Water turbine adjusting system sliding mode control method based on finite time disturbance observation

Abstract

The invention discloses a sliding-mode control method for a water turbine adjusting system based on finite time disturbance observation, and the method comprises the steps: 1), building a fourth-order nonlinear model of the water turbine adjusting system of an ultra-long diversion tunnel according to a kinetic equation of actual components of the system; 2) establishing a finite time observer for power grid load disturbance existing in the water turbine adjusting system; 3) establishing a finite time observer for a noise signal existing in a control input channel of the water turbine adjusting system; 4) constructing a non-singular terminal sliding mode surface, ensuring that the rotating speed of a rotor of the water-turbine generator set is converged to a set value within finite time, and meanwhile, avoiding the problem of singularity; according to the control method, the anti-interference capacity of a super-long diversion tunnel water turbine adjusting system is improved, the response time of the super-long diversion tunnel water turbine adjusting system to power grid load changes is shortened, and therefore the dynamic response of a water turbine is accelerated.

Description

Technical Field

[0001] The invention relates to the technical field of water turbine regulation, and in particular to a sliding mode control method of a water turbine regulation system based on finite time interference observation. Background Art

[0002] As a clean and renewable energy source, hydropower does not produce pollutants and greenhouse gases during the power generation process, which helps reduce air pollution and avoids the problem of fossil fuel depletion. Therefore, the development and research of hydropower has received more and more attention.

[0003] In hydropower generation, a hydropower plant has good peak-shaving capability, can flexibly adjust the power generation according to the load demand of the power grid, store energy when the power supply is sufficient, release energy during peak hours, and provide support for the stable operation of the power grid. Therefore, it is at the core of the whole process of hydropower generation. In the operation of a hydropower plant, the hydro-turbine governing system is responsible for automatically or manually adjusting the operating state of the turbine according to the load demand and head conditions of the power grid to ensure power generation efficiency and system stability. With the rapid development of the power industry and the widespread application of renewable energy, the hydro-turbine governing system is also constantly improving to meet higher performance requirements. Among them, a type of super-long diversion tunnel hydro-turbine governing system has special advantages: the super-long diversion tunnel can more effectively guide the water source to the turbine and reduce the water energy loss along the way; the increased diversion distance can control the water flow in a larger range, better respond to the load changes of the power grid, and achieve precise power regulation.

[0004] The structure of the super-long water diversion tunnel turbine regulation system is complex, usually composed of a turbine unit, a generator unit, a super-long water diversion tunnel, a surge chamber, a pressure pipeline, an electro-hydraulic servo mechanism, a guide vane and a control system, and has strong nonlinear characteristics. Among them, the rotor speed of the turbine generator unit plays a decisive role in the final output power. At present, most turbine regulation systems use a control system with PID algorithm as the core to control the rotor speed. The PID algorithm does not rely on an accurate system model, has good robustness and high reliability. However, in practical applications, the rotor speed and output power will be affected by changes in the grid load. The turbine needs to respond quickly to changes in grid demand, which means that the rotor speed and output power must be able to be adjusted in a short time, and the constantly changing grid load makes the parameter design of the PID controller difficult. In addition, there are inevitable noise signals in the control input channel of the turbine regulation system. The PID algorithm has relatively poor resistance to this type of control signal noise, causing the operating state of the turbine generator unit to deviate from the set value and unable to adapt to complex working conditions. Summary of the invention

[0005] The purpose of the present invention is to overcome the defects of the prior art and provide a sliding mode control method for a turbine regulating system based on finite time disturbance observation, which can improve the observation accuracy of disturbances, shorten the dynamic response time of the turbine, and suppress noise.

[0006] The object of the present invention is achieved by: a sliding mode control method for a turbine regulating system based on finite time interference observation, comprising the following steps:

[0007] Step 1) According to the actual physical characteristics of the components of the super-long water diversion tunnel turbine regulation system, the nonlinear dynamic equation of the super-long water diversion tunnel, the dynamic equation of the surge chamber, the turbine torque and flow equation, the first-order dynamic equation of the synchronous generator, and the dynamic equation of the electro-hydraulic servo mechanism are established, and based on this, a nonlinear fourth-order system model of the super-long water diversion tunnel turbine regulation system is established with the tunnel flow, surge chamber water level change, turbine generator unit speed, and guide vane opening as state variables;

[0008] Step 2) for the grid load disturbance existing in the turbine regulation system of the ultra-long water diversion tunnel, a finite time observer is established to achieve real-time and accurate observation of the load disturbance;

[0009] Step 3) for the noise signal existing in the system control input channel, a finite time observer is established to achieve real-time and accurate observation of the noise signal;

[0010] Step 4) Based on the nonlinear fourth-order system model established in step 1), a non-singular terminal sliding surface is constructed to ensure that the rotor speed of the hydro-generator unit converges to the set value within a limited time and avoid singularity problems;

[0011] Step 5) Combining the designed disturbance observer and the non-singular terminal sliding surface, a sliding mode controller for the rotor speed of the hydro-generator set is designed to complete the rotor speed control of the hydro-generator set.

[0012] As a further limitation of the present invention, the nonlinear dynamic equation of the ultra-long water diversion tunnel in step 1) is:

[0013]

[0014] In the formula, is the water flow inertia time constant of the super-long diversion tunnel, L H is the length of the super-long diversion tunnel, Q H0 is the initial flow value of the super-long diversion tunnel, g is the acceleration of gravity, H is the turbine head, H 0 As its initial value, f H is the cross-sectional area of ​​the super-long diversion tunnel; is the relative value of the flow rate of the super-long diversion tunnel, Q His the flow rate of the super-long diversion tunnel; is the relative value of the water level change in the surge chamber; h H is the head loss of the super-long diversion tunnel, is its initial value;

[0015] The dynamic equation of the surge chamber in step 1) is:

[0016]

[0017] In the formula, is the relative value of the pressure pipeline flow, Q p is the flow rate of the pressure pipe, is its initial value; is the time constant of the surge chamber, and F is the cross-sectional area of ​​the surge chamber;

[0018] The torque and flow equation of the turbine in step 1) is:

[0019]

[0020] In the formula, is the relative value of the turbine torque, M t is the turbine torque, is its initial value; is the relative value of the speed of the turbine generator set, N is the speed of the turbine generator set, and N 0 is its initial value; is the relative value of the guide vane opening, Y is the guide vane opening, Y 0 is its initial value; e h、 e x and e y is the torque transfer coefficient of the turbine, e qh 、e qx and e qy is the flow transfer coefficient of the turbine;

[0021] The first-order dynamic equation of the synchronous generator in step 1) is:

[0022]

[0023] Where, T a is the inertia time constant of the turbine generator set, e g is the load self-regulation coefficient;

[0024] The dynamic equation of the electro-hydraulic servo mechanism in step 1) is:

[0025]

[0026] Where, T yis the inertia time constant of the electro-hydraulic servo system, and u is the output of the controller;

[0027] The fourth-order nonlinear system model of the super-long water diversion tunnel turbine regulation system described in step 1) with tunnel flow, surge chamber water level change, turbine generator speed, and guide vane opening as state variables is:

[0028]

[0029] In the formula, is the relative value of the grid load disturbance, M g is the turbine resistance torque, is its initial value; d u Control the noise signal present in the input channel for the system.

[0030] As a further limitation of the present invention, in step 2), the finite time observer is established for the grid load disturbance existing in the turbine regulation system of the ultra-long water diversion tunnel as follows:

[0031]

[0032] In the formula, the grid load disturbance m g is a non-matching interference with r-order bounded derivative; and are the grid load disturbance m g and the estimated value of its r-1 derivative;

[0033]

[0034] in, and is the disturbance observer coefficient greater than 0, and sgn(·) is the sign function.

[0035] As a further limitation of the present invention, in step 3), the finite time observer is established for the noise signal existing in the control input channel as follows:

[0036]

[0037] In the formula, the noise signal d in the control input channel u To match the disturbance, it has a bounded derivative of order q; and are the noise signals d u and estimates of its q-1 derivative;

[0038]

[0039] Among them are: and is a disturbance observer coefficient greater than 0, and sgn(·) is a sign function.

[0040] As a further limitation of the present invention, the non-singular terminal sliding surface in step 4) is:

[0041]

[0042] In the formula, x d is the set speed of the turbine generator rotor, β s >0 is the design constant of the non-singular terminal sliding surface, τ s and σ s is a positive odd number and satisfies:

[0043] As a further limitation of the present invention, the hydro-generator rotor speed sliding mode controller in step 5) is:

[0044]

[0045] In the formula, K u >0 is the sliding mode controller gain, β u Satisfy the condition 0<β u <1, and there is a function Among them, the function parameter vector

[0046] The present invention adopts the above technical solution, and compared with the prior art, the beneficial effects are as follows: 1) the non-singular terminal sliding surface constructed by the present invention can make the rotor speed of the hydro-generator unit converge to the speed setting value within a limited time, and can avoid the singular problem at the same time;

[0047] 2) The finite-time observer for power grid load disturbance constructed by the present invention can realize accurate observation of load disturbance within a limited time, shorten the response time of the turbine regulation system of the ultra-long water diversion tunnel to the time-varying power grid load, thereby improving the dynamic response capability;

[0048] 3) The finite time observer for controlling the input channel noise signal constructed by the present invention can realize accurate observation of the noise signal within a finite time, thereby improving the system's resistance to channel noise. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is the system structure diagram of the turbine regulation system for the ultra-long water diversion tunnel.

[0050] Figure 2 This is the simulation result diagram of the actual value, set value and tracking error of the rotor speed of the turbine regulating system in the ultra-long water diversion tunnel.

[0051] Figure 3 The simulation results of the actual value, observed value and observation error of the grid load disturbance of the turbine regulation system in the ultra-long water diversion tunnel are shown in the figure.

[0052] Figure 4 This is the simulation result diagram of the actual value, observed value and observation error of the noise signal in the control channel of the turbine regulation system of the ultra-long water diversion tunnel. DETAILED DESCRIPTION

[0053] Table 1 shows the relevant parameters of the turbine regulation system in the super-long diversion tunnel.

[0054]

[0055]

[0056] In the above table, “*” indicates that the value needs to be obtained through real-time measurement, and “ / ” indicates that the value has no unit.

[0057] A sliding mode control method for a hydraulic turbine regulating system based on finite-time disturbance observation comprises the following steps:

[0058] Step 1) According to the actual physical characteristics of the components of the super-long water diversion tunnel turbine regulation system, the nonlinear dynamic equation of the super-long water diversion tunnel, the dynamic equation of the surge chamber, the turbine torque and flow equation, the first-order dynamic equation of the synchronous generator, and the dynamic equation of the electro-hydraulic servo mechanism are established, and based on this, a nonlinear fourth-order system model of the super-long water diversion tunnel turbine regulation system is established with the tunnel flow, surge chamber water level change, turbine generator unit speed, and guide vane opening as state variables;

[0059] like Figure 1 As shown in the figure, the super-long water diversion tunnel turbine regulation system consists of a water turbine unit, a generator unit, a super-long water diversion tunnel, a surge chamber, a pressure pipe, an electro-hydraulic servo mechanism, a guide vane and a control system. According to the actual physical properties of its components, we can first obtain:

[0060] The dynamic equation of the super-long diversion tunnel is:

[0061]

[0062] The dynamic equation of the surge chamber is:

[0063]

[0064] The dynamic equations of turbine torque and pressure pipe flow are:

[0065]

[0066] The dynamic equation of the synchronous generator is:

[0067]

[0068] The dynamic equation of the electro-hydraulic servo mechanism is:

[0069]

[0070] Furthermore, considering the grid load disturbance and the noise in the control input channel, the tunnel flow q can be obtained. H The fourth-order nonlinear model of the turbine regulation system in the super-long water diversion tunnel with the water level change Z in the surge chamber, the speed x of the turbine generator set, and the guide vane opening y as state variables is:

[0071]

[0072] Step 2) for the grid load disturbance existing in the turbine regulation system of the ultra-long water diversion tunnel, a finite time observer is established to achieve real-time and accurate observation of the load disturbance;

[0073] Assuming that the grid load disturbance m in the turbine regulation system of the super-long diversion tunnel g With r-order derivatives and the r-th order derivative is bounded, in order to accurately obtain m g The estimated value of The finite-time observer equation for is:

[0074]

[0075] In the formula, and m g and estimates of its 1st to r-1th derivatives;

[0076]

[0077] in and is the observer coefficient greater than 0, sgn(·) is the sign function; is the estimated value of the rotor speed x of the hydro-generator set.

[0078] Based on the designed finite-time disturbance observer, Able to track the actual grid load disturbance m within a limited time g , which can be guaranteed in a limited time Afterwards, there also, express The first derivative of and Will be used in controller design.

[0079] Step 3) for the noise signal existing in the system control input channel, a finite time observer is established to achieve real-time and accurate observation of the noise signal;

[0080] Assume that the noise signal d in the control input channel of the turbine regulation system of the super-long water diversion tunnel u With q-order derivative and the q-th order derivative is bounded, in order to accurately obtain the noise signal d u The estimated value of The finite-time observer equation for is:

[0081]

[0082] In the formula, and They are d u and estimates of its 1st to q-1th derivatives;

[0083]

[0084] in, and is the observer coefficient greater than 0, sgn(·) is the sign function; is the estimated value of the guide vane opening y.

[0085] Based on the designed finite-time disturbance observer, Able to track the actual noise signal d in a limited time u , which can be guaranteed in a limited time Afterwards, there Will be used in controller design.

[0086] Step 4) Based on the system model established in step 1), a non-singular terminal sliding surface is constructed to ensure that the rotor speed of the hydro-generator unit converges to a set value within a limited time and avoids singularity problems;

[0087] To ensure that the rotor speed x of the generator set of the turbine regulating system of the super-long water diversion tunnel tracks the set value x within a limited time d , while avoiding the singularity problem, construct a non-singular terminal sliding surface:

[0088]

[0089] In the formula, x d is the set speed of the turbine generator rotor, β s >0 is the design constant of the non-singular terminal sliding surface, τ s and σ s Is a positive odd number and satisfies the conditions:

[0090] The above non-singular terminal sliding surface can ensure that: After the sliding mode s converges to s = 0, After that there is xx d = 0, that is, the rotor speed x of the turbine generator set tracks its set value x d .

[0091] Step 5) combining the designed disturbance observer and the non-singular terminal sliding surface, designing a sliding mode controller to complete the rotor speed control of the hydro-generator set;

[0092] In order to ensure that the sliding mode s reaches s = 0 within a finite time, the sliding mode control law is designed as:

[0093]

[0094] In the formula, K u >0 is the sliding mode controller gain, β u Satisfy the condition 0<β u <1, and there is a function Among them, the function parameter vector

[0095] Under the intervention of the designed sliding mode control law u, the sliding mode s can reach s = 0 in a finite time. The following is the proof process:

[0096] Select the Lyapunov function:

[0097]

[0098] V s After taking the derivative and substituting the fourth-order dynamic equation of the turbine regulating system into it, we get:

[0099]

[0100] Substituting the sliding mode control law u into the above formula, we get:

[0101]

[0102] The information constructed in steps 2) and 3) and The observer of is finite-time stable, so the observation error and In limited time The inner is bounded, that is, it exists Make and Established. And because τ s , σ s is a positive odd number and has τs >σ s , so τ s -σ s is a positive even number, then In addition, because z, x, y, and Continuously divisible, so In limited time is bounded, assuming its maximum value is Then we can further get:

[0103]

[0104] make When |s|≥1, we have holds; when |s|<1, Established. Combining the two situations, we get: Solve That is, the sliding mode s will not exist in a finite time escape.

[0105] exist After that, the interference observation error converges to 0, and there is and Established, then we can get:

[0106]

[0107] From the above analysis, we can get: And only when When , it is equal to 0. When V s Can converge to 0 in a finite time; when When , we can get:

[0108]

[0109] From the above, we can see that is not an attractor. So we can get: V s It can converge to 0 in a finite time, that is, there is a finite time t s So that for any time s(t)=0 holds.

[0110] After the sliding mode s converges to the non-singular terminal sliding mode surface s = 0 constructed in step 4), There are xx d = 0, then we can further obtain: for any time have It is established, that is, the rotor speed of the hydro-generator set can track its set value within a limited time.

[0111] The sliding mode controller of the turbine regulation system in the ultra-long water diversion tunnel based on the finite time disturbance observer is designed, and the control effect of the invention is verified by simulation comparison in Simulink. Specifically, the simulation time is set to 100s; the initial value N of the rotor speed of the turbine generator set is set 0 The rotor speed of the turbine generator set is set to 55r / s. The speed relative value setting signal Setting the turbine resistance torque Relative value of grid load disturbance Set the control channel noise signal In the finite time observer of the grid load disturbance and the control channel noise signal, take the fifth-order derivative and set the observer coefficient and is equal to 30, Equal to 60, set the observer coefficient and is equal to 50, Equal to 80; set the non-singular terminal sliding surface constant: β s =20,τ s =5 and σ s =3; Set the sliding mode controller gain: K u =65 and controller constant: β u =0.6.

[0112] In the rotor speed setting value of the hydro-turbine generator set, two speed sudden increase conditions are set. Figure 2 The simulation results of the relative value, set value and tracking error of the rotor speed of the turbine regulating system in the super-long water diversion tunnel are shown in the figure. In the grid load disturbance, five operating conditions are set: linear increase, constant value, sinusoidal change, sawtooth wave change and sudden drop. Figure 3 The simulation results of the relative value, observation value and observation error of the grid load disturbance in the turbine regulation system of the super-long water diversion tunnel are shown in the figure. In the control channel noise signal, five working conditions are set: constant value, sinusoidal change, sawtooth wave change, sudden increase and sudden decrease. Figure 4 This is a simulation result diagram of the control channel noise signal, observation value and observation error in the turbine regulation system of the ultra-long water diversion tunnel.

[0113] It can be seen from the simulation results that, in general, the finite-time observer can achieve accurate observation of grid load disturbance and control channel noise signal, has better tracking performance of the generator rotor speed, and has faster dynamic response of the turbine.

[0114] The present invention provides a sliding mode control method for a turbine regulating system based on finite-time disturbance observation, which includes: 1) establishing a finite-time observer for two disturbances in a super-long water diversion tunnel turbine regulating system, which can realize accurate observation within a finite time; 2) adopting a method of constructing a non-singular terminal sliding mode surface, so that the rotor speed of the hydro-generator unit converges to a set value within a finite time, and can avoid singular problems. From the perspective of improving the control algorithm, the present invention improves the anti-interference ability of the super-long water diversion tunnel turbine regulating system, shortens its response time to changes in power grid load, and thus accelerates the dynamic response of the turbine. Since there is no need to increase the hardware structure too much, the method has good feasibility, can produce obvious economic benefits, and improves the practicability of the control algorithm of the super-long water diversion tunnel turbine regulating system in actual engineering applications.

[0115] The present invention is not limited to the above-mentioned embodiments. On the basis of the technical solution disclosed in the present invention, technicians in this field can make some substitutions and deformations to some technical features therein according to the disclosed technical content without creative labor, and these substitutions and deformations are all within the protection scope of the present invention.

Claims

1. A sliding mode control method for a hydraulic turbine regulating system based on finite time disturbance observation, characterized in that: The following steps are involved: Step 1) According to the actual physical characteristics of the components of the super-long water diversion tunnel turbine regulation system, the nonlinear dynamic equation of the super-long water diversion tunnel, the dynamic equation of the surge chamber, the turbine torque and flow equation, the first-order dynamic equation of the synchronous generator, and the dynamic equation of the electro-hydraulic servo mechanism are established, and based on this, a nonlinear fourth-order system model of the super-long water diversion tunnel turbine regulation system is established with the tunnel flow, surge chamber water level change, turbine generator unit speed, and guide vane opening as state variables; Step 2) for the grid load disturbance existing in the turbine regulation system of the ultra-long water diversion tunnel, a finite time observer is established to achieve real-time and accurate observation of the load disturbance; Step 3) for the noise signal existing in the system control input channel, a finite time observer is established to achieve real-time and accurate observation of the noise signal; Step 4) Based on the nonlinear fourth-order system model established in step 1), a non-singular terminal sliding surface is constructed to ensure that the rotor speed of the hydro-generator unit converges to the set value within a limited time and avoid singularity problems; Step 5) Combining the designed disturbance observer and the non-singular terminal sliding surface, a sliding mode controller for the rotor speed of the hydro-generator set is designed to complete the rotor speed control of the hydro-generator set.

2. The sliding mode control method for a water turbine regulating system based on finite time disturbance observation according to claim 1 is characterized in that: The nonlinear dynamic equation of the ultra-long water diversion tunnel in step 1) is: In the formula, is the water flow inertia time constant of the super-long diversion tunnel, L H is the length of the super-long diversion tunnel, is the initial value of the flow rate of the super-long diversion tunnel, g is the gravity acceleration, H is the turbine head, H0 is its initial value, and f H is the cross-sectional area of ​​the super-long diversion tunnel; is the relative value of the flow rate of the super-long diversion tunnel, Q H is the flow rate of the super-long diversion tunnel; is the relative value of the water level change in the surge chamber; h H is the head loss of the super-long diversion tunnel, is its initial value; The dynamic equation of the surge chamber in step 1) is: In the formula, is the relative value of the pressure pipeline flow, Q p is the flow rate of the pressure pipe, is its initial value; is the time constant of the surge chamber, and F is the cross-sectional area of ​​the surge chamber; The torque and flow equation of the turbine in step 1) is: In the formula, is the relative value of the turbine torque, M t is the turbine torque, is its initial value; is the relative value of the speed of the hydro-generator set, N is the speed of the hydro-generator set, and N0 is its initial value; is the relative value of the guide vane opening, Y is the guide vane opening, and Y0 is its initial value; h、 e x and e y is the torque transfer coefficient of the turbine, e qh 、e qx and e qy is the flow transfer coefficient of the turbine; The first-order dynamic equation of the synchronous generator in step 1) is: Where, T a is the inertia time constant of the turbine generator set, e g is the load self-regulation coefficient; The dynamic equation of the electro-hydraulic servo mechanism in step 1) is: Where, T y is the inertia time constant of the electro-hydraulic servo system, and u is the output of the controller; The fourth-order nonlinear system model of the super-long water diversion tunnel turbine regulation system described in step 1) with tunnel flow, surge chamber water level change, turbine generator speed, and guide vane opening as state variables is: In the formula, is the relative value of the grid load disturbance, M g is the turbine resistance torque, is its initial value; d u Control the noise signal present in the input channel for the system.

3. The sliding mode control method for a water turbine regulating system based on finite time disturbance observation according to claim 1 is characterized in that: In step 2), for the grid load disturbance existing in the turbine regulation system of the ultra-long diversion tunnel, a finite time observer is established as follows: In the formula, the grid load disturbance m g is a non-matching interference with r-order bounded derivative; and …、 are the grid load disturbance m g and the estimated value of its r-1 derivative; in, and is the disturbance observer coefficient greater than 0, and sgn(·) is the sign function.

4. The sliding mode control method for a water turbine regulating system based on finite time disturbance observation according to claim 1 is characterized in that: In step 3), the finite time observer is established for the noise signal existing in the control input channel as follows: In the formula, the noise signal d in the control input channel u To match the disturbance, it has a bounded derivative of order q; and are the noise signals d u and estimates of its q-1 derivative; Among them are: and is a disturbance observer coefficient greater than 0, and sgn(·) is a sign function.

5. The sliding mode control method for a water turbine regulating system based on finite time disturbance observation according to claim 1, characterized in that: The non-singular terminal sliding surface in step 4) is: In the formula, x d is the set speed of the turbine generator rotor, β s >0 is the design constant of the non-singular terminal sliding surface, τ s and σ s is a positive odd number and satisfies:

6. The sliding mode control method for a water turbine regulating system based on finite time disturbance observation according to claim 1, characterized in that: The sliding mode controller of the rotor speed of the hydro-generator set in step 5) is: In the formula, K u >0 is the sliding mode controller gain, β u Satisfy the condition 0<β u <1, and there is a function Among them, the function parameter vector

Images