CN105631086A - Damping circuit optimization design method for inhibiting numerical oscillation in simulation - Google Patents

Abstract

The invention designs an optimal design method of a damping circuit used for suppressing numerical oscillation in simulation, and belongs to the field of electromagnetic transient state. It is aimed at the problem that in the electromagnetic transient simulation platform without interpolation function, the non-state variables of energy storage elements such as inductors or capacitors produce numerical oscillations due to the action of switching devices. The core technical solution of the present invention is: through the damping circuit parameter optimization method designed in the present invention, for different simulation circuits, design different damping circuit configuration schemes according to specific circuit parameters and topological structures. The basic process is: 1. First determine the damping Selection of the initial value of the circuit; 2. Find the characteristic information representing the degree of oscillation through the custom module; 3. Use the simplex module to design the iterative loop, continuously update the R and C values through multiple nonlinear iterations, and observe the characteristic information to determine the suppression The R and C values when the effect is optimal can achieve the purpose of parameter optimization of the damping circuit.

Description

An Optimal Design Method for Damping Circuits Used to Suppress Numerical Oscillations in Simulation

technical field

The invention belongs to the field of electromagnetic transient simulation, and is especially used for solving the numerical oscillation problem in simulation.

Background technique

At present, there are many kinds of electromagnetic transient simulation calculation software for power system, and the application occasions are also different. According to different types of application requirements, it can be roughly divided into offline simulation platform and real-time simulator. The off-line simulation platform includes various common software packages, such as PSCAD/EMTDC, EMTP-RV, ATP, MicroTran, etc. Representatives of real-time simulators mainly include RTDS, RT-LAB, etc., which are widely used in industry and academia. Although the application places are different, most of the integration methods used in offline simulation or real-time simulation are implicit trapezoidal integration methods. The implicit trapezoidal integration method has the advantages of high precision and good stability. However, due to the influence of its integral characteristics, during the simulation process, the switches and devices operate, and the network structure changes, which will cause non-state variables of energy storage components such as inductors and capacitors. After the event, it swings abnormally around the true solution, that is, the numerical oscillation phenomenon in the electromagnetic transient simulation.

Contents of the invention

The invention proposes an RC damping circuit design method suitable for suppressing numerical oscillation in a simulation platform without an interpolation function. First, starting from the analysis of the mechanism of numerical oscillation, the initial value configuration of the damping circuit is determined, and a quantization function SUM that describes the severity of numerical oscillation is proposed, and the configuration of the RC circuit is optimized by combining it with the simplex algorithm, which greatly improves the performance of the damping circuit. The damping effect of the damping circuit on numerical oscillations.

Technical solution feature of the present invention comprises the following steps:

Step 1: Analyze the mechanism of numerical oscillation. For different circuit parameters, combine the integration method used by the simulation platform, and consider the interaction between circuit parameters and simulation time steps to determine the initial value configuration of the RC circuit.

Step 2: According to the electrical quantity oscillation when the numerical oscillation is generated, use the Fortran language to write a custom module to generate a characteristic signal for judging the degree of oscillation.

Step 3: Take the characteristic signal generated in step 2 as an input, combine it with the simplex module, design an iterative loop, and obtain the optimal configuration of the RC circuit through nonlinear iteration.

Through the above three steps, the present invention can determine the initial value of the RC damping circuit, further optimize its parameters through the simplex algorithm, and find the optimal solution, effectively solving the numerical oscillation problem in the simulation.

Description of drawings

Figure 1 is the RL test circuit with RC damping circuit added. In the figure, E _s , R , L are the specific parameter configurations of the RL test circuit, D is the diode, R _snb , C _snb are the specific parameter configurations of the damping circuit.

Figure 2 is the Thevenin deformation circuit of the damping circuit, RSNB is the equivalent _resistance of the damping circuit, V _{SNB_EQ} is the equivalent historical voltage source of the damping circuit, and its value is determined by the capacitor voltage, capacitor current, damping resistance and simulation step at the previous moment long decision.

Fig. 3 is a flow chart of the generation of the characteristic signal SUM function. In the figure, V _L is the voltage across the inductor L in Figure 1, and I _L is the current flowing through the inductor L in Figure 1.

Figure 4 is an optimization flow chart of the RC damping circuit, where the quantization function SUM is the characteristic signal described in Figure 2. It is worth noting that, with the continuous update of the quantization function SUM , the values of R and C of the damping circuit will also be Then it is updated to obtain the optimal configuration of the RC circuit.

detailed description

A damping circuit design method based on the optimization algorithm simplex involved in the present invention will be described in detail below. It should be emphasized that the following description is only exemplary and not intended to limit the scope of the invention and its application.

The technical problem to be solved by the present invention is to suppress the numerical oscillation problem in the simulation to the greatest extent by optimizing the parameters of the damping circuit, and to make the simulation result more accurate and credible without resorting to additional simulation algorithms. The present invention adopts following technical scheme to realize:

The present invention realizes through following three steps:

Step 1: As shown in Figure 1, connect the damping circuit in parallel to both ends of the diode, and then determine the initial values of R and C of the damping circuit.

Determination of the initial value of R: Based on the theory of binary resistance, the on-resistance of the diode is very small, generally R _on =0.01Ω; the off-resistance of the diode is large, generally R _off =1MΩ. Obviously, regardless of the on-off state of the diode, the RC damping circuit should minimize its impact on the diode, that is, R _snb should not be too small to avoid shunting when the diode is on or flowing when the diode is off. R _snb should not be too large, otherwise it will be similar to the effect of R _off so as to weaken the effect of the transient damping function. Based on the above analysis, the initial value of R _snb is roughly taken as 5kΩ, which is just in the order of magnitude between 0.01Ω and 10MΩ.

Determination of the initial value of C : It can be seen from Figure 2 that at the moment of turning off, the voltage across the diode can also be equivalent to a series connection of an equivalent resistance R _SNB and an equivalent historical voltage source V _{SNB_EQ} . It can be found that, in order to make the current i (t+Δt) not really zero, R _SNB should be of the same order of magnitude as R _snb , and should not be too large, and because the simulation situation discussed here is mainly an off-line simulation with a step size larger than 10μs, so Here the initial value of C is roughly selected as 0.02μF. It can be inferred that with the optimization work in the later stage, in order to reflect the effect of the RC damping circuit, the selection of the value of C _snb will converge upwards, because the initial value of C _snb =0.02μF is a conservative configuration close to an open circuit , and in order to suppress the abnormal fluctuation of the inductor voltage, the value of the capacitor C _snb should be large enough to suppress its fluctuation. Of course, the value of C _snb should not be too large, because once it is larger than a certain range, the damping circuit is equivalent to a short circuit state for high-frequency currents. Even if the optimization algorithm is added later, it may not be able to converge to its exact value.

Step 2: Use the Fortran language to write the custom module SUM , where the SUM function value represents the intensity of numerical oscillation. As shown in Figure 3, the inductor voltage and inductor current are collected as raw information and further processed. The output quantization function SUM value is used as the characteristic information, which is the input information for the next iterative update of the RC value. The specific algorithm of the function SUM is shown in formula (1).

(1)

Step 3: Use the quantization function SUM value obtained in step 2 as the input quantity, use the simplex optimization module, design the program according to the flow shown in Figure 4, and write the iterative loop. Through continuous iteration, the most suitable RC value is found according to the change of the SUM value, so that the effect of suppressing the oscillation is optimal.

The above is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present invention can easily think of changes or Replacement should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be determined by the protection scope of the claims.

Claims (2)

1. one kind for suppressing the antihunt circuit Optimization Design of numerical oscillation in emulating, it is characterized in that without in the emulation platform of interpolation function, for the numerical oscillation problem that switching device action causes, on the basis of existing RC antihunt circuit, a kind of brand-new method for optimally designing parameters is proposed, make RC antihunt circuit parameter value possess certain theoretical basis, also make the effect suppressing vibration reach optimum, comprise the following steps:

Step 1: by circuit is made a concrete analysis of, it is considered to the order of magnitude relation between simulation step length, integration method and circuit parameter, completes the initial value configuration of RC circuit;

Step 2: the phenomenon characteristic produced according to numerical oscillation, uses Fortran language to write suitable custom block, gathers current electric parameters and generates a characteristic signal that can characterize degree of oscillation;

Step 3: characteristic signal step 2 generated, as input quantity, optimizes the input quantity of module, and finds R, C value one group optimum by nonlinear iteration repeatedly as simplex so that suppress the effect of vibration to reach optimum.

2. a kind of for suppressing the antihunt circuit Optimization Design of numerical oscillation in emulating based on described in claim 1, it is characterized in that step 1,2 and 3 entirety are as summary of the invention, the design making antihunt circuit has had certain theoretical foundation, can significantly more efficient solving without numerical oscillation problem in interpolation emulation platform, three steps are organic indivisible entirety.