Method for improving kp value of agc control of power plant

Abstract

The application discloses a method for improving the AGC control KP value of a power plant, and comprises the following steps: step one, based on the coordinated control system of the power plant, establishing a short-term load precision prediction module, a regulation speed improvement module, a regulation precision improvement module and a response time module; step two, using the historical data of the load of the power plant for prediction; step three, adjusting the feedforward amount of the primary air of the boiler according to the load predicted by the short-term load precision prediction module; step four, increasing the load deviation of the main control of the boiler and simultaneously reducing the insensitive area; and step five, adjusting the opening of the main steam valve and the regulating valve in the steam turbine unit. The method effectively shortens the combustion hysteresis, adjusts the hysteresis of the air volume and the coal volume, realizes the rapid tracking of the main steam pressure of the steam turbine unit to adapt to the needs of the continuous change of the load, reduces the deviation between the actual load deviation and the dispatching instruction, maintains the stability of the main steam pressure of the steam turbine unit, reduces the load response time and prevents the regulation from being too large, and improves the stability of the overall operation of the power plant.

Description

TECHNICAL FIELD

[0001] The present application relates to power plant AGC control KP value technical field, more particularly to a kind of method for improving power plant AGC control KP value. BACKGROUND

[0002] The generation and perfection of automatic generation control technology promote the improvement of system reliability and economy.The English name of automatic generation control technology is Automatic Generation Control, abbreviated as AGC.Due to the fluctuation of power grid load, the situation that power generation and power consumption cannot be balanced synchronously often occurs, which will eventually lead to the collapse of system frequency and voltage.The emergence of AGC technology eliminates the imbalance between power generation and power consumption in real time.AG system uses computer system and feedback control to form a closed-loop generation control system according to the fluctuation of load.AGC is divided into decision control layer and instruction execution layer according to function layer.For the generator unit of power plant side, the target load of unit is given according to the needs of power grid, and the power plant side automatically responds to the change of target load at the required rate.The input of AGC system makes great changes in the peak shaving mode of thermal power unit, which often requires the unit to quickly respond to the large amplitude change of target load within a specified time, which puts high requirements on the automatic regulation quality of generator unit.

[0003] According to the "Implementation Details of North China Regional Power Plant Grid Connection Operation Management" and "Implementation Details of North China Regional Grid Connection Power Plant Auxiliary Service Management" of the Electricity Regulatory Commission in North China Grid, referred to as "two rules", the Kp parameter in the "two rules" is a comprehensive evaluation index of AGC function.Shandong Electric Power Dispatching and Control Center publishes the ranking of AGC performance of each unit according to Kp parameter every month, and gives economic compensation of tens of thousands of yuan to the unit with high Kp parameter every month, and carries out the same penalty assessment to the unit with low Kp parameter every month.

[0004] Due to the large lag of power plant boiler and steam turbine unit regulation, the regulation speed of steam turbine unit side is faster than that of power plant boiler side, which causes the regulation mismatch of steam turbine unit and power plant boiler and affects Kp value.Therefore, how to effectively control KP value is a problem to be solved by those skilled in the art. SUMMARY

[0005] Therefore, the present application provides a method for improving power plant AGC control KP value, which effectively controls KP value.

[0006] To achieve the above purpose, the present application provides the following technical scheme:

[0007] Preferably, the above method for improving power plant AGC control KP value comprises:

[0008] Step one, based on the power plant coordinated control system, a short-term load accurate prediction module, a regulation speed improvement module, a regulation accuracy improvement module and a response time module are established to improve the regulation rate K1, the regulation accuracy K2 and the response time K3, and then control the KP value, wherein KP = K1*K2*K3;

[0009] Step two, the short-term load accurate prediction module uses the historical data of the power plant load for prediction;

[0010] Step three, the regulation speed improvement module adjusts the feedforward amount of the boiler primary air according to the load predicted by the short-term load accurate prediction module to improve the regulation rate K1;

[0011] Step four, the regulation accuracy improvement module increases the load deviation of the boiler master control, when the load changes, the load deviation signal in the boiler master control and the pressure deviation of the steam turbine unit are calculated together, and the set value of the steam turbine unit pressure is modified according to the calculation result, the coal quantity and air quantity entering the boiler are changed; at the same time, the insensitive area is reduced, and the regulation accuracy K2 is improved;

[0012] Step five, based on the power plant coordinated control system and the steam turbine digital electro-hydraulic control system, the steam turbine unit master control PID adjusts the opening of the main steam valve and the throttle valve in the steam turbine unit according to the deviation value of the given load and the actual power of the steam turbine unit, and increases the feedforward and pressure correction loop in the adjustment process to improve the response time K3.

[0013] Preferably, the short-term load accurate prediction module uses the historical data of the power plant load for prediction, including:

[0014] Step one, the historical data of the power plant load is processed for abnormal data and normalized;

[0015] Step two, determine the optimization parameters in the LSTM model: the number of neurons L1, the learning rate ε and the training iteration times K, as the optimization variables of the PSO algorithm, and determine the optimization range of the optimization parameters respectively;

[0016] Step three, initialize the parameters of the PSO algorithm, including: the initial speed, position, learning weight, training times and scale of the PSO particle;

[0017] Step four, determine the fitness function of the PSO particle;

[0018] Step five, compare the fitness value of the PSO particle;

[0019] Step six, judging whether the maximum iteration number is reached, when the maximum iteration number is reached, the optimal parameters obtained are transmitted to the LSTM model, and training and prediction are performed, when the maximum iteration number is not reached, step five is executed.

[0020] Preferably, the abnormal data processing and normalization processing of the historical data of the power plant load comprises:

[0021] Step one, unreasonable values and null values in the historical data are replaced by average values;

[0022] Step two, data with too large value difference in the historical data is eliminated for normalization processing, the maximum value in the historical data is set as Xmax, the minimum value is set as Xmin, Xmax and Xmin are linearly transformed to a value Xnorm between intervals [0, 1] through the historical data X, and the calculation formula is:

[0023]

[0024] Preferably, the fitness function of the PSO particle comprises:

[0025] Step one, the mean absolute percentage error MAPE is calculated, and the formula is:

[0026] Wherein, Yactual is the actual value of the power plant load, Ypre is the predicted value of the power plant load, and n is the number of power plant load prediction points;

[0027] Step two, the fitness function f(x) = min(MAPE) is set.

[0028] Preferably, the speed adjustment and improvement module adjusts the feedforward amount of the boiler primary air according to the load predicted by the short-term load accurate prediction module, comprising:

[0029] The load A predicted by the short-term load accurate prediction module is used to adjust the feedforward amount of the boiler primary air, a preset load matrix A0 is set, A0 includes A1, A2, A3, and A4, wherein A1 is the first preset load, A2 is the second preset load, A3 is the third preset load, and A4 is the fourth preset load, and A1>A2>A3>A4;

[0030] The speed adjustment and improvement module adjusts the feedforward amount B of the boiler primary air according to the load A predicted by the short-term load accurate prediction module, a preset feedforward amount matrix B0 includes B1, B2, B3, and B4, wherein B1 is the first preset feedforward amount, B2 is the second preset feedforward amount, B3 is the third preset feedforward amount, and B4 is the fourth preset feedforward amount, and B1>B2>B3>B4;

[0031] The regulating speed improving module adjusts the feedforward amount of the boiler primary air according to the relationship between the load A predicted by the short-term accurate prediction module and the feedforward amount B of the boiler primary air.

[0032] When A1≥A>A2, the regulating speed improving module adjusts the feedforward amount of the boiler primary air to the first preset feedforward amount B1.

[0033] When A2≥A>A3, the regulating speed improving module adjusts the feedforward amount of the boiler primary air to the second preset feedforward amount B2.

[0034] When A3≥A>A4, the regulating speed improving module adjusts the feedforward amount of the boiler primary air to the third preset feedforward amount B3.

[0035] When A4≥A, the regulating speed improving module adjusts the feedforward amount of the boiler primary air to the fourth preset feedforward amount B4.

[0036] Preferably, the load deviation signal in the boiler master control and the turbine unit pressure deviation are jointly calculated, and the set value of the turbine unit pressure is modified according to the calculation result, comprising:

[0037] The calculation result C of the joint calculation of the load deviation signal in the boiler master control and the turbine unit pressure deviation, the preset calculation result matrix C0 is set to include C1, C2, C3, and C4, wherein C1 is the first preset calculation result, C2 is the second preset calculation result, C3 is the third preset calculation result, and C4 is the fourth preset calculation result, and C1>C2>C3>C4;

[0038] The regulating precision improving module modifies the set value D of the turbine unit pressure according to the calculation result C of the joint calculation of the load deviation signal in the boiler master control and the turbine unit pressure deviation, and the preset set value matrix D0 includes D1, D2, D3, and D4, wherein D1 is the first preset set value, D2 is the second preset set value, D3 is the third preset set value, and D4 is the fourth preset set value, and D1>D2>D3>D4;

[0039] The regulating precision improving module modifies the set value D of the turbine unit pressure according to the relationship between the calculation result C of the joint calculation of the load deviation signal in the boiler master control and the turbine unit pressure deviation and the set value D of the turbine unit pressure.

[0040] When C1≥C>C2, the regulating precision improving module modifies the set value of the turbine unit pressure to the first preset set value D1.

[0041] When C2≥C>C3, the regulation accuracy improvement module modifies the set value of the turbine unit pressure to the second preset set value D2;

[0042] When C3≥C>C4, the regulation accuracy improvement module modifies the set value of the turbine unit pressure to the third preset set value D3;

[0043] When C4≥C, the regulation accuracy improvement module modifies the set value of the turbine unit pressure to the fourth preset set value D4.

[0044] Preferably, the turbine unit main control PID adjusts the opening of the main valve and the governing valve in the turbine unit according to the deviation value of the turbine unit given load and the actual power, comprising:

[0045] The turbine unit main control PID adjusts the opening F of the main valve and the governing valve in the turbine unit according to the deviation value E of the turbine unit given load and the actual power, and the preset opening matrix F0 includes F1, F2, F3, and F4, wherein F1 is the first preset set value, F2 is the second preset set value, F3 is the third preset set value, and F4 is the fourth preset set value, and F1>F2>F3>F4;

[0046] The turbine unit main control PID adjusts the opening F of the main valve and the governing valve in the turbine unit according to the relationship between the deviation value E of the turbine unit given load and the actual power and the opening F of the main valve and the governing valve in the turbine unit;

[0047]

[0048] When E1≥E>E2, the turbine unit main control PID adjusts the opening of the main valve and the governing valve in the turbine unit to the first preset opening F1;

[0049] When E2≥E>E3, the turbine unit main control PID adjusts the opening of the main valve and the governing valve in the turbine unit to the second preset opening F2;

[0050] When E3≥E>E4, the turbine unit main control PID adjusts the opening of the main valve and the governing valve in the turbine unit to the third preset opening F3;

[0051] When E4≥E, the turbine unit main control PID adjusts the opening of the main valve and the governing valve in the turbine unit to the fourth preset opening F4.

[0052] ​Via the technical solution, compared with the prior art, the beneficial effects of the present application are:

[0053] 1. By adding short-term load precise prediction module, regulation speed improvement module, regulation precision improvement module and response time improvement module in the original power plant coordinated control system, the KP value is controlled, the combustion hysteresis is effectively shortened, the hysteresis of air volume and coal volume is adjusted, the rapid tracking of the main steam pressure of the steam turbine unit is realized to adapt to the continuous change of the load, the deviation between the actual load deviation and the dispatching instruction is reduced, the contradiction between the fast action of the steam turbine unit in the power plant coordinated control system and the large inertia of the boiler is solved, the stability of the main steam pressure of the steam turbine unit is maintained, the load response time is reduced to prevent over-regulation, the boiler heat meets the instantaneous demand of the steam turbine unit, the load control performance of the unit is effectively improved, the rapidity and accuracy of the load regulation response of the thermal power unit to the dispatching load are ensured, the peak regulation capacity is ensured, and the stability of the overall operation of the power plant is improved.

[0054] 2. By setting the PSO-LSTM short-term load precise prediction module, the method of combining PSO and LSTM neural network is to take the three key hyperparameters (neuron number L1, learning rate epsilon and training iteration number k) of LSTM as the optimization variables of the PSO particles, the speed and position of the particles are updated, so that the fitness value of the load prediction is minimized, and the better model parameters are obtained. BRIEF DESCRIPTION OF DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of the provided drawings.

[0056] Figure 1 The accompanying drawings are schematic diagrams of the overall structure of the present application.

[0057] Figure 2 The accompanying drawings are schematic diagrams of the short-term load precise prediction model of the present application. DETAILED DESCRIPTION

[0058] The technical solutions in the embodiments of the present application will be described in detail below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0059] As Figure 1As shown, the embodiment of the present application discloses a method for improving the AGC control KP value of a power plant, comprising:

[0060] Step one, based on the coordinated control system of the power plant, a short-term load accurate prediction module, a regulation speed improvement module, a regulation accuracy improvement module and a response time module are established to improve the regulation rate K1, the regulation accuracy K2 and the response time K3, and then control the KP value, wherein KP=K1*K2*K3;

[0061] Step two, the short-term load accurate prediction module uses the historical data of the power plant load for prediction;

[0062] Step three, the regulation speed improvement module adjusts the feedforward amount of the primary air of the boiler according to the load predicted by the short-term load accurate prediction module to improve the regulation rate K1;

[0063] Step four, the regulation accuracy improvement module increases the load deviation of the boiler master control, when the load changes, the load deviation signal in the boiler master control and the pressure deviation of the steam turbine unit are jointly calculated, the set value of the steam turbine unit pressure is modified according to the calculation result, the coal quantity and air quantity entering the boiler are changed; at the same time, the insensitive area is reduced, and the regulation accuracy K2 is improved;

[0064] Step five, based on the coordinated control system of the power plant and the digital electro-hydraulic control system of the steam turbine, the steam turbine unit master control PID adjusts the opening of the main steam valve and the regulating valve in the steam turbine unit according to the deviation value of the given load and the actual power of the steam turbine unit, increases the feedforward and pressure correction loop in the regulation process, and improves the response time K3.

[0065] Wherein, the regulation rate K1 is the rate of the steam turbine unit responding to the set point instruction, the calculation formula is: K1=2-V / V1, wherein V is the regulation rate, and V1 is the reference value;

[0066] Wherein, the regulation accuracy K2 is the difference between the actual output and the set point output after the steam turbine unit responds stably, the calculation formula is: K2=2-△P / 0.01PN, wherein △P is the unit regulation deviation, and PN is the rated capacity of the steam turbine unit;

[0067] Wherein, the response time K3 is the time for the steam turbine unit to cross the regulation dead zone on the basis of the original output point after the energy management system issues an instruction, the calculation formula is: K3=2-△T / TN, wherein △T is the actual response time of the steam turbine unit, and TN is the original output point;

[0068] The beneficial effects of the above embodiment are that: by adding the short-term load precise prediction module, the regulation speed improvement module, the regulation precision improvement module, and the response time improvement module in the original power plant coordinated control system, the KP value is controlled; the combustion hysteresis and the hysteresis of the air volume and the coal volume are effectively shortened, the rapid tracking of the main steam pressure of the steam turbine unit is realized to adapt to the needs of the continuous change of the load, the deviation between the actual load deviation and the dispatching instruction is reduced, the contradiction between the fast action of the governing valve of the steam turbine unit in the power plant coordinated control system and the large inertia of the boiler is solved, the stability of the main steam pressure of the steam turbine unit is maintained, the load response time is reduced to prevent over-regulation, the boiler heat meets the instantaneous demand of the steam turbine unit, the load control performance of the unit is effectively improved, the rapidity and accuracy of the load regulation response of the thermal power unit to the dispatching load are ensured, the peak regulation capacity of the thermal power unit is ensured, and the stability of the overall operation of the power plant is improved.

[0069] As shown in Figure 2 In one embodiment, the short-term load precise prediction module uses the historical data of the power plant load for prediction, including:

[0070] Step one: performing abnormal data processing and normalization processing on the historical data of the power plant load;

[0071] Step two: determining the optimization parameters in the LSTM model: the number of neurons L1, the learning rate ε, and the training iteration number K, as the optimization variables of the PSO algorithm, and respectively determining the optimization range of the optimization parameters;

[0072] Step three: initializing the parameters of the PSO algorithm, including: the initial speed, position, learning weight, training number, and scale of the PSO particle;

[0073] Step four: determining the fitness function of the PSO particle;

[0074] Step five: comparing the fitness values of the PSO particles;

[0075] Step six: judging whether the maximum iteration number is reached, when the maximum iteration number is reached, the optimal parameters obtained are transmitted to the LSTM model, and training and prediction are performed, and when the maximum iteration number is not reached, step five is executed.

[0076] The beneficial effects of the above embodiment are that: by setting the PSO-LSTM short-term load precise prediction module, the method of combining PSO and LSTM neural network is to take the three key hyperparameters of LSTM (the number of neurons L1, the learning rate ε, and the training iteration number k) as the optimization variables of the PSO particle, and by updating the speed and position of the particle, the fitness value of the load prediction is minimized, and more optimal model parameters are obtained.

[0077] In one embodiment, the historical data of the power plant load is subjected to abnormal data processing and normalization processing, including:

[0078] Step one, unreasonable values and null values in the historical data are replaced by average values;

[0079] Step two, data with too large value difference in the historical data is subjected to normalization processing, setting the maximum value in the historical data as Xmax and the minimum value as Xmin, Xmax and Xmin are linearly converted to values between interval [0, 1] through a historical data X, and the calculation formula is:

[0080]

[0081] The above embodiment has the beneficial effect that the abnormal data and data with too large difference in the historical data are processed, and the accuracy of the load prediction is improved.

[0082] In one embodiment, the fitness function of the PSO particle is determined, including:

[0083] Step one, the mean absolute percentage error MAPE is calculated, and the formula is:

[0084] Wherein, Yactual is the actual value of the power plant load, Ypre is the predicted value of the power plant load, and n is the number of power plant load prediction points;

[0085] Step two, the fitness function is set as f(x) = min(MAPE).

[0086] In one embodiment, the speed adjustment and improvement module adjusts the feedforward amount of the boiler primary air according to the load predicted by the short-term load accurate prediction module, including:

[0087] The load A predicted by the short-term load accurate prediction module is used to adjust the feedforward amount of the boiler primary air, a preset load matrix A0 is set, A0 includes A1, A2, A3, and A4, wherein A1 is the first preset load, A2 is the second preset load, A3 is the third preset load, and A4 is the fourth preset load, and A1>A2>A3>A4;

[0088] The speed adjustment and improvement module adjusts the feedforward amount B of the boiler primary air according to the load A predicted by the short-term load accurate prediction module, and a preset feedforward amount matrix B0 includes B1, B2, B3, and B4, wherein B1 is the first preset feedforward amount, B2 is the second preset feedforward amount, B3 is the third preset feedforward amount, and B4 is the fourth preset feedforward amount, and B1>B2>B3>B4;

[0089] The adjusting speed improving module adjusts the feedforward amount of the boiler primary air according to the relationship between the load A predicted by the short-term accurate prediction module and the feedforward amount B of the boiler primary air.

[0090] When A1≥A>A2, the adjusting speed improving module adjusts the feedforward amount of the boiler primary air to the first preset feedforward amount B1.

[0091] When A2≥A>A3, the adjusting speed improving module adjusts the feedforward amount of the boiler primary air to the second preset feedforward amount B2.

[0092] When A3≥A>A4, the adjusting speed improving module adjusts the feedforward amount of the boiler primary air to the third preset feedforward amount B3.

[0093] When A4≥A, the adjusting speed improving module adjusts the feedforward amount of the boiler primary air to the fourth preset feedforward amount B4.

[0094] The above embodiment has the beneficial effect that accurate primary air feedforward amount is given according to different predicted loads, and the accuracy of the adjusting speed is improved.

[0095] In one embodiment, the load deviation signal in the boiler master control and the turbine unit pressure deviation are jointly calculated, and the set value of the turbine unit pressure is modified according to the calculation result, including:

[0096] The calculation result C of the joint calculation of the load deviation signal in the boiler master control and the turbine unit pressure deviation, the preset calculation result matrix C0 is set to include C1, C2, C3, and C4, wherein C1 is the first preset calculation result, C2 is the second preset calculation result, C3 is the third preset calculation result, and C4 is the fourth preset calculation result, and C1>C2>C3>C4;

[0097] The adjusting precision improving module modifies the set value D of the turbine unit pressure according to the calculation result C of the joint calculation of the load deviation signal in the boiler master control and the turbine unit pressure deviation, and the preset set value matrix D0 includes D1, D2, D3, and D4, wherein D1 is the first preset set value, D2 is the second preset set value, D3 is the third preset set value, and D4 is the fourth preset set value, and D1>D2>D3>D4;

[0098] The adjusting precision improving module modifies the set value D of the turbine unit pressure according to the relationship between the calculation result C of the joint calculation of the load deviation signal in the boiler master control and the turbine unit pressure deviation and the set value D of the turbine unit pressure.

[0099] When C1≥C>C2, the adjusting precision improving module modifies the set value of the turbine unit pressure to the first preset set value D1.

[0100] When C2≥C>C3, the regulating accuracy improvement module modifies the set value of the pressure of the steam turbine unit to a second preset set value D2;

[0101] When C3≥C>C4, the regulating accuracy improvement module modifies the set value of the pressure of the steam turbine unit to a third preset set value D3;

[0102] When C4≥C, the regulating accuracy improvement module modifies the set value of the pressure of the steam turbine unit to a fourth preset set value D4.

[0103] The above embodiment has the beneficial effect that the set value of the pressure of the steam turbine unit is modified according to different calculation results, and the accuracy of the regulating accuracy is improved.

[0104] In one embodiment, the steam turbine unit main control PID adjusts the opening of the main valve and the governing valve in the steam turbine unit according to the deviation value of the given load and the actual power of the steam turbine unit, including:

[0105] The steam turbine unit main control PID adjusts the opening F of the main valve and the governing valve in the steam turbine unit according to the deviation value E of the given load and the actual power of the steam turbine unit, and the preset deviation matrix E0 includes E1, E2, E3, and E4, wherein E1 is a first preset deviation value, E2 is a second preset deviation value, E3 is a third preset deviation value, and E4 is a fourth preset deviation value, and E1>E2>E3>E4;

[0106] The steam turbine unit main control PID adjusts the opening F of the main valve and the governing valve in the steam turbine unit according to the deviation value E of the given load and the actual power of the steam turbine unit, and the preset opening matrix F0 includes F1, F2, F3, and F4, wherein F1 is a first preset set value, F2 is a second preset set value, F3 is a third preset set value, and F4 is a fourth preset set value, and F1>F2>F3>F4;

[0107] The steam turbine unit main control PID adjusts the opening F of the main valve and the governing valve in the steam turbine unit according to the relationship between the deviation value E of the given load and the actual power of the steam turbine unit and the opening F of the main valve and the governing valve in the steam turbine unit;

[0108] When E1≥E>E2, the steam turbine unit main control PID adjusts the opening of the main valve and the governing valve in the steam turbine unit to a first preset opening F1;

[0109] When E2≥E>E3, the steam turbine unit main control PID adjusts the opening of the main valve and the governing valve in the steam turbine unit to a second preset opening F2;

[0110] When E3≥E>E4, the steam turbine unit main control PID adjusts the opening of the main valve and the governing valve in the steam turbine unit to a third preset opening F3;

[0111] When E4≥E, the main control PID of the steam turbine unit adjusts the opening of the main steam valve and the governing valve in the steam turbine unit to the fourth preset opening F4.

[0112] The beneficial effect of the above-mentioned embodiments is that the opening of the main steam valve and the governing valve in the steam turbine unit is adjusted according to different deviation values, thereby reducing the response time of the steam turbine unit.

[0113] The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the difference from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0114] The above description of disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to the embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for increasing the AGC control KP value of a power plant, characterized by, The utility model relates to a kind of power plant coordinated control system and turbine digital electro-hydraulic control system based on short-term load precision prediction module, regulation speed promotion module, regulation accuracy promotion module and response time module, which are established based on power plant coordinated control system, to adjust rate K1, regulation accuracy K2 and response time K3 are promoted, and then control KP value, wherein KP=K1*K2*K3. Step two, the short-term load precision prediction module uses the historical data of power plant load for prediction. Step three, the regulation speed promotion module adjusts the feedforward amount of boiler primary air according to the load predicted by the short-term load precision prediction module, and improves the regulation speed K1. Step four, the regulation accuracy promotion module increases load deviation to the boiler master control, when load changes, load deviation signal in the boiler master control and turbine unit pressure deviation are calculated together, and the set value of turbine unit pressure is modified according to the calculation result, the amount of coal and air entering the boiler are changed; at the same time, the insensitive area is reduced, and the regulation accuracy K2 is improved. Step five, based on the power plant coordinated control system and the turbine digital electro-hydraulic control system, the turbine unit master control PID adjusts the opening of the main steam valve and the regulating valve in the turbine unit according to the deviation value of the given load and the actual power of the turbine unit, and increases the feedforward and pressure correction loop in the adjustment process to improve the response time K3. The short-term load precision prediction module uses the historical data of power plant load for prediction, including: Step one, abnormal data processing and normalization processing are performed on the historical data of power plant load. Step two, determine the optimization parameters in the LSTM model: the number of neurons L1, the learning rate ε and the training iteration number K as the optimization variables of PSO algorithm, and determine the optimization range of the optimization parameters respectively. ​ Step three, initialize the parameters of the PSO algorithm, including: the initial velocity of the PSO particle, the position, the learning weight, the training times and the scale; Step four, determine the fitness function of the PSO particle; Step five, compare the fitness value of the PSO particle; Step six, judge whether the maximum iteration number is reached, when the maximum iteration number is reached, the optimal parameter obtained is transmitted to the LSTM model, and training and prediction are performed, when the maximum iteration number is not reached, the step five is executed.

2. The method for improving the AGC control KP value of a power plant according to claim 1, characterized in that, The abnormal data processing and normalization processing of the historical data of the power plant load, comprising: Step one, replacing the unreasonable values and null values in the historical data with the average value; Step two, eliminating the data with too large value difference in the historical data for normalization processing, setting the maximum value in the historical data as Xmax, the minimum value as Xmin, and Xmax and Xmin through a linear transformation of the historical data X to the value Xnorm between the interval [0, 1], the calculation formula is:

3. The method for improving the AGC control KP value of a power plant according to claim 2, characterized in that, The determination of the fitness function of the PSO particle, comprising: Step one, calculating the mean absolute percentage error MAPE, the formula is: Wherein, Yactual is the actual value of the power plant load, Ypre is the predicted value of the power plant load, and n is the number of power plant load prediction points; Step two, setting the fitness function, f(x) = min(MAPE).

4. The method for improving the AGC control KP value of a power plant according to claim 1, characterized in that, The speed adjustment module adjusts the feedforward amount of the boiler primary air according to the load predicted by the short-term load accurate prediction module, comprising: The load A predicted by the short-term load accurate prediction module is used to adjust the feedforward amount of the boiler primary air, a preset load matrix A0 is set, A0 includes A1, A2, A3 and A4, wherein A1 is the first preset load, A2 is the second preset load, A3 is the third preset load, and A4 is the fourth preset load, and A1>A2>A3>A4; The speed adjustment module adjusts the feedforward amount of the boiler primary air according to the relationship between the load A predicted by the short-term load accurate prediction module and the feedforward amount B of the boiler primary air; When A1≥A>A2, the speed adjustment module adjusts the feedforward amount of the boiler primary air to the first preset feedforward amount B1; When A2≥A>A3, the speed adjustment module adjusts the feedforward amount of the boiler primary air to the second preset feedforward amount B2; When A3≥A>A4, the speed adjustment module adjusts the feedforward amount of the boiler primary air to the third preset feedforward amount B3; When A4≥A, the speed adjustment module adjusts the feedforward amount of the boiler primary air to the fourth preset feedforward amount B4.

5. The method for increasing the AGC control KP value of a power plant according to claim 1, characterized in that, The load deviation signal in the boiler master control and the turbine unit pressure deviation jointly calculate, and the set value of the turbine unit pressure is modified according to the calculation result, comprising: The calculation result C of the load deviation signal in the boiler master control and the pressure deviation of the turbine unit is calculated together, a preset calculation result matrix C0 is designed, C0 is set to include C1, C2, C3, and C4, wherein C1 is a first preset calculation result, C2 is a second preset calculation result, C3 is a third preset calculation result, C4 is a fourth preset calculation result, and C1>C2>C3>C4; The regulation accuracy improvement module modifies the set value D of the turbine unit pressure according to the relationship between the calculation result C of the load deviation signal in the boiler master control and the pressure deviation of the turbine unit and the set value D of the turbine unit pressure; When C1≥C>C2, the regulation accuracy improvement module modifies the set value of the turbine unit pressure to the first preset set value D1; When C2≥C>C3, the regulation accuracy improvement module modifies the set value of the turbine unit pressure to the second preset set value D2; When C3≥C>C4, the regulation accuracy improvement module modifies the set value of the turbine unit pressure to the third preset set value D3; When C4≥C, the regulation accuracy improvement module modifies the set value of the turbine unit pressure to the fourth preset set value D4.

6. The method for improving the AGC control KP value of a power plant according to claim 1, characterized in that, The turbine unit master control PID adjusts the opening of the main valve and the regulating valve in the turbine unit according to the deviation value between the given load and the actual power of the turbine unit, including: The turbine unit master control PID adjusts the opening of the main valve and the regulating valve in the turbine unit according to the deviation value E between the given load and the actual power of the turbine unit, a preset deviation value matrix E0 is designed, E0 is set to include E1, E2, E3, and E4, wherein E1 is a first preset deviation value, E2 is a second preset deviation value, E3 is a third preset deviation value, E4 is a fourth preset deviation value, and E1>E2>E3>E4; The turbine unit master control PID adjusts the opening F of the main valve and the regulating valve in the turbine unit according to the relationship between the deviation value E between the given load and the actual power of the turbine unit and the opening F of the main valve and the regulating valve in the turbine unit; When E1≥E>E2, the turbine unit master control PID adjusts the opening of the main valve and the regulating valve in the turbine unit to the first preset opening F1; When E2≥E>E3, the turbine unit master control PID adjusts the opening of the main valve and the regulating valve in the turbine unit to the second preset opening F2; When E3≥E>E4, the turbine unit master control PID adjusts the opening of the main valve and the regulating valve in the turbine unit to the third preset opening F3; When E4≥E, the turbine unit master control PID adjusts the opening of the main valve and the regulating valve in the turbine unit to the fourth preset opening F4.

Images