Gas turbine IGV control optimization method combining PID controller and single neuron

Abstract

The invention provides a gas turbine IGV control optimization method combining a PID controller and a single neuron. The method comprises the steps of obtaining a first system error of the PID controller in a kth control period, a second system error of the PID controller in a (k-1) th control period and a third system error of the PID controller in a (k-2) th control period; generating a proportional control input item, an integral control input item and a differential control input item of the PID controller according to the first system error, the second system error and the third system error; the single neuron determines a first control increment of the PID controller in the kth control period based on a first weight coefficient, a proportional control input item, an integral control input item and a differential control input item, and the first weight coefficient comprises a first proportional coefficient, a first integral coefficient and a first differential coefficient in the kth control period; and adjusting the first weight coefficient according to the first system error and the first control increment to obtain a new weight coefficient. Therefore, online adaptive adjustment of the PID parameters can be realized.

Description

Technical Field

[0001] The present disclosure relates to the technical field of electronic equipment, and in particular to a gas turbine IGV control optimization method combining a PID controller and a single neuron. Background Art

[0002] Gas-fired power generation uses gas turbines as power machinery to generate electricity. It has received widespread attention as a clean technology. When the gas turbine is in operation, especially in the start-stop state, surge is prone to occur due to the mismatch between the rotor speed and the intake volume and direction. During the application process, if the intake volume and direction can be adjusted quickly and accurately, a higher combined cycle efficiency can be obtained. The intake volume and direction of the gas turbine are mainly adjusted by the IGV system in the compressor. When the compressor is running, after the flow rate through the compressor decreases to a certain value, the compressor can no longer continue to work stably. At this time, the flow rate through the compressor will fluctuate violently, and the pressure ratio will also fluctuate with the fluctuation of the flow rate, causing the unit to vibrate violently, accompanied by a low-frequency, wind-like roar. This phenomenon is called "surge". The IGV system in the compressor can adjust the intake volume and intake direction of the compressor according to the actual speed of the compressor, so that the compressor can operate away from the surge boundary line; in the combined cycle unit, the exhaust temperature of the gas turbine needs to be kept near the optimal operating point to obtain higher efficiency, and by adjusting the opening of the compressor IGV to change the air flow into the compressor, the exhaust temperature of the gas turbine can be controlled within a suitable range. In addition, if the position adjustment error of the IGV is too large, or the adjustment process has a large lag and cannot adapt to the variable operating conditions of the gas turbine, the exhaust temperature of the gas turbine will fluctuate greatly, which will cause the gas turbine alarm to trip and cause losses. In more serious cases, it may even affect the safe operation of the unit and cause accidents. Therefore, the control of IGV is crucial.

[0003] At present, IGV control mainly uses traditional PID control. The key to the design of traditional PID controller is how to select P, I, and D parameters, and the setting of these parameters also limits the application of PID controller. In addition, PID controller does not have adaptive and self-correction functions. When the model parameters change or there is an external disturbance, it cannot be adjusted online in real time. Summary of the invention

[0004] The present disclosure aims to solve one of the technical problems in the related art at least to some extent.

[0005] To this end, the purpose of the present disclosure is to propose a gas turbine IGV control optimization method, device, computer equipment and storage medium combining a PID controller and a single neuron, thereby enabling online adaptive adjustment of PID parameters to make the output more accurate and quickly reach the target value.

[0006] To achieve the above-mentioned purpose, the gas turbine IGV control optimization method combining a PID controller and a single neuron is proposed in the first aspect of the present disclosure, comprising:

[0007] Obtain a first system error of the PID controller in a kth control period, a second system error in a k-1th control period, and a third system error in a k-2th control period, wherein k is an integer greater than 2;

[0008] generating a proportional control input term, an integral control input term, and a differential control input term of the PID controller according to the first system error, the second system error, and the third system error;

[0009] The first control increment of the PID controller in the kth control period is determined by a single neuron based on a first weight coefficient, the proportional control input term, the integral control input term, and the differential control input term, wherein the first weight coefficient includes:

[0010] The first proportional coefficient, the first integral coefficient and the first differential coefficient of the kth control period; the first weight coefficient is adjusted according to the first system error and the first control increment to obtain a new weight coefficient.

[0011] To achieve the above-mentioned purpose, a gas turbine IGV control optimization device combining a PID controller and a single neuron is proposed in a second aspect of the present disclosure, comprising:

[0012] An acquisition module, used to acquire a first system error of the PID controller in a kth control period, a second system error in a k-1th control period, and a third system error in a k-2th control period, wherein k is an integer greater than 2;

[0013] A generating module, configured to generate a proportional control input item, an integral control input item, and a differential control input item of the PID controller according to the first system error, the second system error, and the third system error;

[0014] A determination module, configured to determine, by a single neuron, a first control increment of the PID controller in a kth control period based on a first weight coefficient, the proportional control input term, the integral control input term, and the differential control input term, wherein the first weight coefficient includes: a first proportional coefficient, a first integral coefficient, and a first differential coefficient in the kth control period;

[0015] An adjustment module is used to adjust the first weight coefficient according to the first system error and the first control increment to obtain a new weight coefficient.

[0016] The computer device proposed in the third aspect embodiment of the present disclosure includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the gas turbine IGV control optimization method combining a PID controller and a single neuron as proposed in the first aspect embodiment of the present disclosure.

[0017] The fourth aspect embodiment of the present disclosure proposes a non-temporary computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, it implements the gas turbine IGV control optimization method combining a PID controller and a single neuron as proposed in the first aspect embodiment of the present disclosure.

[0018] The fifth aspect embodiment of the present disclosure proposes a computer program product. When the instructions in the computer program product are executed by a processor, the gas turbine IGV control optimization method combining a PID controller and a single neuron as proposed in the first aspect embodiment of the present disclosure is executed.

[0019] The present disclosure provides a gas turbine IGV control optimization method, device, computer equipment and storage medium combining a PID controller and a single neuron, by obtaining the first system error of the PID controller in the kth control cycle, the second system error in the k-1th control cycle and the third system error in the k-2th control cycle; generating a proportional control input item, an integral control input item and a differential control input item of the PID controller according to the first system error, the second system error and the third system error; determining the first control increment of the PID controller in the kth control cycle by a single neuron based on the first weight coefficient, the proportional control input item, the integral control input item and the differential control input item, wherein the first weight coefficient includes: the first proportional coefficient, the first integral coefficient and the first differential coefficient of the kth control cycle; adjusting the first weight coefficient according to the first system error and the first control increment to obtain a new weight coefficient. Thus, online adaptive adjustment of PID parameters can be achieved, so that the output is more accurate and reaches the target value quickly.

[0020] Additional aspects and advantages of the present disclosure will be given in part in the following description and in part will be obvious from the following description or learned through practice of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The above and / or additional aspects and advantages of the present disclosure will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0022] Figure 1 It is a flow chart of a gas turbine IGV control optimization method combining a PID controller and a single neuron proposed in an embodiment of the present disclosure;

[0023] Figure 2is a flow chart of a gas turbine IGV control optimization method combining a PID controller and a single neuron proposed in another embodiment of the present disclosure;

[0024] Figure 3 is a schematic diagram of the structure of a single neuron model proposed in the present disclosure;

[0025] Figure 4 is a block diagram of the neuron PID controller proposed in the present disclosure;

[0026] Figure 5 It is a structural schematic diagram of a gas turbine IGV control optimization device combining a PID controller and a single neuron proposed in one embodiment of the present disclosure;

[0027] Figure 6 A block diagram of an exemplary computer device suitable for implementing embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0028] Embodiments of the present disclosure are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present disclosure, and are not to be construed as limitations of the present disclosure. On the contrary, the embodiments of the present disclosure include all changes, modifications, and equivalents that fall within the spirit and connotation of the appended claims.

[0029] Figure 1 It is a flow chart of a gas turbine IGV control optimization method combining a PID controller and a single neuron proposed in an embodiment of the present disclosure.

[0030] Among them, it should be noted that the executor of the gas turbine IGV control optimization method combining PID controller and single neuron in this embodiment is a gas turbine IGV control optimization device combining PID controller and single neuron, which can be implemented by software and / or hardware. The device can be configured in a computer device, and the computer device can include but is not limited to a terminal, a server, etc. For example, the terminal can be a mobile phone, a handheld computer, etc.

[0031] like Figure 1 As shown, the gas turbine IGV control optimization method combining PID controller and single neuron includes:

[0032] S101: Obtain a first system error of a PID controller in a kth control cycle, a second system error in a k-1th control cycle, and a third system error in a k-2th control cycle, wherein k is an integer greater than 2.

[0033] The control period can be used to indicate the control actions of the PID controller at different times.

[0034] The system error may refer to the error value between the system setting value and the system output value, and the first system error, the second system error and the third system error may refer to the system errors in different control cycles respectively.

[0035] In the embodiment of the present disclosure, when the first system error of the PID controller in the kth control cycle, the second system error in the k-1th control cycle, and the third system error in the k-2th control cycle are obtained, reliable data support can be provided for the subsequent generation of the proportional control input items, integral control input items, and differential control input items of the PID controller.

[0036] S102: Generate a proportional control input item, an integral control input item, and a differential control input item of a PID controller according to the first system error, the second system error, and the third system error.

[0037] Among them, the proportional control input term, the integral control input term and the differential control input term can refer to the input of the proportional, integral and differential parts in the incremental PID control. The proportional (P), integral (I) and differential (D) of the PID control algorithm refer to the three key components of the algorithm, which work together in the control system to achieve precise regulation of the system output.

[0038] In the embodiment of the present disclosure, after the proportional control input item, the integral control input item and the differential control input item of the PID controller are generated according to the first system error, the second system error and the third system error, reliable input data can be provided for determining the first control increment of the PID controller in the kth control cycle.

[0039] S103: Determine the first control increment of the PID controller in the kth control cycle by a single neuron based on the first weight coefficient, proportional control input item, integral control input item and differential control input item, wherein the first weight coefficient includes: the first proportional coefficient, the first integral coefficient and the first differential coefficient of the kth control cycle.

[0040] The control increment may refer to the change in the control signal or control input between two consecutive control actions, and the first control increment refers to the control increment of the PID controller in the kth control cycle relative to the k-1th control cycle.

[0041] Among them, the proportional coefficient, integral coefficient and differential coefficient refer to the proportional, integral and differential coefficients in the PID control algorithm, which can be adjusted according to the specific application scenario to achieve the best control effect. The first proportional coefficient, the first integral coefficient and the first differential coefficient may refer to the proportional coefficient, integral coefficient and differential coefficient applied in the kth control cycle.

[0042] Optionally, in some embodiments, when a single neuron determines the first control increment of a PID controller in the kth control period based on the first weight coefficient, the proportional control input item, the integral control input item, and the differential control input item, the proportional control input item, the integral control input item, and the differential control input item may be weighted and summed based on the first weight coefficient to obtain an initial control increment; determine the first neuron coefficient of the single neuron in the kth control period; and determine the product value of the initial control increment and the first neuron coefficient as the first control increment. Thus, the initial control increment can be flexibly adjusted based on the first neuron coefficient to ensure the practicality of the obtained first control increment.

[0043] The initial control increment refers to the control increment obtained by weighted summing the proportional control input item, the integral control input item and the differential control input item based on the first weight coefficient.

[0044] The neuron coefficient refers to a parameter related to weights, which is used to adjust the effect of the input signal on the neuron output. The first neuron coefficient refers to the neuron coefficient applied in the kth control cycle.

[0045] Optionally, in some embodiments, when performing weighted summation of the proportional control input item, the integral control input item and the differential control input item based on the first weight coefficient to obtain the initial control increment, it can be: determining the sum of the absolute values ​​of the first proportional coefficient, the first integral coefficient and the first differential coefficient; determining the ratio of the first proportional coefficient to the sum of the absolute values ​​as the new first proportional coefficient; determining the ratio of the first integral coefficient to the sum of the absolute values ​​as the new first integral coefficient; determining the ratio of the first differential coefficient to the sum of the absolute values ​​as the new first differential coefficient; performing weighted summation of the proportional control input item, the integral control input item and the differential control input item based on the new first proportional coefficient, the new first integral coefficient and the new first differential coefficient to obtain the initial control increment. In this way, the normalization of different coefficients can be achieved, which can effectively improve the indication accuracy of the obtained initial control increment.

[0046] Optionally, in some embodiments, the overshoot and response speed of the PID controller in the kth control cycle can be determined; the first neuron coefficient is adjusted according to the overshoot and the response speed to obtain a new neuron coefficient, wherein the new neuron coefficient is used in the control increment calculation process of the k+1th control cycle. In this way, the neuron coefficient can be adjusted online to adapt to personalized application scenarios and improve the control effect.

[0047] Overshoot is the amount by which a system's response exceeds a target value. In a control system, high overshoot means that the system may swing above or below the target value when reaching a steady state.

[0048] Among them, the response speed can be used to indicate the time required for the system to reach a stable state.

[0049] It is understandable that in different application scenarios, neuron coefficients with different values ​​may be applicable. Therefore, the overshoot and response speed of the PID controller in the kth control cycle can be determined; the first neuron coefficient is adjusted according to the overshoot and response speed to obtain a new neuron coefficient, thereby realizing real-time adjustment of the neuron coefficient.

[0050] S104: adjusting the first weight coefficient according to the first system error and the first control increment to obtain a new weight coefficient.

[0051] That is to say, in the embodiment of the present disclosure, after determining the first system error and the first control increment, the first weight coefficient can be adjusted according to the first system error and the first control increment to obtain a new weight coefficient, thereby providing a more suitable weight coefficient for the control calculation process of the next control cycle.

[0052] In this embodiment, the first system error of the PID controller in the kth control cycle, the second system error in the k-1th control cycle, and the third system error in the k-2th control cycle are obtained; the proportional control input item, the integral control input item, and the differential control input item of the PID controller are generated according to the first system error, the second system error, and the third system error; the first control increment of the PID controller in the kth control cycle is determined by a single neuron based on the first weight coefficient, the proportional control input item, the integral control input item, and the differential control input item, wherein the first weight coefficient includes: the first proportional coefficient, the first integral coefficient, and the first differential coefficient of the kth control cycle; the first weight coefficient is adjusted according to the first system error and the first control increment to obtain a new weight coefficient. In this way, the online adaptive adjustment of the PID parameters can be realized, so that the output is more accurate and reaches the target value quickly.

[0053] Figure 2It is a flow chart of a gas turbine IGV control optimization method combining a PID controller and a single neuron proposed in another embodiment of the present disclosure.

[0054] like Figure 2 As shown, the gas turbine IGV control optimization method combining PID controller and single neuron includes:

[0055] S201: Obtain a first system error of the PID controller in a kth control period, a second system error in a k-1th control period, and a third system error in a k-2th control period, wherein k is an integer greater than 2.

[0056] S202: Generate a proportional control input item, an integral control input item, and a differential control input item of a PID controller according to the first system error, the second system error, and the third system error.

[0057] S203: Determine the first control increment of the PID controller in the kth control cycle by a single neuron based on the first weight coefficient, proportional control input item, integral control input item and differential control input item, wherein the first weight coefficient includes: the first proportional coefficient, the first integral coefficient and the first differential coefficient of the kth control cycle.

[0058] The description of S201 - S203 can be specifically referred to the above embodiment, which will not be repeated here.

[0059] S204: Processing the first system error and the first control increment based on a preset system error weighting coefficient and a control increment weighting coefficient to obtain a quadratic performance indicator of the kth control period.

[0060] The system error weighting coefficient and the control increment weighting coefficient refer to coefficients used to perform weighted processing on the first system error and the first control increment, respectively.

[0061] Among them, the quadratic performance index refers to an index commonly used in control systems, optimization problems and statistics. It is usually used to measure the deviation of the system response or state in the form of a quadratic equation, which can provide effective information about the system performance.

[0062] In an embodiment of the present disclosure, when processing the first system error and the first control increment based on the preset system error weighting coefficient and the control increment weighting coefficient to obtain the quadratic performance indicator of the kth control cycle, the first system error and the first control increment can be input into a preset calculation formula to obtain the corresponding quadratic performance indicator.

[0063] S205: Determine a first learning rate corresponding to each first weight coefficient in the kth control cycle.

[0064] The learning rate is an important hyperparameter in the gradient descent method, which can be used to determine the magnitude of updating the model parameters in each iteration. The first learning rate refers to the learning rate when correcting the corresponding weight coefficient based on the gradient descent method.

[0065] In the embodiment of the present disclosure, when determining the first learning rate corresponding to each first weight coefficient in the kth control cycle, reliable reference information can be provided for the subsequent determination of the weight coefficient adjustment value corresponding to each first weight coefficient.

[0066] S206: Determine a weight coefficient adjustment value corresponding to each first weight coefficient based on the quadratic performance indicator and the first learning rate.

[0067] The weight coefficient adjustment value may be used to describe the value of the first weight coefficient that needs to be adjusted.

[0068] In the embodiment of the present disclosure, when determining the weight coefficient adjustment value corresponding to each first weight coefficient based on the quadratic performance indicator and the first learning rate, the first weight coefficient, the quadratic performance indicator and the first learning rate can be substituted into a preset formula to obtain the corresponding weight coefficient adjustment value.

[0069] S207: Determine the sum of each first weight coefficient and the corresponding weight coefficient adjustment value as a new weight coefficient, wherein the new weight coefficient is used in the control increment calculation process of the k+1th control period.

[0070] That is to say, in the embodiment of the present disclosure, after obtaining the first system error and the first control increment, the first system error and the first control increment can be processed based on the preset system error weighting coefficient and the control increment weighting coefficient to obtain the quadratic performance index of the kth control cycle; determine the first learning rate corresponding to each first weight coefficient in the kth control cycle; determine the weight coefficient adjustment value corresponding to each first weight coefficient based on the quadratic performance index and the first learning rate; determine the sum of each first weight coefficient and the corresponding weight coefficient adjustment value as the new weight coefficient, wherein the new weight coefficient is used in the control increment calculation process of the k+1th control cycle. Thus, the reliability and rationality of the weight coefficient adjustment process can be effectively improved.

[0071] S208: Determine the overshoot and response speed of the PID controller in the kth control cycle.

[0072] S209: adjusting the first learning rate according to the overshoot and the response speed to obtain a new learning rate, wherein the new learning rate is used in the weight coefficient adjustment process of the k+1th control cycle.

[0073] That is, in the embodiment of the present disclosure, the overshoot and response speed of the PID controller in the kth control cycle can be determined; the first learning rate is adjusted according to the overshoot and the response speed to obtain a new learning rate, wherein the new learning rate is used in the weight coefficient adjustment process of the k+1th control cycle. Thus, the learning rate can be adjusted in real time to ensure that the obtained new learning rate is suitable for personalized use scenarios, which can effectively improve the robustness of the control optimization process.

[0074] In this embodiment, the first system error and the first control increment are processed based on the preset system error weighting coefficient and the control increment weighting coefficient to obtain the quadratic performance index of the kth control cycle; the first learning rate corresponding to each first weight coefficient in the kth control cycle is determined; based on the quadratic performance index and the first learning rate, the weight coefficient adjustment value corresponding to each first weight coefficient is determined; the sum of each first weight coefficient and the corresponding weight coefficient adjustment value is determined as a new weight coefficient, wherein the new weight coefficient is used for the control increment calculation process of the k+1th control cycle. Thus, the reliability and rationality of the weight coefficient adjustment process can be effectively improved. By determining the overshoot and response speed of the PID controller in the kth control cycle; adjusting the first learning rate according to the overshoot and the response speed to obtain a new learning rate, wherein the new learning rate is used for the weight coefficient adjustment process of the k+1th control cycle. Thus, the real-time adjustment of the learning rate can be achieved, ensuring that the obtained new learning rate is suitable for personalized use scenarios, and the robustness of the control optimization process can be effectively improved.

[0075] Based on the above embodiments, the gas turbine IGV control optimization method combining PID controller and single neuron proposed in the present disclosure is improved based on incremental PID, and the incremental PID control expression is as follows:

[0076] Δu(k)=K P [e(k)-e(k-1)]+K I e(k)+K D [e(k)-2e(k-1)+e(k-2)]

[0077] Single neurons have autonomous learning and adaptive capabilities, such as Figure 3 As shown, Figure 3 Schematic diagram of the single neuron model structure proposed in this disclosure. The incremental PID is improved based on the neuron control model to obtain a neuron PID controller, such as Figure 4 As shown, Figure 4 It is a structural block diagram of the neuron PID controller proposed in the present disclosure.

[0078] In the figure, x1(k), x2(k), and x3(k) correspond to the input of the proportional, integral, and differential parts in the incremental PID control, respectively. The weight coefficients w1, w2, and w3 correspond to the proportional, integral, and differential coefficients. The self-learning and adaptive adjustment of the parameters can be achieved by adjusting the weight coefficients in a targeted manner.

[0079] x1(k)=e(k)-e(k-1)

[0080] x2(k)=e(k)

[0081] x3(k)=e(k)-2e(k-1)+e(k-2)

[0082] Normalize u(k) and the expression is as follows:

[0083]

[0084] Where: K is the proportional coefficient of the neuron.

[0085] The quadratic performance index is introduced in the neuron PID control weight adjustment, and the weighted square sum of the output error and the control increment is used to optimize the weight coefficient to realize the constraint of the error and the control increment, and the optimized P, I, D parameter values ​​and the output value with smaller error are obtained. The quadratic performance index is as follows:

[0086]

[0087] e(k)=y d (k)-y(k)

[0088] Where: P is the system error weighting coefficient, Q is the control increment weighting coefficient, y d (k) is the target value and y(k) is the actual output.

[0089] According to the gradient descent method, the corresponding weight coefficient w is corrected i (k), that is, E(k) searches and adjusts the negative gradient direction of the weight coefficient. The adaptive process of the control system can be achieved by adjusting the weight coefficient (P, I, D) in a targeted manner. The weight coefficient w i (k) is adjusted as follows:

[0090]

[0091] Where: η i (η P , η I , η D ) is the learning rate of the algorithm's proportional, integral, and differential weighting coefficients.

[0092] According to the above formula, the output after iterative adjustment of the corresponding weight coefficient can be obtained as follows:

[0093]

[0094] Where: b0 is the first value of the output response.

[0095] The system output is calculated using the modified and adjusted w1, w2, w3 (P, I, D) parameter values ​​until the number of iterations or the error size meets the control requirements, and the entire optimization process ends.

[0096] From the above analysis, we can see that combining neurons with traditional PID controllers and optimizing the quadratic performance can, to a certain extent, solve the defect that traditional PID controllers are not easy to adjust online real-time parameters. It can better adapt to changes in system conditions or external environment and effectively control the system.

[0097] The relevant parameter adjustment rules are as follows:

[0098] 1. The initial weight coefficients w1, w2, w3 (P, I, D) can be selected arbitrarily.

[0099] 2. The neuron coefficient K is a relatively sensitive parameter. You can first determine an initial value and then adjust it according to the actual control. If the output has a large overshoot, the K value should be reduced. If the response rise time is long and the speed is slow, the K value should be increased.

[0100] 3. Learning rate η P , η I , η D , η P The larger the value, the smaller the overshoot, but the slower the response speed. I The larger the value, the faster the response will be, but the overshoot will also be larger. D It has little effect on the control effect of the controller.

[0101] 4. Weighting coefficients P and Q. P determines the proportion of the output error term in the optimization index and has a direct impact on system stability, while Q is used to limit the drastic changes in the control increment Δu(k). Both values ​​should not be too large and are generally selected to be relatively small.

[0102] The innovative points proposed in this disclosure include at least:

[0103] 1. By utilizing the self-learning and self-adjusting characteristics of neurons, the PID parameters can be adaptively adjusted when the parameters of the gas turbine variable operating model change or there is an external disturbance, so that the control system can be adjusted quickly, thereby avoiding excessive IGV position adjustment errors or large lags in the adjustment process, affecting the safe operation of the unit and causing losses.

[0104] 2. Introduce quadratic performance indicators in parameter adjustment, and adjust parameters according to the minimum weighted square sum of output error and control increment, so as to achieve control of output error and control increment. Compared with traditional PID control method, this method has stronger adaptive adjustment function and stronger parameter setting accuracy.

[0105] The technical method proposed in the present disclosure combines the PID controller with a single neuron with self-learning and self-adaptive capabilities to form a neuron PID controller. By utilizing the self-learning characteristics of neurons, it can adapt to environmental changes to achieve self-adjustment and correction of P, I, and D parameters, and intelligent control process, which just makes up for the shortcomings of traditional PID algorithms. At the same time, with the help of quadratic performance indicators, parameter optimization is carried out to make the setting and adjustment of PID parameters reasonable and accurate, so that the IGV system control is more accurate and rapid, and its dynamic performance is improved. While improving the system control quality, it can also reduce the probability of unit tripping accidents.

[0106] Figure 5 It is a structural schematic diagram of a gas turbine IGV control optimization device combining a PID controller and a single neuron proposed in one embodiment of the present disclosure.

[0107] like Figure 5 As shown, the gas turbine IGV control optimization device 50 combining a PID controller and a single neuron comprises:

[0108] An acquisition module 501 is used to acquire a first system error of the PID controller in a kth control period, a second system error in a k-1th control period, and a third system error in a k-2th control period, wherein k is an integer greater than 2;

[0109] A generating module 502, for generating a proportional control input item, an integral control input item and a differential control input item of a PID controller according to the first system error, the second system error and the third system error;

[0110] A determination module 503 is used to determine a first control increment of the PID controller in a kth control period based on a first weight coefficient, a proportional control input item, an integral control input item, and a differential control input item by a single neuron, wherein the first weight coefficient includes: a first proportional coefficient, a first integral coefficient, and a first differential coefficient in the kth control period;

[0111] The adjustment module 504 is used to adjust the first weight coefficient according to the first system error and the first control increment to obtain a new weight coefficient.

[0112] It should be noted that the above explanation of the gas turbine IGV control optimization method combining a PID controller and a single neuron is also applicable to the gas turbine IGV control optimization device combining a PID controller and a single neuron in this embodiment, and will not be repeated here.

[0113] In this embodiment, the first system error of the PID controller in the kth control cycle, the second system error in the k-1th control cycle, and the third system error in the k-2th control cycle are obtained; the proportional control input item, the integral control input item, and the differential control input item of the PID controller are generated according to the first system error, the second system error, and the third system error; the first control increment of the PID controller in the kth control cycle is determined by a single neuron based on the first weight coefficient, the proportional control input item, the integral control input item, and the differential control input item, wherein the first weight coefficient includes: the first proportional coefficient, the first integral coefficient, and the first differential coefficient of the kth control cycle; the first weight coefficient is adjusted according to the first system error and the first control increment to obtain a new weight coefficient. In this way, the online adaptive adjustment of the PID parameters can be realized, so that the output is more accurate and reaches the target value quickly.

[0114] Figure 6 A block diagram of an exemplary computer device suitable for implementing embodiments of the present disclosure is shown. Figure 6 The computer device 12 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0115] like Figure 6 As shown, the computer device 12 is in the form of a general-purpose computing device. The components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that connects various system components (including the system memory 28 and the processing unit 16).

[0116] The bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor or a local bus using any of a variety of bus structures. For example, these architectures include but are not limited to Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus and Peripheral Component Interconnection (PCI) bus.

[0117] The computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0118] The memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 6 Not shown, often called a "hard drive").

[0119] although Figure 6 Not shown, a disk drive for reading and writing a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing a removable non-volatile optical disk (e.g., a compact disc read only memory (Compact Disc Read Only Memory; hereinafter referred to as: CD-ROM), a digital versatile disc read only memory (Digital Video Disc Read Only Memory; hereinafter referred to as: DVD-ROM) or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present disclosure.

[0120] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in the memory 28, such program modules 42 including but not limited to an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. The program modules 42 generally perform the functions and / or methods of the embodiments described in the present disclosure.

[0121] The computer device 12 may also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), one or more devices that enable a human body to interact with the computer device 12, and / or any device that enables the computer device 12 to communicate with one or more other computing devices (e.g., network card, modem, etc.). Such communication may be performed via an input / output (I / O) interface 22. In addition, the computer device 12 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 20. As shown, the network adapter 20 communicates with other modules of the computer device 12 via a bus 18. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the computer device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0122] The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, such as implementing the gas turbine IGV control optimization method combining a PID controller and a single neuron mentioned in the aforementioned embodiment.

[0123] In order to implement the above-mentioned embodiments, the present disclosure also proposes a non-temporary computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, it implements the gas turbine IGV control optimization method combining a PID controller and a single neuron as proposed in the aforementioned embodiments of the present disclosure.

[0124] In order to implement the above embodiments, the present disclosure also proposes a computer program product. When the instruction processor in the computer program product is executed, it executes the gas turbine IGV control optimization method combining PID controller and single neuron as proposed in the above embodiments of the present disclosure.

[0125] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are to be considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0126] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

[0127] It should be noted that, in the description of the present disclosure, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present disclosure, unless otherwise specified, the meaning of "plurality" is two or more.

[0128] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present disclosure includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present disclosure belong.

[0129] It should be understood that the various parts of the present disclosure can be implemented in hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0130] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0131] In addition, each functional unit in each embodiment of the present disclosure may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0132] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0133] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0134] Although the embodiments of the present disclosure have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present disclosure. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present disclosure.

Claims

1. A gas turbine IGV control optimization method combining PID controller and single neuron, characterized in that: include: Obtain a first system error of the PID controller in a kth control period, a second system error in a k-1th control period, and a third system error in a k-2th control period, wherein k is an integer greater than 2; generating a proportional control input term, an integral control input term, and a differential control input term of the PID controller according to the first system error, the second system error, and the third system error; Determining, by a single neuron, a first control increment of the PID controller in a kth control period based on a first weight coefficient, the proportional control input term, the integral control input term, and the differential control input term, wherein the first weight coefficient includes: a first proportional coefficient, a first integral coefficient, and a first differential coefficient in the kth control period; The first weight coefficient is adjusted according to the first system error and the first control increment to obtain a new weight coefficient.

2. The method according to claim 1, characterized in that The adjusting the first weight coefficient according to the first system error and the first control increment to obtain a new weight coefficient includes: Processing the first system error and the first control increment based on a preset system error weighting coefficient and a control increment weighting coefficient to obtain a quadratic performance indicator of a kth control period; Determine a first learning rate corresponding to each of the first weight coefficients in the kth control period; Determining a weight coefficient adjustment value corresponding to each of the first weight coefficients based on the quadratic performance indicator and the first learning rate; The sum of each of the first weight coefficients and the corresponding weight coefficient adjustment value is determined as the new weight coefficient, wherein the new weight coefficient is used in the control increment calculation process of the k+1th control cycle.

3. The method according to claim 2, characterized in that The method further comprises: Determining the overshoot and response speed of the PID controller in the kth control cycle; The first learning rate is adjusted according to the overshoot and the response speed to obtain a new learning rate, wherein the new learning rate is used in a weight coefficient adjustment process of a k+1th control cycle.

4. The method according to claim 1, characterized in that The determining, by a single neuron based on a first weight coefficient, the proportional control input term, the integral control input term, and the differential control input term, of a first control increment of the PID controller in a kth control period comprises: Performing weighted summation on the proportional control input item, the integral control input item, and the differential control input item based on the first weight coefficient to obtain an initial control increment; Determine the first neuron coefficient of the single neuron in the kth control cycle; A product value of the initial control increment and the first neuron coefficient is determined as the first control increment.

5. The method according to claim 4, characterized in that The step of performing weighted summation on the proportional control input item, the integral control input item, and the differential control input item based on the first weight coefficient to obtain an initial control increment includes: determining a sum of absolute values ​​of the first proportional coefficient, the first integral coefficient, and the first differential coefficient; determining a ratio of the first proportional coefficient to the sum of the absolute values ​​as a new first proportional coefficient; determining a ratio of the first integral coefficient to the sum of the absolute values ​​as a new first integral coefficient; determining a ratio of the first differential coefficient to the sum of the absolute values ​​as a new first differential coefficient; The initial control increment is obtained by performing weighted summation on the proportional control input term, the integral control input term and the differential control input term based on the new first proportional coefficient, the new first integral coefficient and the new first differential coefficient.

6. The method according to claim 4, characterized in that The method further comprises: Determining the overshoot and response speed of the PID controller in the kth control cycle; The first neuron coefficient is adjusted according to the overshoot and the response speed to obtain a new neuron coefficient, wherein the new neuron coefficient is used in the control increment calculation process of the k+1th control cycle.

7. A gas turbine IGV control optimization device combining PID controller and single neuron, characterized in that: include: An acquisition module, used to acquire a first system error of the PID controller in a kth control period, a second system error in a k-1th control period, and a third system error in a k-2th control period, wherein k is an integer greater than 2; A generating module, configured to generate a proportional control input item, an integral control input item, and a differential control input item of the PID controller according to the first system error, the second system error, and the third system error; A determination module, configured to determine, by a single neuron, a first control increment of the PID controller in a kth control period based on a first weight coefficient, the proportional control input term, the integral control input term, and the differential control input term, wherein the first weight coefficient includes: a first proportional coefficient, a first integral coefficient, and a first differential coefficient in the kth control period; An adjustment module is used to adjust the first weight coefficient according to the first system error and the first control increment to obtain a new weight coefficient.

8. A computer device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: in, The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The invention comprises a computer program which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 6.

Images