CN114759557A - Coal-fired unit AGC adjustment performance prediction method - Google Patents

Coal-fired unit AGC adjustment performance prediction method Download PDF

Info

Publication number
CN114759557A
CN114759557A CN202210452271.5A CN202210452271A CN114759557A CN 114759557 A CN114759557 A CN 114759557A CN 202210452271 A CN202210452271 A CN 202210452271A CN 114759557 A CN114759557 A CN 114759557A
Authority
CN
China
Prior art keywords
agc
set value
target load
coal
adjusting
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210452271.5A
Other languages
Chinese (zh)
Inventor
卢建刚
戴月
林玥廷
曾凯文
李世明
余志文
郭文鑫
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangdong Power Grid Co Ltd
Electric Power Dispatch Control Center of Guangdong Power Grid Co Ltd
Original Assignee
Guangdong Power Grid Co Ltd
Electric Power Dispatch Control Center of Guangdong Power Grid Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guangdong Power Grid Co Ltd, Electric Power Dispatch Control Center of Guangdong Power Grid Co Ltd filed Critical Guangdong Power Grid Co Ltd
Priority to CN202210452271.5A priority Critical patent/CN114759557A/en
Publication of CN114759557A publication Critical patent/CN114759557A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/004—Generation forecast, e.g. methods or systems for forecasting future energy generation
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/007—Arrangements for selectively connecting the load or loads to one or several among a plurality of power lines or power sources
    • H02J3/0075—Arrangements for selectively connecting the load or loads to one or several among a plurality of power lines or power sources for providing alternative feeding paths between load and source according to economic or energy efficiency considerations, e.g. economic dispatch
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/12—Circuit arrangements for ac mains or ac distribution networks for adjusting voltage in ac networks by changing a characteristic of the network load
    • H02J3/14—Circuit arrangements for ac mains or ac distribution networks for adjusting voltage in ac networks by changing a characteristic of the network load by switching loads on to, or off from, network, e.g. progressively balanced loading
    • H02J3/144—Demand-response operation of the power transmission or distribution network
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
    • H02J3/46—Controlling of the sharing of output between the generators, converters, or transformers
    • H02J3/466—Scheduling the operation of the generators, e.g. connecting or disconnecting generators to meet a given demand
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/10—Power transmission or distribution systems management focussing at grid-level, e.g. load flow analysis, node profile computation, meshed network optimisation, active network management or spinning reserve management
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]

Abstract

The invention discloses a method for predicting AGC (automatic gain control) adjusting performance of a coal-fired unit, which comprises the following steps: selecting historical working condition data of the coal-fired unit under a stable working condition within a preset time period according to historical working condition parameters of the coal-fired unit; identifying each AGC adjusting process without frequency modulation action according to the historical working condition data to obtain AGC performance indexes corresponding to each AGC adjusting process; inputting AGC performance indexes corresponding to all AGC adjusting processes and corresponding historical working condition parameters serving as sample data into a preset BP neural network for training; and obtaining a prediction result of the AGC (automatic gain control) adjustment performance of the coal-fired unit by combining the current operating condition parameters under the stable condition and the BP neural network. By adopting the method and the device, the AGC adjusting performance of the unit under different operating conditions is dynamically mastered, and the AGC adjusting performance is accurately predicted in advance by combining a neural network.

Description

Coal-fired unit AGC adjustment performance prediction method
Technical Field
The invention relates to the technical field of coal-fired unit control, in particular to a prediction method for AGC (automatic gain control) adjusting performance of a coal-fired unit.
Background
The peak-valley difference of the power load is continuously increased, the new energy is rapidly developed, and the power proportion of the new energy is continuously increased. Due to the randomness, intermittency and fluctuation of new energy sources, higher requirements are put on the load regulation performance of the traditional power supply. The coal-fired unit is used as a stabilizer and a peak shaving main unit of the power system, and the AGC (automatic gain control) adjusting performance of the coal-fired unit is directly related to the quality of electric energy and the reliable supply of the electric energy. However, the AGC adjustment performance of current coal-fired units is mainly tested by periodic performance tests, and typically only some typical conditions are manually selected for testing. The AGC adjusting performance of the coal-fired unit cannot be predicted based on the current actual operating condition, and data support cannot be provided for the power dispatching department to fully explore the load adjusting capacity of the coal-fired unit.
The AGC adjusting performance of the coal-fired unit is mainly detected by a regular performance test. The performance test is carried out by selecting some typical working conditions, and the test result can represent the AGC adjusting performance of the unit under the current working state. However, because the test interval period is generally long, the characteristics of the unit and the equipment slightly change along with time, so that the original test result is difficult to accurately reflect the AGC (automatic gain control) adjustment performance of the unit under the current working condition, and the AGC adjustment performance is not required to be predicted. In other words, the AGC adjustment performance detected by the prior art is less time-sensitive and may introduce some human error.
Disclosure of Invention
The embodiment of the invention provides a method for predicting the AGC (automatic gain control) adjusting performance of a coal-fired unit, which is used for dynamically mastering the AGC adjusting performance of the unit under different operating conditions by automatically identifying the AGC adjusting process of the coal-fired unit and accurately predicting the AGC adjusting performance in advance by combining a neural network.
In order to achieve the above object, a first aspect of the embodiments of the present application provides a method for predicting AGC adjustment performance of a coal-fired unit, including:
selecting historical working condition data of the coal-fired unit under a stable working condition within a preset time period according to the historical working condition parameters of the coal-fired unit;
identifying each AGC adjusting process without frequency modulation action according to the historical working condition data to obtain AGC performance indexes corresponding to each AGC adjusting process;
inputting AGC performance indexes corresponding to all AGC adjusting processes and corresponding historical working condition parameters serving as sample data into a preset BP neural network for training until a correction weight and a threshold of the BP neural network meet error requirements;
and obtaining a prediction result of the AGC adjusting performance of the coal-fired unit by combining the current operating condition parameters under the stable condition and the BP neural network.
In a possible implementation manner of the first aspect, the identifying, according to the historical operating condition data, each AGC adjusting process that does not include a frequency modulation action to obtain an AGC performance index corresponding to each AGC adjusting process specifically includes:
extracting AGC target load set value, actual load and adjustment instruction information in historical working condition data;
calculating the deviation between the AGC target load set value and the actual load;
judging the adjusting direction of the AGC adjusting process according to the value of the deviation and the dead zone; the dead zone is related to the rated capacity of the coal-fired unit;
judging the end time of the AGC adjusting process according to the AGC target load set value, the actual load and the dead zone;
the starting time of the AGC adjusting process is the sending time of the last adjusting instruction in the adjusting instruction information;
and calculating the response time, the adjusting speed and the adjusting precision of the AGC adjusting process according to the adjusting direction, the ending time and the starting time.
In a possible implementation manner of the first aspect, the determining, according to the value of the deviation and the dead zone, an adjustment direction of the AGC adjustment process specifically includes:
and the deviation between the AGC target load set value and the current load is a positive deviation, the absolute value of the deviation is greater than the dead zone, and the regulating direction of the AGC regulating process is load increase.
In a possible implementation manner of the first aspect, the determining an end time of the AGC adjustment process according to the AGC target load setting value, the actual load, and the dead zone specifically includes:
if the actual load is increased to the range of subtracting the dead zone from the AGC target load set value for the first time, and an adjusting instruction is received to adjust the AGC target load set value in the period, wherein the time when the AGC target load set value changes is the end time of the AGC adjusting process;
and if the actual load is increased to the range of subtracting the dead zone from the AGC target load set value for the first time and no adjusting instruction is received during the period to adjust the AGC target load set value, taking the end time of the last waveform of three continuous waveforms in the range of +/-dead zone of the AGC target load set value as the end time of the AGC adjusting process.
In a possible implementation manner of the first aspect, the determining, according to the AGC target load setting value, the actual load, and the dead zone, an end time of the AGC adjustment process includes:
and if the actual load does not rise to the range of subtracting the dead zone from the AGC target load set value, and an adjusting instruction is received to adjust the AGC target load set value in the period, wherein the adjusted AGC target load set value is larger than the actual load, and the end time of the last waveform of three continuous waveforms in the range of +/-the dead zone of the adjusted AGC target load set value is taken as the end time of the AGC adjusting process.
In a possible implementation manner of the first aspect, the determining, according to the value of the deviation and the dead zone, an adjustment direction of the AGC adjustment process specifically includes:
the deviation between the AGC target load set value and the current load is a negative deviation, the absolute value of the deviation is larger than the dead zone, and the regulating direction of the AGC regulating process is load reduction.
In a possible implementation manner of the first aspect, the determining an end time of the AGC adjustment process according to the AGC target load setting value, the actual load, and the dead zone specifically includes:
if the actual load is firstly reduced to the range of the AGC target load set value plus the dead zone, and an adjusting instruction is received during the period to adjust the AGC target load set value, wherein the time when the AGC target load set value changes is the end time of the AGC adjusting process;
and if the actual load is firstly reduced to the range of the AGC target load set value plus the dead zone, and no adjusting instruction is received during the period to adjust the AGC target load set value, taking the end time of the last waveform of three continuous waveforms in the range of the AGC target load set value plus or minus the dead zone as the end time of the AGC adjusting process.
In a possible implementation manner of the first aspect, the determining, according to the AGC target load setting value, the actual load, and the dead zone, an end time of the AGC adjustment process includes:
and if the actual load does not fall into the range of the AGC target load set value plus the dead zone, and an adjusting instruction is received to adjust the AGC target load set value in the period, wherein the adjusted AGC target load set value is smaller than the actual load, and the end time of the last waveform of three continuous waveforms in the range of the adjusted AGC target load set value plus or minus the dead zone is taken as the end time of the AGC adjusting process.
In a possible implementation manner of the first aspect, the determining process of the stable operating condition is:
in a preset time range, the parameter fluctuation ranges of actual load, environment temperature, main steam pressure, main steam temperature, reheated steam temperature, vacuum, water supply temperature, exhaust oxygen amount, hearth negative pressure, steam drum water level and fired coal heat value are in a preset range.
In a possible implementation manner of the first aspect, the obtaining, by combining the current operating condition parameter under the stable condition and the BP neural network, a prediction result of the AGC adjustment performance of the coal-fired unit specifically includes:
and inputting AGC target load set value, actual load, environment temperature, main steam pressure, main steam temperature, reheated steam temperature, vacuum, exhaust gas oxygen amount, hearth negative pressure, steam drum water level and fired coal heat value as input quantities into the trained BP neural network to obtain predicted response time, regulation rate and regulation precision.
Compared with the prior art, the coal-fired unit AGC regulation performance prediction method provided by the embodiment of the invention has the advantages that data preprocessing is firstly carried out on AGC historical data, and a normal operation time period is found out; then, automatically identifying each AGC adjusting process of the AGC unit based on a preset rule to obtain each AGC adjusting process identification under a normal working condition, and calculating the performance index of each AGC adjusting process; then, defining the unit working condition division parameters, and selecting the time period when the unit normally operates and the regulating process is normal to carry out BP neural network model training to obtain a sample database; and finally, predicting the AGC adjusting performance based on the current actual operation condition parameters. In the whole process, the AGC adjusting performance of the coal-fired unit under different operating conditions in historical data is extracted for analysis by identifying the AGC adjusting process of the coal-fired unit, and a BP neural network is trained; and finally, predicting the AGC real-time regulation performance according to the actual working condition, thereby providing data support for the improvement of the network source coordination capacity and the consumption of new energy electric power, and better playing the power generation regulation role of the coal-fired unit in the novel electric power system.
Drawings
Fig. 1 is a schematic flow chart of a method for predicting AGC adjustment performance of a coal-fired unit according to an embodiment of the present invention;
fig. 2 is a schematic diagram illustrating a process for identifying AGC adjustment processes of a coal-fired unit according to an embodiment of the present invention;
fig. 3 is a schematic flow chart of training a BP neural network according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, an embodiment of the present invention provides a method for predicting AGC adjustment performance of a coal-fired unit, including:
s10, selecting historical working condition data of the coal-fired unit under the stable working condition within a preset time period according to the historical working condition parameters of the coal-fired unit.
And S11, identifying each AGC adjusting process without frequency modulation action according to the historical working condition data, and obtaining AGC performance indexes corresponding to each AGC adjusting process.
And S12, inputting AGC performance indexes corresponding to the AGC adjusting processes and corresponding historical working condition parameters serving as sample data into a preset BP neural network for training until the correction weight and the threshold of the BP neural network meet the error requirement.
And S13, obtaining a prediction result of the AGC adjusting performance of the coal-fired unit by combining the current operating condition parameters under the stable condition and the BP neural network.
The main factors influencing the AGC adjusting performance of the coal-fired unit are as follows: the operation mode of the unit, the adjustment performance of DCS, the actual load, the heat value of coal entering a furnace, the number of running coal mills, the pressure of main steam, the temperature of reheated steam, the negative pressure of a hearth and other operation parameters. The influence factors are numerous, coal-fired power generation is a complex process, the operation parameters are thousands of, and the parameters are correlated and influenced with each other, so that the difficulty is increased for dividing the working condition for people; in addition, the characteristics of the unit and the equipment are changed along with the time, so that the uncertainty of the AGC regulation performance under the test working condition is increased.
In order to improve timeliness and accuracy of AGC (automatic gain control) adjusting performance, firstly, data preprocessing is carried out on historical data of AGC, and an alarm time interval is recorded; then, the AGC fixed value is compared with an empirical value or a historical value for analysis, automatic checking is realized, and an alarm time interval is recorded; then, automatically identifying each AGC adjusting process of the unit based on a certain rule, identifying the AGC adjusting process in the alarm time period, and calculating the performance index of each AGC adjusting process; then, defining the unit working condition division parameters, and selecting the time period when the unit normally operates and the regulating process is normal to perform model training to obtain a sample database; and finally, predicting the AGC adjusting performance based on the current actual operation condition parameters. The historical working condition data in the S10 refers to the working condition data of the coal-fired unit under the stable working condition after the alarm time period data are removed.
It should be noted that, the acquisition of the historical operating condition data requires that a fixed value parameter basic information table such as a load upper limit, a load lower limit, an adjustment rate upper limit and the like is established based on the characteristics of the unit, the fixed value parameters are subjected to rough range detection by using empirical values, and the fixed value parameters are identified and automatically checked according to earlier historical operating data of the unit. And the inspection is performed once every week, and the working condition data of the alarm period is deleted.
Generally speaking, the embodiment dynamically masters the AGC adjusting performance of the units under different operating conditions by automatically identifying the AGC adjusting process of the coal-fired unit, and predicts the AGC adjusting performance of the coal-fired unit in advance according to the BP neural network, thereby providing data support for the promotion of network source coordination capacity and the consumption of new energy electric power, and better playing the roles of a stabilizer and ballast stones of the coal-fired unit in a novel electric power system.
Referring to fig. 2, the identifying, according to the historical working condition data, each AGC adjustment process that does not include a frequency modulation action to obtain an AGC performance index corresponding to each AGC adjustment process specifically includes:
extracting AGC target load set value, actual load and adjustment instruction information in historical working condition data;
calculating the deviation between the AGC target load set value and the actual load;
judging the adjusting direction of the AGC adjusting process according to the value of the deviation and the dead zone; the dead zone is related to the rated capacity of the coal-fired unit;
judging the end time of the AGC adjusting process according to the AGC target load set value, the actual load and the dead zone;
the starting time of the AGC adjusting process is the sending time of the last adjusting instruction in the adjusting instruction information;
and calculating the response time, the adjusting speed and the adjusting precision of the AGC adjusting process according to the adjusting direction, the ending time and the starting time.
Illustratively, the determining the adjustment direction of the AGC adjustment process according to the value of the deviation and the dead zone specifically includes:
the deviation between the AGC target load set value and the current load is positive deviation, the absolute value of the deviation is larger than the dead zone, and the regulating direction of the AGC regulating process is load rising.
Illustratively, the determining the ending time of the AGC adjusting process according to the AGC target load setting value, the actual load and the dead zone specifically includes:
if the actual load is increased to the range of subtracting the dead zone from the AGC target load set value for the first time, and an adjusting instruction is received to adjust the AGC target load set value in the period, wherein the time when the AGC target load set value changes is the end time of the AGC adjusting process;
and if the actual load is increased to the range of subtracting the dead zone from the AGC target load set value for the first time and no adjusting instruction is received in the period to adjust the AGC target load set value, taking the end time of the last waveform of three continuous waveforms in the range of +/-dead zone of the AGC target load set value as the end time of the AGC adjusting process.
Illustratively, the determining the ending time of the AGC adjusting process according to the AGC target load setting value, the actual load and the dead zone specifically includes:
and if the actual load does not rise to the range of subtracting the dead zone from the AGC target load set value, and an adjusting instruction is received to adjust the AGC target load set value in the period, wherein the adjusted AGC target load set value is larger than the actual load, and the end time of the last waveform of three continuous waveforms in the range of +/-dead zone of the adjusted AGC target load set value is taken as the end time of the AGC adjusting process.
Illustratively, the determining the adjustment direction of the AGC adjustment process according to the value of the deviation and the dead zone specifically includes:
the deviation between the AGC target load set value and the current load is a negative deviation, the absolute value of the deviation is larger than the dead zone, and the regulating direction of the AGC regulating process is load reduction.
Exemplarily, the determining the ending time of the AGC adjusting process according to the AGC target load setting value, the actual load and the dead zone specifically includes:
if the actual load is firstly reduced to the range of the AGC target load set value plus the dead zone, and an adjusting instruction is received during the period to adjust the AGC target load set value, wherein the time when the AGC target load set value changes is the end time of the AGC adjusting process;
and if the actual load is firstly reduced to the range of the AGC target load set value plus the dead zone, and no adjusting instruction is received during the period to adjust the AGC target load set value, taking the end time of the last waveform of three continuous waveforms in the range of the AGC target load set value plus or minus the dead zone as the end time of the AGC adjusting process.
Illustratively, the determining the ending time of the AGC adjusting process according to the AGC target load setting value, the actual load and the dead zone specifically includes:
and if the actual load does not fall into the range of the AGC target load set value plus the dead zone, and an adjusting instruction is received to adjust the AGC target load set value in the period, wherein the adjusted AGC target load set value is smaller than the actual load, and the end time of the last waveform of three continuous waveforms in the range of the adjusted AGC target load set value plus or minus the dead zone is taken as the end time of the AGC adjusting process.
The precondition for identifying the AGC adjusting process is that a CCS coordination mode or AGC is in an input state, data which are not input in the CCS coordination mode are not distinguished, and the identifying process specifically comprises the following steps:
1. when the absolute value of the deviation between the AGC target load set value and the current load is within the dead zone (1% rated capacity of the unit), the actual load is not adjusted according to the AGC target load set value, namely, the AGC target load set value is not considered to be effective, and an AGC adjusting process is not triggered.
2. When the deviation between the AGC target load set value and the current load is positive deviation and is larger than the dead zone, if the actual load of the unit is increased to the range of subtracting the dead zone (1% rated capacity of the unit) from the AGC target load set value for the first time, and the AGC target load set value is updated in the period, the variation moment of the AGC target load set value is the ending moment of the AGC adjustment process; and if the AGC target load set value is not updated all the time, the last wave peak or wave trough is the end time of the AGC adjustment process when 3 continuous wave peaks or wave troughs are within +/-dead zones of the AGC target load set value.
3. Similarly, when the deviation between the AGC target load set value and the current load is a negative deviation and the absolute value is greater than the dead zone, if the actual load of the unit is initially reduced to the range of the AGC target load set value plus the dead zone (1% of rated capacity of the unit), and the AGC target load set value is updated in the period, the variation moment of the AGC target load set value is the ending moment of the AGC adjustment process; and if the AGC target load set value is not updated all the time, the last wave peak or wave trough is the end time of the AGC adjustment process when 3 continuous wave peaks or wave troughs are all within the range of the AGC target load set value +/-dead zone.
4. When the deviation between the AGC target load set value and the current load is positive deviation and is greater than the dead zone, if the actual load of the unit does not rise to the range of subtracting the dead zone from the AGC target load set value (1% of rated capacity of the unit), and the AGC target load set value is updated and is greater than the current actual load in the period, when 3 continuous wave crests or wave troughs are all within the range of the last AGC target load set value +/-dead zone, the last wave crest or wave trough is taken as the ending time of the AGC adjustment process.
5. Similarly, when the deviation between the AGC target load setting value and the current load is a negative deviation and the absolute value thereof is greater than the dead zone, if the actual load of the unit has not decreased to the range of the AGC target load setting value plus the dead zone (1% of the rated capacity of the unit), and the AGC target load setting value is updated and is smaller than the current actual load in the period, when 3 continuous wave crests or wave troughs are all within the range of the last AGC target load setting value plus or minus the dead zone, the last wave crest or wave trough is taken as the ending time of the AGC adjustment process.
6. When the absolute value of the deviation between the AGC target load set value and the current load is larger than the dead zone, in the process of increasing the load and in the range of not reaching the AGC target load set value minus the dead zone (1% rated capacity of the unit), or in the process of reducing the load and in the range of not reaching the AGC target load set value plus the dead zone (1% rated capacity of the unit), the AGC target load set value is updated and is respectively smaller than the current actual load or respectively larger than the current actual load, the time of sending a reverse instruction for the last time is taken as the starting time of the AGC regulation process, and the ending time is judged according to the 3 rd point or the 4 th point; if the command is repeatedly reversed in the adjusting process, the principle is also followed.
The following illustrates the backward instruction: if the 1 st set value of the AGC target load is larger than (actual load of the unit + dead zone), namely the current unit is in load-up, but the actual load of the unit is not yet raised to the 1 st set value of the AGC target load, the 2 nd set value of the AGC target load is sent out, and the value is smaller than the actual load of the unit, and the 2 nd set value of the AGC target load at the moment is a reverse instruction. In general, the load is increased, but the load is decreased without increasing to the target. Or the load is increased when the load is reduced but not reduced to the target.
The following illustrates the repeated reversal: if the 1 st set value of the AGC target load is larger than (actual load of the unit + dead zone), namely the current unit is in load-up, but the actual load of the unit is not yet raised to the 1 st set value of the AGC target load, the 2 nd set value of the AGC target load is sent out, and the value is smaller than the actual load of the unit, and the 2 nd set value of the AGC target load at the moment is a reverse instruction. In general, the load is increased, but the load is decreased without increasing to the target. Or the load is increased when the load is reduced but not reduced to the target. And repeating the process, namely, repeatedly reversing the instruction.
Illustratively, the steady state determination process includes:
in a preset time range, the parameter fluctuation ranges of actual load, environment temperature, main steam pressure, main steam temperature, reheated steam temperature, vacuum, water supply temperature, exhaust oxygen amount, hearth negative pressure, steam drum water level and fired coal heat value are in a preset range.
In the embodiment, abnormal condition data of the locking increasing mode, the locking decreasing mode, the main auxiliary machine tripping mode and the non-CCS mode are eliminated. Selecting actual load, environment temperature, main steam pressure, main steam temperature, reheated steam temperature, vacuum, water supply temperature, exhaust gas oxygen amount, hearth negative pressure, drum water level (drum furnace) and fired coal heat value as stable working condition judgment parameters, and screening working conditions by judging the parameter fluctuation range within 5 minutes. Generally speaking, the load variation range is +/-10 MW, the environment temperature, the main steam temperature, the reheat steam temperature and the feedwater temperature variation range are +/-5 ℃, the main steam pressure variation range is +/-1 MPa, the vacuum variation range is +/-1%, the exhaust gas oxygen quantity variation range is +/-1%, the hearth negative pressure variation range is +/-30 Pa, the drum water level variation range is +/-10 mm, the fired coal heat value variation range is +/-1 MJ/kg, and stable working conditions are identified.
Illustratively, the obtaining of the prediction result of the AGC adjustment performance of the coal-fired unit by combining the current operating condition parameters under the stable condition and the BP neural network specifically includes:
and inputting AGC target load set values, actual load, environment temperature, main steam pressure, main steam temperature, reheated steam temperature, vacuum, exhaust gas oxygen quantity, hearth negative pressure, steam drum water level and fired coal heat value as input quantities into the trained BP neural network to obtain predicted response time, regulation rate and regulation precision.
The BP neural network training process in this embodiment is shown in fig. 3: the method comprises the steps of taking an AGC target load set value, an actual load, an environment temperature, a main steam pressure, a main steam temperature, a reheated steam temperature, vacuum, exhaust gas oxygen quantity, hearth negative pressure, a steam drum water level (a steam drum furnace), a coal heat value of a fired coal, an AGC input state, a CCS input state and the number of running coal mills as input quantities, taking adjusting directions (load rising and load falling), response time, adjusting speed and adjusting precision as output quantities, constructing a three-layer BP neural network, wherein an input layer comprises 14 nodes, a hidden layer comprises 8 nodes, an output comprises 4 nodes, and a learning error is 0.01. Firstly, network initialization is carried out, and random numbers in an interval (-1, 1) are respectively assigned to each connection weight. Then, selecting an input sample, and calculating output and output errors of each layer; then calculating local gradients of each layer, and finally repeatedly correcting the weight and the threshold value to enable the error to meet the requirement.
Compared with the prior art, the coal-fired unit AGC regulation performance prediction method provided by the embodiment of the invention has the advantages that data preprocessing is firstly carried out on AGC historical data, and a normal operation time period is found out; then, automatically identifying each AGC adjusting process of the AGC unit based on a preset rule to obtain each AGC adjusting process identification under a normal working condition, and calculating the performance index of each AGC adjusting process; then, defining the unit working condition division parameters, and selecting the time period when the unit normally operates and the regulating process is normal to carry out BP neural network model training to obtain a sample database; and finally, predicting the AGC adjusting performance based on the current actual operation condition parameters. In the whole process, the AGC adjusting process of the coal-fired unit is identified, the AGC adjusting performance of the unit under different operating conditions in historical data is extracted for analysis, and then a BP neural network is trained; and finally, predicting the AGC real-time regulation performance according to the actual working condition, thereby providing data support for the improvement of the network source coordination capacity and the consumption of new energy electric power, and better playing the power generation regulation role of the coal-fired unit in the novel electric power system.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention.

Claims (10)

1. A coal-fired unit AGC adjustment performance prediction method is characterized by comprising the following steps:
selecting historical working condition data of the coal-fired unit under a stable working condition within a preset time period according to the historical working condition parameters of the coal-fired unit;
identifying each AGC adjusting process without frequency modulation action according to the historical working condition data to obtain AGC performance indexes corresponding to each AGC adjusting process;
inputting AGC performance indexes corresponding to each AGC adjusting process and corresponding historical working condition parameters serving as sample data into a preset BP neural network for training until a correction weight and a threshold of the BP neural network meet error requirements;
and obtaining a prediction result of the AGC adjusting performance of the coal-fired unit by combining the current operating condition parameters under the stable condition and the BP neural network.
2. The method for predicting the AGC adjustment performance of the coal-fired unit according to claim 1, wherein the identifying each AGC adjustment process not including a frequency modulation action according to the historical working condition data to obtain an AGC performance index corresponding to each AGC adjustment process specifically includes:
extracting AGC target load set value, actual load and adjustment instruction information in historical working condition data;
calculating the deviation between the AGC target load set value and the actual load;
judging the adjusting direction of the AGC adjusting process according to the value of the deviation and the dead zone; the dead zone is related to the rated capacity of the coal-fired unit;
judging the end time of the AGC adjusting process according to the AGC target load set value, the actual load and the dead zone;
the starting time of the AGC adjusting process is the sending time of the last adjusting instruction in the adjusting instruction information;
and calculating the response time, the adjusting speed and the adjusting precision of the AGC adjusting process according to the adjusting direction, the ending time and the starting time.
3. The method for predicting the AGC (automatic gain control) adjustment performance of the coal-fired unit according to claim 2, wherein the step of judging the adjustment direction of the AGC adjustment process according to the deviation value and the dead zone specifically comprises the following steps:
and the deviation between the AGC target load set value and the current load is a positive deviation, the absolute value of the deviation is greater than the dead zone, and the regulating direction of the AGC regulating process is load increase.
4. The method for predicting the AGC adjustment performance of the coal-fired unit according to claim 3, wherein the determining the end time of the AGC adjustment process according to the AGC target load setting value, the actual load, and the dead zone specifically includes:
if the actual load is increased to the range of subtracting the dead zone from the AGC target load set value for the first time, and an adjusting instruction is received to adjust the AGC target load set value in the period, wherein the time when the AGC target load set value changes is the end time of the AGC adjusting process;
and if the actual load is increased to the range of subtracting the dead zone from the AGC target load set value for the first time and no adjusting instruction is received during the period to adjust the AGC target load set value, taking the end time of the last waveform of three continuous waveforms in the range of +/-dead zone of the AGC target load set value as the end time of the AGC adjusting process.
5. The method for predicting the AGC adjustment performance of the coal-fired unit according to claim 3, wherein the determining the end time of the AGC adjustment process according to the AGC target load setting value, the actual load, and the dead zone specifically includes:
and if the actual load does not rise to the range of subtracting the dead zone from the AGC target load set value, and an adjusting instruction is received to adjust the AGC target load set value in the period, wherein the adjusted AGC target load set value is larger than the actual load, and the end time of the last waveform of three continuous waveforms in the range of +/-dead zone of the adjusted AGC target load set value is taken as the end time of the AGC adjusting process.
6. The method for predicting the AGC (automatic gain control) adjustment performance of the coal-fired unit according to claim 2, wherein the step of judging the adjustment direction of the AGC adjustment process according to the deviation value and the dead zone specifically comprises the following steps:
the deviation between the AGC target load set value and the current load is a negative deviation, the absolute value of the deviation is larger than the dead zone, and the regulating direction of the AGC regulating process is load reduction.
7. The method for predicting the AGC adjustment performance of the coal-fired unit according to claim 6, wherein the determining the end time of the AGC adjustment process according to the AGC target load setting value, the actual load, and the dead zone specifically includes:
if the actual load is firstly reduced to the range of the AGC target load set value plus the dead zone, and an adjusting instruction is received during the period to adjust the AGC target load set value, wherein the time when the AGC target load set value changes is the end time of the AGC adjusting process;
and if the actual load is firstly reduced to the range of the AGC target load set value plus the dead zone, and no adjusting instruction is received during the period to adjust the AGC target load set value, taking the end time of the last waveform of three continuous waveforms in the range of the AGC target load set value plus or minus the dead zone as the end time of the AGC adjusting process.
8. The method for predicting the AGC adjustment performance of the coal-fired unit according to claim 6, wherein the determining the end time of the AGC adjustment process according to the AGC target load setting value, the actual load, and the dead zone specifically includes:
and if the actual load does not fall into the range of the AGC target load set value plus the dead zone, and an adjusting instruction is received to adjust the AGC target load set value in the period, wherein the adjusted AGC target load set value is smaller than the actual load, and the end time of the last waveform of three continuous waveforms in the range of the adjusted AGC target load set value plus or minus the dead zone is taken as the end time of the AGC adjusting process.
9. The method for predicting the AGC (automatic gain control) regulation performance of the coal-fired unit according to claim 1, wherein the judgment process of the stable working condition is as follows:
in a preset time range, the parameter fluctuation ranges of actual load, environment temperature, main steam pressure, main steam temperature, reheated steam temperature, vacuum, water supply temperature, exhaust oxygen amount, hearth negative pressure, steam drum water level and fired coal heat value are in a preset range.
10. The method for predicting the AGC (automatic gain control) regulation performance of the coal-fired unit according to claim 1, wherein the step of obtaining the prediction result of the AGC regulation performance of the coal-fired unit by combining the current operating condition parameters under the stable condition and the BP neural network specifically comprises the following steps:
and inputting AGC target load set values, actual load, environment temperature, main steam pressure, main steam temperature, reheated steam temperature, vacuum, exhaust gas oxygen quantity, hearth negative pressure, steam drum water level and fired coal heat value as input quantities into the trained BP neural network to obtain predicted response time, regulation rate and regulation precision.
CN202210452271.5A 2022-04-27 2022-04-27 Coal-fired unit AGC adjustment performance prediction method Pending CN114759557A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210452271.5A CN114759557A (en) 2022-04-27 2022-04-27 Coal-fired unit AGC adjustment performance prediction method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210452271.5A CN114759557A (en) 2022-04-27 2022-04-27 Coal-fired unit AGC adjustment performance prediction method

Publications (1)

Publication Number Publication Date
CN114759557A true CN114759557A (en) 2022-07-15

Family

ID=82332806

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210452271.5A Pending CN114759557A (en) 2022-04-27 2022-04-27 Coal-fired unit AGC adjustment performance prediction method

Country Status (1)

Country Link
CN (1) CN114759557A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115239200A (en) * 2022-08-31 2022-10-25 华能莱芜发电有限公司 Unit load comprehensive frequency modulation method and system based on network source cooperation

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115239200A (en) * 2022-08-31 2022-10-25 华能莱芜发电有限公司 Unit load comprehensive frequency modulation method and system based on network source cooperation
CN115239200B (en) * 2022-08-31 2023-12-01 华能莱芜发电有限公司 Network source cooperation-based unit load comprehensive frequency modulation method and system

Similar Documents

Publication Publication Date Title
CN107526292B (en) A method of the regulation ammonia spraying amount based on inlet NOx concentration prediction
CN110263395A (en) The power plant's denitration running optimizatin method and system analyzed based on numerical simulation and data
CN113433911B (en) Accurate control system and method for ammonia spraying of denitration device based on accurate concentration prediction
CN112180851B (en) Method and device for controlling production line, computer equipment and storage medium
CN114759557A (en) Coal-fired unit AGC adjustment performance prediction method
CN112285287A (en) Flue gas online monitoring data preprocessing method
CN112383045B (en) Transient stability out-of-limit probability calculation method and device for new energy power generation uncertainty
CN113440990A (en) EMD-LSTM based outlet SO2Concentration prediction method
CN111242469B (en) Method and system for coupling ultralow emission and high-efficiency energy-saving operation of boiler or kiln
CN112215464A (en) Prediction balance scheduling system for blast furnace gas under multiple working conditions
CN112576326B (en) Sliding pressure optimal operation control method, device and equipment for thermal power generating unit
CN111008735B (en) Predictive emissions management system and method
US6230495B1 (en) Method for optimizing fossil-fueled power stations
CN117637046A (en) Converter TSO prediction method
CN112023650A (en) Full-working-condition self-adaptive pH adjusting system
CN116739147A (en) BIM-based intelligent energy consumption management and dynamic carbon emission calculation combined method and system
CN103714262B (en) A kind of thermal technology's soft-sensing model update method based on buffer stopper timing Design
CN115809732A (en) Distributed photovoltaic power generation power prediction method and system
CN111174824B (en) Control platform that acid mist discharged
Jin et al. A causal model-based scheduling approach for coke oven gas system in steel industry
CN114742300B (en) Denitration device boiler flue resistance early warning method and system
CN112107977A (en) Automatic pH value adjusting method, system and device based on desulfurization system
CN112819224A (en) Unit output prediction and confidence evaluation method based on deep learning fusion model
CN114413249B (en) Data analysis method for power station boiler efficiency benchmarking optimization
CN112287535B (en) Method and system for determining minimum operation load of thermal power generating unit

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination