1. A method for defending the instability of a converter level of a direct current distribution network is characterized by comprising the following steps:

s1, after the instability is detected, an initial point is obtained by using Hilbert spectrum analysis, and an initial neighborhood and an initial temperature are determined according to initial setting;

s2, substituting the bus harmonic amplitude into the system to obtain the harmonic amplitude of the bus under the action of the virtual impedance of the initial frequency point;

s3, randomly selecting a new frequency solution in the neighborhood, substituting the new frequency solution into the system, and obtaining the harmonic amplitude of the bus under the action of the virtual impedance of a new frequency point;

s4, comparing whether the bus harmonic amplitude of the new solution and the original solution is reduced or not, if the amplitude is reduced, adopting the new solution, and if the amplitude is not reduced, adopting Metropolis acceptance criterion, and probabilistically accepting the new solution;

s5, judging whether the basic iteration times are reached, if so, entering a step S6, and if not, entering a step S7 to normally generate a new solution;

s6, judging whether a termination condition of three-level detection is met, if so, locking the current frequency to finish circulation, and if not, continuing to generate a new solution step S7 and resetting the number of three-level iterations;

s7, generating a fuzzy neural network Q1 through a neighborhood, and generating a neighborhood and a temperature value of a new solution through a temperature generation fuzzy neural network Q2;

s8, repeating S3-S7 until the frequency is locked by a three-level method.

And S9, at the searched negative resistance alternating-current cutoff frequency, changing the input impedance characteristic of the load converter from the original negative resistance characteristic to the positive resistance characteristic to perform instability defense on the system.

2. The method for defending against converter-level instability of a direct current distribution network according to claim 1, wherein the step S1. obtaining the initial point using Hilbert spectrum analysis comprises:

s11, performing Hilbert signal spectrum analysis;

s12, reading a leading harmonic frequency point of system oscillation;

s13, setting the virtual impedance on a corresponding frequency point, and recording a harmonic amplitude corresponding to the bus;

s14, selecting a dominant harmonic frequency point with the minimum harmonic amplitude as an initial point.

3. The method for defending against converter-level instability of a direct current distribution network according to claim 1, wherein the step S4. using Metropolis acceptance criteria comprises the following specific steps:

s41, calculating the acceptance probability p

Wherein, Δ V_TFor bus harmonic amplitude variation

S42, randomly generating or manually setting a threshold q

S43, comparing the size relation between p and q, if p is larger than q, accepting a new solution, and if p is smaller than or equal to q, keeping the original solution.

4. The method for defending against converter-level instability of a direct current distribution network according to claim 1, wherein the step S6 includes:

s61, taking the current frequency solution when the three-level judgment is carried out as an initial point, setting an adaptive step length coefficient k (setting the k value to enable the three-level step length to be smaller than the simulated annealing step length) and calculating the initial step length delta f to delta V_r*k；

S62, calculating a new point f ═ delta f + f, and substituting the new point f ═ delta f + f into the virtual impedance to obtain a new solution bus harmonic amplitude;

and S63, comparing whether the bus harmonic amplitude of the new solution and the original solution is reduced or not, if the amplitude is reduced, adopting the new solution, and if the amplitude is not reduced, changing the frequency in the opposite direction. In addition, whether the variation amplitude of the harmonic amplitude of the bus is larger than a threshold value or not is observed, if so, the three-level judgment is quitted, the iteration times are reset, and the simulation annealing method is returned;

s64, repeating S62-S63 until the bus harmonic amplitude variation delta V of the current solution and the previous solution is found_r(ΔV_r< 0) and new solutionsBus harmonic amplitude variation delta V of current solution_r’(ΔV_r' > 0) the difference is less than a threshold value V_rTI.e. by

|DV_r-DV_r′|＜V_rT (2)

Wherein, V_rTThe setting of (c) is designed in consideration of the shape of the curve itself, being a small amount close to 0.

And S65, keeping the current solution size, repeatedly taking the symmetrical points of which the left side and the right side are the Delta f' for 2 times, locking the current frequency if the symmetrical points meet the requirement of the formula (2), and returning to S63 if the symmetrical points do not meet the requirement of the formula (2).

5. The method for defending against converter-level instability of a direct current distribution network according to claim 1, wherein the step S7 includes:

and S71, inputting the current bus harmonic amplitude and the bus harmonic amplitude variation into Q2 to obtain the temperature variation.

S72, calculating to obtain a new temperature value

And S73, inputting the new temperature value, the current bus harmonic amplitude and the bus harmonic amplitude variation into Q1 to obtain a new neighborhood solution.

Background

With the increasing use of new energy resources such as photovoltaic energy, wind energy and the like and the increasing of direct current loads such as automobiles, illumination and the like, the loads of a direct current power distribution network are more complex. Although each converter in the system is stable when individually designed and tested, when they actually make up a dc distribution system, oscillation and even instability of the overall system can be induced due to mismatch between the different converters. This instability is often due to the load converter input negative resistance characteristic causing an intercept with the source system frequency characteristic.

At present, a common method is to connect a virtual impedance in parallel at a negative resistance-to-alternating cutoff frequency to change the negative resistance characteristic in the frequency range. However, since a common power distribution network system is connected to a plurality of loads and a bus contains noise, an actual harmonic amplitude-frequency curve is a multi-extreme curve, and therefore a fast global optimization algorithm is required to find the required negative impedance intercept frequency.

A patent document 112670992 a entitled "method and system for analyzing stability and correcting instability of a power distribution network including an energy router", adopts a method for correcting instability, which comprises: in the main electric network controller, a virtual resistor is connected in parallel on the output side of the main electric network, so that the output impedance of the main electric network is improved, and the operation characteristic of a load is not influenced. This patent is directed to a power distribution network containing energy routers.

Disclosure of Invention

In view of the above, the present invention provides a method for monitoring the stability of a dc power distribution system in real time directly according to bus data information, comprising:

a method of transformer level instability protection employing an improved simulated annealing algorithm incorporating fuzzy logic control and an adaptive virtual impedance technique, the method comprising:

and (3) taking a classical simulation annealing method as a framework, and searching for an intersection frequency causing the instability of the converter. The searched intersection frequency is represented as the frequency of the global lowest point of the harmonic amplitude-frequency curve;

the improvement of the method on the basic simulated annealing can be divided into three parts: when an initial point is selected, adopting Hilbert spectrum analysis to obtain a dominant harmonic frequency point, and taking the dominant harmonic frequency point as an initial value of a simulated annealing algorithm; generating a new frequency solution and a new temperature in the iterative process of the simulated annealing algorithm through a fuzzy logic controller based on neural networking; the repeated iteration process of the simulation annealing algorithm at low temperature is quickly skipped through a three-level step-changing method.

And at the searched negative resistance AC cut-off frequency, changing the input impedance of the load converter from the original negative resistance to the positive resistance to defend the system from instability.

The cross-over frequency is found to represent the lowest point in the amplitude of the system harmonics when the virtual impedance provided in the transformer traverses the full frequency domain.

The basic simulated annealing frame mainly comprises the steps of randomly generating initial frequency points, and manually setting initial temperature, temperature updating step length and neighborhood range of each step; randomly selecting a new point in the neighborhood of the current frequency point, if the harmonic amplitude of the new point system is smaller, adopting the new point, and if the harmonic amplitude of the new point system is higher, adopting Metropolis acceptance criterion to accept a worse frequency solution with probability; updating the temperature; and repeating iteration until the set temperature value or the set iteration number value is reached.

The improved Hilbert spectrum is resolved to initial points including: hilbert frequency spectrum analysis is carried out on the system signals; reading a leading harmonic frequency point of system oscillation; setting the virtual impedance at a corresponding frequency point, and recording a corresponding harmonic amplitude; and selecting a dominant harmonic frequency point with the minimum harmonic amplitude as an initial point.

The improved fuzzy logic controller based on neural network generates a new frequency solution and a new temperature in the iteration process of the simulated annealing algorithm, and comprises the following steps: the neighborhood generation fuzzy neural network Q1 and the temperature generation fuzzy neural network Q2 are respectively provided, wherein Q1 inputs comprise a current harmonic amplitude, a harmonic amplitude change value and a temperature update value, Q1 outputs comprise a neighborhood size, Q2 inputs comprise a current harmonic amplitude and a harmonic amplitude change value, and Q2 outputs comprise a temperature update step.

Optionally, the two neural networks can be trained in a strengthened training mode, so that the simulated annealing algorithm can enter convergence more quickly and accurately, and repeated training is performed by taking the convergence speed and accuracy as reward and punishment standards. The frame design adopts a fuzzy logic mode, fuzzification, fuzzy reasoning and defuzzification.

The application process comprises the following steps: when a new frequency solution is received, recording the current harmonic amplitude and the change value of the harmonic amplitude, inputting the current harmonic amplitude and the change value of the harmonic amplitude into a Q2 neural network, and generating a temperature updating step length; and obtaining a temperature updating value according to the temperature updating step length and the current temperature, inputting the temperature updating value, the current harmonic amplitude and the harmonic amplitude change value into a Q1 neural network, and generating to obtain the size of the new field. And then continuing to perform the iterative judgment process of simulated annealing.

The 'three-level' step-variable method jumping-out iteration comprises the following steps: after the iteration times reach a set threshold value, updating frequency points by adopting smaller self-adaptive step length, and simultaneously only accepting a more optimal solution and not accepting a differential solution, namely a traditional disturbance observation method optimization method; in the disturbance optimization process, if the amplitude change of a new solution, namely a current solution, and the harmonic amplitude of the current solution, namely a previous solution, are observed to be smaller than a set threshold value, points are repeatedly searched near the current solution, and if the conditions are met, the current solution is locked as a negative resistance cross-cut frequency.

Drawings

Fig. 1 is a schematic diagram of a converter-level real-time defense method for an improved simulated annealing method and an adaptive virtual impedance of an embodiment.

Detailed Description

The exemplary embodiments of the present invention will now be described with reference to the accompanying drawings, however, the invention may be embodied in many different forms and is not limited to the embodiments described herein, which are provided for complete and complete disclosure of the invention and to fully convey the scope of the invention to those skilled in the art. The terminology used in the exemplary embodiments illustrated in the accompanying drawings is not intended to be limiting of the invention. In the drawings, the same units/elements are denoted by the same reference numerals.

Unless otherwise defined, terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Further, it will be understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense.

The invention is further illustrated by the following examples:

the method of the invention, as shown in fig. 1, comprises:

s1, after the instability is detected, an initial point is obtained by using Hilbert spectrum analysis, and an initial neighborhood and an initial temperature are determined according to initial setting.

And S2, substituting the signal into the system to obtain the harmonic amplitude of the bus under the action of the virtual impedance of the initial frequency point.

And S3, randomly selecting a new frequency solution in the neighborhood, and substituting the new frequency solution into the system to obtain the harmonic amplitude of the bus under the virtual impedance action of the new frequency point.

And S4, comparing whether the bus harmonic amplitude of the new solution and the original solution is reduced or not, if the amplitude is reduced, adopting the new solution, and if the amplitude is not reduced, adopting Metropolis acceptance criterion and probabilistically accepting the new solution.

S5, judging whether the basic iteration times are reached, if so, entering a three-level judgment S6, and if not, normally generating a new solution, and step S7.

And S6, judging whether a termination condition of three-level detection is met, if so, locking the current frequency to finish circulation, otherwise, continuing to generate a new solution step S7, and resetting the number of three-level iterations.

S7, generating a fuzzy neural network Q1 through the neighborhood, and generating a neighborhood and a temperature value of a new solution through a temperature generation fuzzy neural network Q2.

S8, repeating S3-S7 until the frequency is locked by a three-level method.

And S9, at the searched negative resistance alternating-current cutoff frequency, changing the input impedance characteristic of the load converter from the original negative resistance characteristic to the positive resistance characteristic to defend the system from instability.

The method comprises the following steps of S1, obtaining an initial point by using Hilbert spectrum analysis:

s11. spectral analysis using Hilbert signals

S12, reading a leading harmonic frequency point of system oscillation;

s13, setting the virtual impedance on a corresponding frequency point, and recording a harmonic amplitude corresponding to the bus;

s14, selecting a dominant harmonic frequency point with the minimum harmonic amplitude as an initial point.

S4, using Metropolis acceptance criteria, and specifically comprising the following steps

S41, calculating the acceptance probability p

Wherein, Δ V_TFor bus harmonic amplitude variation

S42, randomly generating or manually setting a threshold q

S43, comparing the size relation between p and q, if p is larger than q, accepting a new solution, and if p is smaller than or equal to q, keeping the original solution.

Step S6 uses three levels to determine whether to lock the frequency, and ends the loop, and the specific steps are:

s61, taking the current frequency solution when the three-level judgment is carried out as an initial point, setting a self-adaptive step length coefficient k (setting the k value to enable the three-level step length to be smaller than the simulated annealing step length) and calculating the initial step length delta f to delta V_r*k

And S62, calculating a new point f ═ Δ f + f, and substituting the new point f ═ Δ f + f into the virtual impedance to obtain a new solution bus harmonic amplitude.

And S63, comparing whether the bus harmonic amplitude of the new solution and the original solution is reduced or not, if the amplitude is reduced, adopting the new solution, and if the amplitude is not reduced, changing the frequency in the opposite direction. And in addition, whether the variation amplitude of the harmonic amplitude of the bus is larger than a threshold value or not is observed, if so, the three-level judgment is quitted, the iteration times are reset, and the simulation annealing method is returned.

S64, repeating S62-S63 until the bus harmonic amplitude variation quantity delta V of the current solution-the previous solution is found_r(△V_r<0) Bus harmonic amplitude variation quantity delta V from new solution to current solution_r’(△V_r’>0) The difference is less than the threshold value V_rTI.e. by

|DV_r-DV_r′|＜V_rT (2)

Wherein, V_rTThe setting of (c) is designed in consideration of the shape of the curve itself, being a small amount close to 0.

And S67, keeping the size of the current solution, repeatedly taking the symmetrical points of the left side and the right side of the current solution, wherein the symmetrical points are delta f' for 2 times, locking the current frequency if the symmetrical points meet the requirement of the formula (2), and returning to S63 if the symmetrical points do not meet the requirement of the formula (2).

Step S7 uses the neighborhood generation fuzzy neural network Q1, and the temperature generation fuzzy neural network Q2 generates a neighborhood and a temperature value of a new solution, wherein the training neural networks Q1 and Q2 may adopt intensive training, and the optimization speed and accuracy are used as reward and punishment criteria. The frame design adopts a fuzzy logic mode, fuzzification, fuzzy reasoning and defuzzification. The specific application process comprises the following steps:

and S71, inputting the current bus harmonic amplitude and the bus harmonic amplitude variation into Q2 to obtain the temperature variation.

S72, calculating to obtain a new temperature value

And S73, inputting the new temperature value, the current bus harmonic amplitude and the bus harmonic amplitude variation into Q1 to obtain a new neighborhood solution.

The method has originality, is beneficial to improving the running stability of the direct-current power distribution network, and has great engineering application value and popularization prospect.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein. The scheme in the embodiment of the application can be implemented by adopting various computer languages, such as object-oriented programming language Java and transliterated scripting language JavaScript.

The present application has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

While the preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all alterations and modifications as fall within the scope of the application. It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of this application and their equivalents, then this application is intended to cover such modifications and variations.