecting the production index value other than the unit load;

[0023] Step S5 also includes: acquiring real-time data of the influencing factor.

[0024] Furthermore, the fitting curve includes a fitting curve of "production index-unit load" or a fitting curve of "production index-unit load-influencing factor".

[0025] Furthermore, the optimization step of the fitting curve includes:

[0026] Step S9: Obtain the real-time unit load, real-time data of influencing factors and corresponding real-time value data of production indicators within a period of time, and calculate the average value of the real-time unit load, real-time data of influencing factors and corresponding real-time value data of production indicators corresponding to each section in the fitting curve segment; add the average value to the fitting data set and update the coefficient of the fitting curve.

[0027] Furthermore, the real-time unit load, real-time data of influencing factors and corresponding real-time value data of production indicators within the fitting curve segment are added to the fitting data set to update the coefficients of the fitting curve.

[0028] Furthermore, before forming the fitting data set in step S1, the method further includes: eliminating data with large unit load fluctuations.

[0029] Furthermore, the data of excluding the unit with large load fluctuations specifically includes:

[0030] All unit load data are sorted by time, and the absolute value of the difference between the unit load obtained at a certain moment and the unit load at the moment before the unit load obtained at this moment is calculated; it is determined whether the absolute value is ≥6MW. If so, the unit load obtained at this moment and the data of its corresponding production index are eliminated; if not, the unit load obtained at this moment and the data of its corresponding production index are added to the fitting data set.

[0031] Furthermore, before step S9 "updating coefficients of the fitting curve", the method further includes: removing historical data of the fitting data set with a time period of two years or more.

[0032] Furthermore, the production indicators include water flow, vacuum, circulating water temperature rise, terminal difference, feed water temperature, water makeup rate, main reheat steam temperature, exhaust gas temperature, exhaust gas oxygen content, fly ash combustibles, outlet CO content, condenser cleanliness, boiler efficiency, and coal consumption; the piecewise function is specifically:

[0033]

[0034] Among them, z 参考值 Indicates the reference value of the production index, x 实时 represents the real-time unit load, y 实时 represents the real-time data of the influencing factor; X0 represents the rated unit load of the generator set, a, b, c represent the coefficients of the piecewise function; y 当日 Indicates the unit load rate for the day.

[0035] For better understanding and implementation, the present invention is described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 The present invention is a flow chart of a method for analyzing production indicators of a thermal power plant based on unit load segmentation. DETAILED DESCRIPTION

[0037] In response to the problem that the reference value of the production index cannot adapt to the large fluctuations in the unit load in the spot trading market, the technicians of the present invention first counted the historical daily average data of the unit load, and based on the daily average data of the existing production indicators, performed fitting analysis on each production indicator and the unit load, and obtained the linear relationship between the unit load and each production indicator, in order to try to use the fitting curve to calculate the production indicator values ​​corresponding to different unit loads under normal operating conditions, and use the calculated value as a reference value to adapt to the large fluctuations in the unit load.

[0038] Although the reference value calculated by the above calculation does change with the load of the random unit and has a high degree of fit, the technicians found that the reference value calculated by the above method still cannot accurately indicate whether an abnormal situation has occurred in the unit. Specifically, the standard deviation between the production index reference value calculated by the above fitting curve and the actual production index value under normal conditions is large. Directly referring to the production index value may cause the crew to misjudge or miss the operation of the generator set.

[0039] The technicians analyzed that the reason why the above reference values ​​still cannot accurately indicate whether the unit has an abnormal situation is that in the historical operation of the unit, the daily unit load is basically stable and rarely changes. The daily average values ​​of the unit load and various production indicators further concentrate the trend of data changes, and some data with drastic changes are not reflected. In response to this, the technicians further improved the analysis method of production indicators: directly using the specific measurement values ​​of each production indicator obtained by daily measurement and its corresponding unit load data for fitting analysis, and obtaining the fitting curve of each production indicator and unit load. However, the fitting degree of this fitting curve is small and not representative.

[0040] In this regard, the technicians tried to use different target fitting functions for fitting, trying to improve the fitting degree of the fitting curve. However, the technicians found that the fitting degree of the fitting curve was still very small. In this regard, the technicians of the present invention carefully analyzed the fitting curve and found that in some fitting curves, although the fitting degree of the entire fitting curve was very small, the fitting degrees corresponding to some unit load sections in the fitting curve were relatively large, representative, and could accurately reflect the relationship between various production indicators and unit loads under normal working conditions, while other unit load sections could use corresponding empirical values ​​as reference values.

[0041] Based on this, see Figure 1 The present invention provides a method for analyzing production indicators of a thermal power plant based on unit load segmentation:

[0042] (I) Calculate the fitting relationship between each production index and unit load

[0043] Step S1: Obtain the measured values ​​z1, z2, z3, ... z of each production indicator n , and its corresponding unit load x1, x2, x3...x n , forming a fitting data set; performing fitting analysis on the data of the fitting data set to obtain a fitting curve of "zx"; preferably, the fitting data set also includes other influencing factors y1, y2, y3...y that affect the production index value n , such as atmospheric temperature, cooling water temperature, etc., to obtain a fitting curve of "zxy" to improve the fitting effect of the fitting curve.

[0044] Step S2: Divide the fitting curve into sections according to the load size of the unit, and calculate the fitting degree R of each unit load section respectively. 2 ;

[0045] Step S3: Obtain the fitness R 2 The empirical value data of production indicators corresponding to the unit load range of <0.8;

[0046] Step S4: Form a piecewise function of the reference value of the production index and the unit load, wherein the piecewise function is divided into a fitting curve segment and an empirical value segment, and the fitting curve segment is a fitting degree R 2 ≥0.8, the corresponding function is the fitting curve; the empirical value section is the fitting degree R 2 The unit load section of <0.8, the corresponding function is the empirical value of the production index; the formula of the piecewise function is:

[0047]

[0048] Among them, z 参考值 Indicates the reference value of the production index, R 2 Indicates the fitting degree of the fitting curve segment.

[0049] (II) Real-time production indicator analysis

[0050] Step S5: Obtain the real-time unit load x of the generator set on that day 实时 , preferably, also includes obtaining real-time data of the influencing factors 实时 ;

[0051] Step S6: determining the range of the real-time unit load;

[0052] Step S7: Calculate the reference value z of the production index according to the piecewise function of the reference value of the production index and the unit load. 参考值 Specifically, if the real-time unit load is in the empirical value section of the piecewise function, then step S7a is executed: the empirical value is the reference value of the production index; if the real-time unit load is in the fitting curve section of the piecewise function, then step S7b is executed: the real-time unit load value is substituted into the formula corresponding to the fitting curve, and the calculation result is the reference value of the production index;

[0053] Step S8: Obtain the real-time value of the production index of the generator set on that day 实时 ; Compare the real-time value of the production index with the reference value of the production index, and adjust the parameters of the generator set or perform maintenance according to the comparison result.

[0054] (III) Optimizing the piecewise function of production index and unit load

[0055] Step S9: Obtain the real-time unit load x within a period of time 实时 , real-time data of impact factors 实时 And the corresponding real-time value data of production indicators z 实时 , calculate the average value x corresponding to each segment in the fitting curve segment 平均 ,y 平均 、z平均 , preferably a daily average; adding the average value to the fitting data set, and updating the coefficients of the fitting curve.

[0056] Optionally, the real-time unit load within the fitting curve segment, the real-time data of the influencing factors and the corresponding real-time value data of the production index may be directly added to the fitting data set to update the coefficient of the fitting curve.

[0057] The method for analyzing production indicators of a thermal power plant based on unit load segmentation of the present invention is described in detail below in conjunction with specific embodiments.

[0058] Example 1

[0059] In this embodiment, the production index is the vacuum value of the steam turbine, and the analysis method of the production index of the thermal power plant based on the unit load segmentation specifically includes the following steps:

[0060] (I) Calculate the fitting relationship between vacuum value and unit load

[0061] Step S1: Obtain the measured value of the vacuum value and its corresponding unit load and cooling water temperature data to form a fitting data set of "vacuum-unit load-cooling water temperature"; set the target fitting function to be z=ax+by+c, where z represents the vacuum value (Pa), x represents the unit load (MW), y represents the cooling water temperature (℃), and a, b, and c represent the coefficients of the target fitting function. Perform fitting analysis on the data of the "vacuum-unit load-cooling water temperature" fitting data set to obtain a fitting curve of "vacuum-unit load-cooling water temperature", specifically:

[0062] z=-0.0131735670349853x-0.380820728720826y+107.473559553764;

[0063] Step S2: Segment the fitting curve according to the unit load size, and calculate the fitting degree of each unit load section respectively. Specifically, X0 represents the rated unit load of the generator set. In this embodiment, the fitting degrees R of the three sections (45% X0, 55% X0), (70% X0, 80% X0), and (90% X0, 100% X0) are 2 ≥0.8, the fitting degree R of the three sections (0%X0, 45%X0), (55%X0, 70%X0), (80%X0, 90%X0) 2 <0.8;

[0064] Step S3: Obtaining empirical value data of vacuum values ​​corresponding to the ranges of (0%X0, 45%X0), (55%X0, 70%X0), and (80%X0, 90%X0);

[0065] Step S4: forming a piecewise function of the reference value of the vacuum value and the unit load, the piecewise function includes a fitting curve segment and an empirical value segment, wherein the fitting curve segment includes: (45%X0, 55%X0), (70%X0, 80%X0), (90%X0, 100%X0) segments; the empirical value segment includes: (0%X0, 45%X0), (55%X0, 70%X0), (80%X0, 90%X0) segments; the piecewise function of the vacuum value and the unit load is specifically:

[0066]

[0067] Among them, z 参考值 Indicates the reference value of vacuum value (Pa), x 实时 represents the real-time unit load (MW), y 实时 represents the real-time cooling water temperature (°C); in this embodiment, a=-0.0131735670349853, b=-0.380820728720826, c=107.473559553764;

[0068] (II) Real-time vacuum value analysis

[0069] Step S5: Obtain the real-time unit load x of the generator set on that day 实时 , Real-time cooling water temperature 实时 ;

[0070] Step S6: determining the range of the real-time unit load;

[0071] Step S7: Calculate the reference value of the vacuum value according to the piecewise function of the vacuum value and the unit load. Specifically: if the segment range of the real-time unit load is in the fitting curve segment of the piecewise function, substitute the real-time unit load value into the formula corresponding to the fitting curve of "vacuum-unit load-cooling water temperature", and the calculation result is the reference value of the vacuum value; if the segment range of the real-time unit load is in the empirical value segment of the piecewise function, the empirical value is the reference value of the vacuum value;

[0072] Step S8: Obtain the real-time vacuum value z of the generator set on that day 实时 ; Compare the real-time value of the vacuum value with the reference value of the vacuum value, and adjust the parameters of the vacuum equipment of the generator set or repair the vacuum equipment according to the comparison result.

[0073] (III) Optimizing the piecewise function of vacuum value and unit load

[0074] Step S9: Obtain the real-time unit load x in the fitting curve segment within a period of time 实时, Real-time cooling water temperature 实时 And the real-time vacuum value z 实时 , calculate the average value x corresponding to each segment of the fitting curve segment (45%X0, 55%X0), (70%X0, 80%X0), (90%X0, 100%X0) 平均 ,y 平均 、z 平均 , preferably a daily average value; adding the average value to the "vacuum-unit load-cooling water temperature" fitting data set, and updating the coefficient of the "vacuum-unit load-cooling water temperature" fitting curve.

[0075] Optionally, the real-time unit load, real-time cooling water temperature and corresponding real-time vacuum value data within the fitting curve segment can be directly added to the "vacuum-unit load-cooling water temperature" fitting data set to update the coefficient of the "vacuum-unit load-cooling water temperature" fitting curve.

[0076] Example 2

[0077] In this embodiment, the production index is the feed water temperature, and the analysis method of the production index of a thermal power plant based on unit load segmentation specifically includes the following steps:

[0078] (I) Calculate the fitting relationship between feed water temperature and unit load

[0079] Step S1: Obtain the measured value of the feed water temperature and its corresponding unit load and air temperature data to form a fitting data set of "feed water temperature-unit load-air temperature"; assume that the target fitting function is z=ax+by+c, where z represents the feed water temperature (℃), x represents the unit load (MW), y represents the air temperature (℃), and a, b, and c represent the coefficients of the target fitting function. Perform fitting analysis on the data of the "feed water temperature-unit load-air temperature" fitting data set to obtain a fitting curve of "feed water temperature-unit load-air temperature", specifically:

[0080] z=0.238803968619097x+0.143178839524597y+194.682348617231;

[0081] Step S2: Segment the fitting curve according to the unit load size, and calculate the fitting degree of each unit load section respectively. Specifically, X0 represents the rated unit load of the generator set. In this embodiment, the fitting degrees R of the three sections (45% X0, 55% X0), (70% X0, 80% X0), and (90% X0, 100% X0) are 2≥0.8, the fitting degree R of the three sections (0%X0,45%X0), (55%X0,70%X0), (80%X0,90%X0) 2 <0.8;

[0082] Step S3: Obtaining empirical value data of the feed water temperature corresponding to the range of (0%X0, 45%X0), (55%X0, 70%X0), (80%X0, 90%X0);

[0083] Step S4: forming a piecewise function of feed water temperature and unit load, the piecewise function includes a fitting curve segment and an empirical value segment, wherein the fitting curve segment includes: (45%X0, 55%X0), (70%X0, 80%X0), (90%X0, 100%X0) segments; the empirical value segment includes: (0%X0, 45%X0), (55%X0, 70%X0), (80%X0, 90%X0) segments; the piecewise function of feed water temperature and unit load is specifically:

[0084]

[0085] Among them, z 参考值 Indicates the reference value of feed water temperature (℃), x 实时 represents the real-time unit load (MW), y 实时 represents the real-time temperature (°C); in this embodiment, a=0.238803968619097, b=0.143178839524597, c=194.682348617231;

[0086] (II) Real-time water supply temperature analysis

[0087] Step S5: Obtain the real-time unit load x of the generator set on that day 实时 , Real-time temperature 实时 ;

[0088] Step S6: determining the range of the real-time unit load;

[0089] Step S7: Calculate the reference value of the feed water temperature according to the piecewise function of the feed water temperature and the unit load. Specifically: if the range of the real-time unit load is in the fitting curve section of the piecewise function, substitute the real-time unit load value into the formula corresponding to the fitting curve of "feed water temperature-unit load-air temperature", and the calculation result is the reference value of the feed water temperature; if the range of the real-time unit load is in the empirical value section of the piecewise function, the empirical value is the reference value of the feed water temperature;

[0090] Step S8: Obtain the real-time feedwater temperature z of the generator set on that day 实时; Compare the real-time value of the feed water temperature with the reference value of the feed water temperature, and adjust the parameters of the generator set feed water temperature equipment or repair the feed water temperature equipment according to the comparison result.

[0091] (III) Optimizing the piecewise function of feed water temperature and unit load

[0092] Step S9: Obtain the real-time unit load x in the fitting curve segment within a period of time 实时 , Real-time temperature 实时 And real-time feed water temperature 实时 , calculate the average value x corresponding to each segment of the fitting curve segment (45%X0, 55%X0), (70%X0, 80%X0), (90%X0, 100%X0) 平均 ,y 平均 、z 平均 , preferably a daily average value; adding the average value to the "feed water temperature-unit load-air temperature" fitting data set, and updating the coefficient of the "feed water temperature-unit load-air temperature" fitting curve.

[0093] Optionally, the real-time unit load, real-time cooling water temperature and corresponding real-time vacuum value data within the fitting curve segment can be directly added to the "feed water temperature-unit load-air temperature" fitting data set to update the coefficient of the "feed water temperature-unit load-air temperature" fitting curve.

[0094] Example 3

[0095] Example 3 is basically the same as Example 1, except that the production index is the exhaust gas temperature. Accordingly, the fitting curve of "exhaust gas temperature-unit load-air temperature" is calculated as follows:

[0096] z=0.125137994861767x+0.726541162932605y+86.5209074713367;

[0097] Among them, z represents the exhaust gas temperature (°C), x represents the unit load (MW), y represents the air temperature (°C), and a, b, and c represent the coefficients of the target fitting function.

[0098] The piecewise function of exhaust gas temperature and unit load is as follows:

[0099]

[0100] Among them, z 参考值 Indicates the reference value of exhaust gas temperature (℃), x 实时 represents the real-time unit load (MW), y 实时represents the real-time temperature (°C); in this embodiment, a=0.125137994861767, b=0.726541162932605, c=86.5209074713367;

[0101] Example 4

[0102] Example 4 is a variation of the above-mentioned Examples 1 to 3, and a fitting curve with a higher degree of fit can be further obtained. The difference between Example 4 and Examples 1 to 3 is that: the production index is the heat consumption affected by the flow rate of condensate per ton (hereinafter referred to as heat consumption), and the fitting curve is the fitting curve of "heat consumption-unit load rate". Among them, the unit load rate refers to the percentage of the average unit load to the maximum unit load during the statistical period. The analysis method of the production index of a thermal power plant based on unit load segmentation specifically includes the following steps:

[0103] (I) Calculation of the fitting relationship between heat consumption and unit load rate

[0104] Step S1: Obtain the measured value of heat rate and its corresponding unit load rate data to form a fitting data set of "heat rate-unit load rate"; set the target fitting function to be z=ay 2 +by+c, where z represents heat consumption (kJ / kWh), a, b, c represent coefficients of the target fitting function, and y represents unit load factor (%), represents the unit load in one day (MW), x max Represents the maximum unit load (MW) in a day. Perform fitting analysis on the data of the "heat consumption-unit load rate" fitting data set to obtain the fitting curve of "heat consumption-unit load rate", which is:

[0105] z=8.2543y 2 -30.347y+32.437;

[0106] Step S2: Segment the fitting curve according to the unit load size, and calculate the fitting degree of each unit load section respectively. Specifically, X0 represents the rated unit load of the generator set. In this embodiment, the fitting degrees R of the three sections (45% X0, 55% X0), (70% X0, 80% X0), and (90% X0, 100% X0) are 2 ≥0.8, the fitting degree R of the three sections (0%X0, 45%X0), (55%X0, 70%X0), (80%X0, 90%X0) 2 <0.8;

[0107] Step S3: Obtaining empirical value data of heat consumption corresponding to the range of (0%X0, 45%X0), (55%X0, 70%X0), (80%X0, 90%X0);

[0108] Step S4: forming a piecewise function of heat rate and unit load, the piecewise function includes a fitting curve segment and an empirical value segment, wherein the fitting curve segment includes: (45%X0, 55%X0), (70%X0, 80%X0), (90%X0, 100%X0) segments; the empirical value segment includes: (0%X0, 45%X0), (55%X0, 70%X0), (80%X0, 90%X0) segments; the piecewise function of heat rate and unit load is specifically:

[0109]

[0110] Among them, z 参考值 Indicates the reference value of heat consumption (kJ / kWh), x 实时 represents the real-time unit load (MW), y 当日 Indicates the unit load factor for the day, a = 8.2543, b = -30.347, c = 32.437;

[0111] (II) Real-time heat consumption analysis

[0112] Step S5: Obtain the real-time unit load x of the generator set on that day 实时 ;

[0113] Step S6: determining the range of the real-time unit load;

[0114] Step S7: Calculate the reference value of the heat rate according to the piecewise function of the heat rate and the unit load. Specifically, if the range of the real-time unit load is within the fitting curve of the piecewise function, first calculate the unit load rate y of the day according to the real-time unit load. 当日 , and then substitute the unit load rate of the day into the formula corresponding to the fitting curve of "heat rate-unit load rate", and the calculation result is the reference value of heat rate; if the section range of the real-time unit load is in the empirical value section of the piecewise function, then the empirical value is the reference value of heat rate;

[0115] Step S8: Obtain the real-time heat consumption z of the generator set on that day 实时 ; Compare the real-time value of the heat consumption with the reference value of the heat consumption, and adjust the parameters of the heat consumption equipment of the generator set or repair the heat consumption equipment according to the comparison result.

[0116] (III) Optimizing the piecewise function of heat consumption and unit load

[0117] Step S9: Obtain the real-time unit load x in the fitting curve segment within a period of time实时 、The unit load factor of the day y 当日 And real-time heat consumption 实时 , calculate the average value x corresponding to each segment of the fitting curve segment (45%X0, 55%X0), (70%X0, 80%X0), (90%X0, 100%X0) 平均 ,y 平均 、z 平均 , preferably a daily average value; adding the average value to the "heat consumption-unit load rate" fitting data set, and updating the coefficient of the "heat consumption-unit load rate" fitting curve.

[0118] Optionally, the real-time unit load, real-time cooling water temperature and corresponding real-time vacuum value data within the fitting curve segment can be directly added to the "heat consumption-unit load rate" fitting data set to update the coefficient of the "heat consumption-unit load rate" fitting curve.

[0119] As a further optimization of the above scheme, before forming the fitting data set, it also includes: eliminating data with large unit load fluctuations to reduce the impact of distorted data generated by load fluctuations on the fitting analysis. Specifically, all unit load data are sorted by time, the unit load obtained at a certain moment is set to x2, and the unit load obtained at the moment before x2 is set to x1, the difference between x2 and x1 is calculated, and the absolute value thereof is calculated; it is determined whether the absolute value is ≥ 6MW, and if so, the unit load x2 obtained at a certain moment and the data of its corresponding production index are eliminated; if not, the unit load x2 obtained at a certain moment and the data of its corresponding production index are added to the fitting data set.

[0120] In the actual use of the analysis method of the production index of a thermal power plant based on unit load segmentation, the technicians found that although the fitting degree of the fitting curve obtained by the fitting analysis is getting higher and higher as the data of the fitting data set increases, it is statistically found that the degree of improvement of the daily fitting degree is getting smaller and smaller. The technicians of the present invention carefully analyzed the reason and found that this may be because the equipment of the generator set ages with the extension of the use time, resulting in the equipment performance of the generator set being lower than the performance when it was just shipped, so that the actual parameter value of the production index cannot reach the size of the historical parameter value, which is manifested as the improvement of the daily fitting degree is getting smaller and smaller.

[0121] In view of the above problems, technicians have further improved the method of the present invention. Specifically, before step S8 "updating the coefficients of the fitting curve", the method also includes: eliminating historical data of the fitting data set with a time of two years or more.

[0122] Compared with the prior art, the present invention realizes the quantitative statistics of production indicators. The statistical reference values ​​of production indicators can change with the changes in unit load. It has the advantages of high general applicability and high precision, which is conducive to the crew's rapid and accurate response to the uneconomical operation, environmentally unfriendly operation and abnormal operation of the generator set. In addition, the analysis and optimization method of the present application, which first fits and then segments and then uses the segmented data for fitting, has the advantages of accurate calculation, convenient and fast data processing and high fitting degree of the fitting curve. The present invention also eliminates data with large unit load fluctuations, thereby reducing the impact of distorted data caused by load fluctuations on the fitting analysis. In addition, the present invention also eliminates historical data with a fitting data set time of two years or more, reducing the impact of aging of generator set equipment on the fitting accuracy.

[0123] The above-mentioned embodiments only express several implementation methods of the present invention, and the description is relatively specific and detailed, but it cannot be understood as limiting the scope of the invention patent. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, and the present invention is also intended to include these modifications and modifications.

Claims

1. A method for analyzing production indicators of a thermal power plant based on unit load segmentation, characterized in that: The following steps are involved: Step S1: Obtain the measured value of each production indicator and its corresponding unit load to form a fitting data set; Perform fitting analysis on the data of the fitting data set to obtain a fitting curve; Step S2: segmenting the fitting curve according to the load size of the unit; Step S3: Obtaining empirical value data of production indicators corresponding to the load sections of each unit; Step S4: forming a piecewise function of the reference value of the production index and the unit load according to the fitting curve and the empirical value, wherein the piecewise function is divided into a fitting curve segment and an empirical value segment; Step S5: obtaining the real-time unit load of the generator set on that day; Step S6: determining the range of the real-time unit load; Step S7: calculating the reference value of the production index according to the piecewise function of the reference value of the production index and the unit load; Step S8: obtaining the real-time value of the production index of the generator set on that day; comparing the real-time value of the production index with the reference value of the production index, and adjusting the parameters of the generator set or performing maintenance according to the comparison result.

2. The method for analyzing production indicators of a thermal power plant based on unit load segmentation according to claim 1 is characterized in that: In step S2, after the fitting curve is segmented according to the unit load size, the following step further includes: calculating the fitting degree R of each unit load segment respectively. 2 ; The step S3 specifically includes: obtaining the fitting degree R 2 The empirical value data of production indicators corresponding to the unit load range of <0.8; In step S4, the fitting curve segment is the fitting degree R 2 ≥0.8 unit load section; the empirical value section is the fitting degree R 2 Unit load section of <0.8; The piecewise function of the reference value of the production index and the unit load formed in step S4 is specifically: Among them, z 参考值 Indicates the reference value of the production index, R 2 Indicates the fitting degree of the fitting curve segment.

3. The method for analyzing production indicators of a thermal power plant based on unit load segmentation according to claim 2 is characterized in that: In step S1, the fitting data set also includes other influencing factors affecting the production index value besides the unit load; Step S5 also includes: acquiring real-time data of the influencing factor.

4. The method for analyzing production indicators of a thermal power plant based on unit load segmentation according to claim 3 is characterized in that: The fitting curve includes a fitting curve of "production index-unit load" or a fitting curve of "production index-unit load-influencing factor".

5. The method for analyzing production indicators of a thermal power plant based on unit load segmentation according to any one of claims 1 to 4, characterized in that: The optimization step of the fitting curve comprises: Step S9: Obtain the real-time unit load, real-time data of influencing factors and corresponding real-time value data of production indicators within a period of time, and calculate the average value of the real-time unit load, real-time data of influencing factors and corresponding real-time value data of production indicators corresponding to each section in the fitting curve segment; add the average value to the fitting data set and update the coefficient of the fitting curve.

6. The method for analyzing production indicators of a thermal power plant based on unit load segmentation according to claims 1-4 is characterized in that: The real-time unit load, the real-time data of the influencing factors and the corresponding real-time value data of the production index within the fitting curve segment are added to the fitting data set, and the coefficient of the fitting curve is updated.

7. The method for analyzing production indicators of a thermal power plant based on unit load segmentation according to claim 5 is characterized in that: Before forming the fitting data set in step S1, the method further includes: eliminating data with large unit load fluctuations.

8. The method for analyzing production indicators of a thermal power plant based on unit load segmentation according to claim 7 is characterized in that: The data for eliminating large load fluctuations of the unit specifically include: All unit load data are sorted by time, and the absolute value of the difference between the unit load obtained at a certain moment and the unit load at the moment before the unit load obtained at this moment is calculated; it is determined whether the absolute value is ≥6MW. If so, the data of the unit load obtained at this moment and its corresponding production index are eliminated; if not, the data of the unit load obtained at this moment and its corresponding production index are added to the fitting data set.

9. The method for analyzing production indicators of a thermal power plant based on unit load segmentation according to claim 7 is characterized in that: Before step S9 "updating the coefficients of the fitting curve", the method also includes: removing historical data of the fitting data set with a time of two years or more.

10. The method for analyzing production indicators of a thermal power plant based on unit load segmentation according to claim 9, characterized in that: The production indicators include water flow, vacuum, circulating water temperature rise, terminal difference, feed water temperature, water replenishment rate, main reheat steam temperature, exhaust gas temperature, exhaust gas oxygen content, fly ash combustibles, outlet CO content, condenser cleanliness, boiler efficiency, and coal consumption; the piecewise function is specifically: Among them, z 参考值 Indicates the reference value of the production index, x 实时 represents the real-time unit load, y 实时 represents the real-time data of the influencing factor; X0 represents the rated unit load of the generator set, a, b, c represent the coefficients of the piecewise function; y 当日 Indicates the unit load rate for the day.

Images