CN114818509A - Filter parameter design method, filter parameter design device, computer equipment and storage medium - Google Patents
Filter parameter design method, filter parameter design device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN114818509A CN114818509A CN202210536043.6A CN202210536043A CN114818509A CN 114818509 A CN114818509 A CN 114818509A CN 202210536043 A CN202210536043 A CN 202210536043A CN 114818509 A CN114818509 A CN 114818509A
- Authority
- CN
- China
- Prior art keywords
- filter
- population
- individual
- operation mode
- local optimal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/20—Design reuse, reusability analysis or reusability optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to a filter parameter design method, a filter parameter design device, computer equipment and a storage medium. The method comprises the following steps: generating an initialization population according to external parameters of a filter to be optimized based on the current operation mode of the power grid simulation model; the external parameters comprise a filter placement position and a filter structure determined according to a power grid simulation model in a full-wiring mode; processing the initialization population by adopting an artificial intelligence model to obtain a local optimal individual corresponding to the current operation mode; switching the current operation mode until traversing each operation mode of the power grid simulation model to obtain local optimal individuals respectively corresponding to each operation mode, and obtaining a local optimal cluster of each local optimal individual; acquiring a global optimal individual based on the objective function value of each individual of the local optimal cluster; the globally optimal individuals are used to characterize the internal parameters of the filter to be optimized. By adopting the method, the resonance risk range of the filter can be enlarged.
Description
Technical Field
The present application relates to the field of power harmonic suppression technologies, and in particular, to a filter parameter design method, device, computer device, and storage medium.
Background
In daily large-scale power grid operation, power electronic devices have indispensable functions, however, many devices, such as power transformers, frequency conversion devices, thyristor rectification devices and the like in power transmission and distribution systems have nonlinear and time-varying characteristics, which can distort waveforms of current and voltage in the power grid, a series of sinusoidal components with frequencies being integral multiples of fundamental waves can be found in frequency domain images, which are called as harmonics, and the occurrence of the harmonics can bring adverse effects, such as reducing power quality and affecting the use safety of the power grid, increasing harmonic power and increasing energy loss.
At present, the conventional method mainly selects a proper filter according to the impedance of an alternating current system, the capacity of a direct current system and a wiring mode, and utilizes a corresponding parameter calculation formula to calculate parameters. However, the existing filter parameter design method has the problems of small frequency range for avoiding resonance risk, and the like.
Disclosure of Invention
In view of the above, it is necessary to provide a filter parameter design method, apparatus, computer device, computer readable storage medium and computer program product capable of improving the avoidance range of resonance risk in view of the above technical problems.
In a first aspect, the present application provides a filter parameter design method. The method comprises the following steps:
generating an initialization population according to external parameters of a filter to be optimized based on the current operation mode of the power grid simulation model; the external parameters comprise a filter placement position and a filter structure determined according to a power grid simulation model in a full-wiring mode;
processing the initialization population by adopting an artificial intelligence model to obtain a local optimal individual corresponding to the current operation mode;
switching the current operation mode until traversing each operation mode of the power grid simulation model to obtain local optimal individuals respectively corresponding to each operation mode, and obtaining a local optimal cluster of each local optimal individual;
acquiring a global optimal individual based on the objective function value of each individual of the local optimal cluster; the globally optimal individuals are used to characterize the internal parameters of the filter to be optimized.
In one embodiment, the step of obtaining the locally optimal individual corresponding to the current operation mode by processing the initialization population with an artificial intelligence model includes:
iteratively updating the initialized population by adopting a genetic algorithm to obtain a current population; processing the iterative optimal solution of the current population by adopting a greedy algorithm to obtain a local optimal individual; the iterative optimal solution is determined according to the objective function values of the individuals of the current population.
In one embodiment, the filter placement position is determined by processing the total harmonic distortion of each bus voltage with a threshold to be filtered; the total harmonic distortion is obtained through a simulation result of a power grid simulation model in a full-wiring mode;
the filter structure is determined according to the number of harmonic ranges by dividing a plurality of harmonic ranges for the harmonic to be filtered in the total harmonic distortion.
In one embodiment, the initialization population is iteratively updated by using a genetic algorithm, and the step of obtaining the current population comprises the following steps:
and if the current evolution iteration number is less than the preset evolution iteration number, performing cross and mutation processing on the last initialized population until a new individual with a target function value superior to that of the last initialized population is obtained, so as to update the initialized population to obtain the current initialized population.
In one embodiment, the step of processing the iterative optimal solution of the current population by using a greedy algorithm to obtain the locally optimal individual comprises the following steps:
processing the iterative optimal solution of the current population by adopting a greedy algorithm to obtain a local optimal solution;
determining individuals corresponding to the local optimal solution meeting the preset precision as local optimal individuals;
if the local optimal solution does not meet the preset precision, eliminating the individuals with low fitness and high concentration of the current population by adopting an immune algorithm; and increasing the preset population evolution iteration times according to the eliminated number of individuals until a local optimal solution meeting the preset precision is obtained.
In one embodiment, the objective function value of an individual is obtained by filtering each harmonic amplitude value through linear normalization processing; and each subharmonic amplitude is obtained through a simulation result of a power grid simulation model accessed to the filter to be optimized.
In a second aspect, the present application further provides a filter parameter design apparatus. The device comprises:
the population obtaining module is used for generating an initialization population according to external parameters of the filter to be optimized based on the current operation mode of the power grid simulation model; the external parameters comprise a filter placement position and a filter structure determined according to a power grid simulation model in a full-wiring mode;
the individual acquisition module is used for processing the initialized population by adopting an artificial intelligence model to obtain a local optimal individual corresponding to the current operation mode;
the cluster acquisition module is used for switching the current operation mode until each operation mode of the power grid simulation model is traversed to obtain local optimal individuals respectively corresponding to each operation mode and obtain a local optimal cluster of each local optimal individual;
the parameter acquisition module is used for acquiring a global optimal individual based on the objective function value of each individual of the local optimal cluster; the globally optimal individuals are used to characterize the internal parameters of the filter to be optimized.
In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the steps of the method described above when executing the computer program.
In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method described above.
In a fifth aspect, the present application further provides a computer program product. The computer program product comprises a computer program which, when executed by a processor, carries out the steps of the method described above.
According to the filter parameter design method, the filter parameter design device, the computer equipment, the storage medium and the computer program product, the initialization population is generated according to the external parameters of the filter to be optimized; processing the initialization population by adopting an artificial intelligence model to obtain a local optimal individual corresponding to the current operation mode; switching the current operation mode until each operation mode of the power grid simulation model is traversed, and obtaining a global optimal individual; the globally optimal individuals are used to characterize the internal parameters of the filter to be optimized. The parameter optimization design of the filter is carried out by using an intelligent algorithm in the field of artificial intelligence, so that harmonic components in a power grid system are suppressed, and resonance avoidance can be effectively carried out in a full frequency band.
Drawings
FIG. 1 is a schematic flow chart diagram of a filter parameter design method in one embodiment;
FIG. 2 is a schematic flow chart of the filter parameter design step in one embodiment;
FIG. 3 is a schematic flow chart of the steps of determining a filter structure in one embodiment;
FIG. 4 is a schematic flow chart showing the filter parameter design step in another embodiment;
FIG. 5 is a block diagram of a filter parameter design apparatus according to an embodiment;
FIG. 6 is a diagram illustrating an internal structure of a computer device according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
It should be noted that the impedance of the filter designed by the existing method is in a parallel relationship with the impedance of the alternating current system, the global impedance is not controllable, the resonance risk cannot be avoided in the full frequency band, and the optimal filter scheme is difficult to determine by the existing theoretical filter design method.
In one embodiment, as shown in FIG. 1, a filter parameter design method is provided. The method comprises the following steps:
step 110, generating an initialization population according to external parameters of a filter to be optimized based on the current operation mode of the power grid simulation model; the external parameters comprise a filter placement position and a filter structure determined according to a power grid simulation model in a full-wiring mode;
specifically, as shown in fig. 2, a power grid simulation model needs to be established first according to the actual power grid condition to analyze and determine the filter placement position and the filter structure (for example, the usage type), so as to facilitate the next step of obtaining the optimized internal parameters of the filter to be optimized by using an artificial intelligence model; in fact, the power grid has multiple operation modes, including a full-wiring mode and various maintenance modes (for example, an N-1 maintenance mode and an N-2 maintenance mode), and the filter structure to be optimized is determined under the condition that the power grid simulation model is in the full-wiring mode. The initialization population comprises a plurality of individuals; generating an initialization population according to external parameters of the filter to be optimized, i.e. generating a group of individuals carrying a fixed number of genetic characteristics, e.g. generating a group of arrays with a certain number of elements, each individual corresponding to a filter, and the genetic characteristics of the individual corresponding to internal parameters of the filter; the individual goodness can be determined by calculating the objective function value of each individual, for example, the filtering effect of the filter to be optimized is determined by calculating the objective function value of each individual.
In some examples, a power grid simulation model is built in simulation software such as Simulink in PSCAD, MATLAB, etc., a time domain waveform of each bus voltage value is obtained according to a simulation result of the power grid simulation model, and each bus voltage value can be converted from a time domain to a frequency domain through a Fast Fourier Transform (FFT), so as to facilitate processing of subsequent frequency domain components. Taking optimization of internal parameters of the single-tuning type-I filter as an example, an individual a [ a1, a2] is generated, where a1 and a2 respectively correspond to values of L and C in the single-tuning type-I filter, and the value types and ranges of L and C can be limited in practical cases, for example, under the condition of ensuring convergence accuracy, the value range of L can be [0,10], and the effective value can be 1 bit after a decimal point.
Step 120, processing the initialization population by adopting an artificial intelligence model to obtain a local optimal individual corresponding to the current operation mode;
specifically, the artificial intelligence model adopted can comprise a genetic algorithm, the thought of the genetic algorithm is derived from a biological evolution process, and the genetic algorithm is a search algorithm based on an information genetic mechanism and a superior-inferior natural selection principle in the evolution process. The genetic algorithm searches in this state space using a probabilistic search process, resulting in new samples. An individual is an object in a problem (e.g., a solution to the problem) that simulates a biological individual, i.e., a point in the search space. The genetic algorithm first encodes the search structure into a string form, with each string structure being made an individual. A population is a population of individuals that mimics a population of organisms, and is typically a small subset of the entire search space. A set of string structures may be referred to as a population. The initialized population is processed by adopting an artificial intelligence model, so that a local optimal individual corresponding to the current operation mode can be obtained, namely, the parameter combination of the filter to be optimized, which can effectively reduce the harmonic component of the power grid, can be obtained. In some examples, an individual corresponding to the locally optimal solution in the current operation mode may be determined as the locally optimal individual.
Step 130, switching the current operation mode until traversing each operation mode of the power grid simulation model to obtain local optimal individuals respectively corresponding to each operation mode, and obtaining a local optimal cluster of each local optimal individual;
specifically, in order to ensure that the obtained internal parameters of the filter to be optimized have good filtering efficiency under different power grid operation modes, the internal parameters of the filter to be optimized are optimized by traversing each operation mode of the power grid simulation model. Different local optimal individuals can be generated by parameter optimization in each operation mode, and the local optimal individuals in each operation mode are stored, so that a local optimal cluster of each local optimal individual is obtained.
In some examples, whether all operation modes of the power grid simulation model are traversed or not is judged, and if yes, a local optimal cluster of each local optimal individual is determined; if not, switching the current operation mode of the power grid simulation model, and obtaining the local optimal individual corresponding to the switched current operation mode.
Step 140, obtaining a global optimal individual based on the objective function value of each individual of the local optimal cluster; the globally optimal individuals are used to characterize the internal parameters of the filter to be optimized.
Specifically, the fitness of each individual of the local optimal cluster is evaluated, and the individual with high fitness is determined as a global optimal individual, so as to obtain internal parameters of the filter to be optimized, which have good filtering effect under all different power grid operation modes; in some examples, fitness of individual ones of the local optimal cluster may be determined from objective function values of the individual ones of the local optimal cluster. The internal parameters of the filter to be optimized may be optimization parameters including one or several values of R (resistance), L (inductance) and C (capacitance).
The method comprises the steps that an initialization population is generated according to external parameters of a filter to be optimized; processing the initialization population by adopting an artificial intelligence model to obtain a local optimal individual corresponding to the current operation mode; switching the current operation mode until each operation mode of the power grid simulation model is traversed, and obtaining a global optimal individual; the globally optimal individuals are used to characterize the internal parameters of the filter to be optimized. The parameter optimization design of the filter is carried out by using an intelligent algorithm in the field of artificial intelligence, so that harmonic components in a power grid system are suppressed, and resonance avoidance can be effectively carried out in a full frequency band.
In one embodiment, the step of obtaining the locally optimal individual corresponding to the current operation mode by processing the initialization population with an artificial intelligence model includes:
iteratively updating the initialized population by adopting a genetic algorithm to obtain a current population; processing the iterative optimal solution of the current population by adopting a greedy algorithm to obtain a local optimal individual; the iterative optimal solution is determined according to the objective function values of the individuals of the current population.
Specifically, the initialized population is iteratively updated by using a genetic algorithm to obtain a current population; determining an iterative optimal solution of the current population according to the objective function values of the individuals of the current population; and processing the iterative optimal solution of the current population by adopting a greedy algorithm to obtain a local optimal individual. The greedy algorithm decomposes the global optimal problem into local optimal problems in each operation mode by acquiring the optimal solution (namely, local optimal individuals) of the power grid simulation model in the current operation mode, so that the global optimal individuals can be acquired conveniently. In some examples, individual objective function values may be used to determine the filtering effect of the filter to be optimized.
In one embodiment, the filter placement position is determined by processing the total harmonic distortion of each bus voltage with a threshold to be filtered; the total harmonic distortion is obtained through a simulation result of a power grid simulation model in a full-wiring mode;
the filter structure is determined according to the number of harmonic ranges by dividing a plurality of harmonic ranges for the harmonic to be filtered in the total harmonic distortion.
Specifically, Total Harmonic Distortion (THD) of each bus voltage is obtained through a simulation result of the power grid simulation model in the full-wiring mode, and if it is determined that the Total Harmonic Distortion (THD) of a certain bus voltage exceeds a preset voltage Harmonic Distortion threshold value (THD) 1 Then, a filter is added to the bus to determine the filter placement.
Further, the harmonic wave distribution rule is carried outThe filter structure can be determined analytically, for example by means of a harmonic voltage amplitude threshold U 1 Finding all the harmonics with the voltage amplitudes higher than the threshold value and the corresponding frequency values thereof, thereby obtaining the harmonics to be filtered; by a frequency difference threshold value deltaf 1 Dividing a plurality of harmonic ranges for the harmonic to be filtered in the total harmonic distortion, and judging the selected filter structure according to the number of the harmonic ranges; if the component distribution of the harmonic to be filtered is more concentrated, the tuning filter can be preferentially selected, and if the component distribution of the harmonic to be filtered is more dispersed, other filters (such as a low-pass filter, a high-pass filter, a band-pass filter or a band-stop filter) are selected to be used.
In some examples, the frequency difference threshold Δ f may be established by 1 For example, if a group of harmonic components is found, the frequency differences between the different harmonic components are all Δ f 1 Within, the group of harmonic components is considered to be a harmonic group (i.e., a harmonic range); the components of the harmonic to be filtered can be arranged on the horizontal axis in the order of the frequency from small to large, and the harmonic component with the smallest frequency (assuming that the frequency value is f) 1 ) Frequency value of (d) plus Δ f 1 Thus determining the range of the harmonic collective 1, i.e. considering all frequency values at f 1 To f 1 +Δf 1 All the harmonic components of (1) are subordinate to the harmonic group 1; find the nearest neighbor harmonic quantity (assuming that its frequency value is f) arranged on the horizontal axis behind the harmonic set 1 2 ) By f 2 Plus Δ f 1 Finding the range of harmonic collective 2, i.e. considering all frequency values at f 2 To f 2 +Δf 1 All the harmonic components of (2) are subordinate to the harmonic group 2; by analogy, all harmonic components to be filtered out can be divided into n harmonic groups.
Further, as shown in fig. 3, whether to select a tuning filter is determined according to the number of the harmonic ranges. If the number n of the harmonic ranges is less than 4, it is indicated that the distribution of the harmonic components to be filtered is concentrated, in this case, the tuning filter can be preferentially used, and if the number n of the harmonic ranges is 1, the single tuning filter is selected; if the number n of the harmonic ranges is 2, using a double-tuned filter; similarly, if the number n of the harmonic ranges is 3, thenUsing a triple-tuned filter; if the number n of the harmonic ranges is greater than or equal to 4, which indicates that the distribution of the harmonic components to be filtered is relatively dispersed, other filters are considered to be used after the tuned filter is not used. Setting a cut-off frequency f k If the frequency of most harmonic components is greater than f k Then a low pass filter is used; if most harmonic components are less than f k Then a high pass filter is used; if this is not the case, the use of band-pass or band-stop filters is to be taken into account, two cut-off frequencies f being set up k1 ,f k2 (f k1 <f k2 ) If the frequency of a part of harmonic components to be filtered is less than f k1 And the other part is larger than f k2 Then a band pass filter is used; similarly, if the frequency of most of the harmonic components to be filtered is concentrated on f k1 And f k2 In between, a band-stop filter is used.
In one embodiment, the initialization population is iteratively updated by using a genetic algorithm, and the step of obtaining the current population comprises the following steps:
and if the current evolution iteration number is less than the preset evolution iteration number, performing cross and mutation processing on the last initialized population until a new individual with a target function value superior to that of the last initialized population is obtained, so as to update the initialized population to obtain the current initialized population.
Specifically, as shown in fig. 4, according to the preset evolution iteration number of the initialized population, when the initialized population performs complete intersection and mutation processing once and determines that a new individual having a target function value superior to the last initialized population is obtained, the step is called one-time evolution, the preset evolution iteration number needs to be reasonably designed, an optimal solution is difficult to find when the preset evolution iteration number is too small, so that the final effectiveness of the filter is affected, and a large amount of calculation time is consumed when the preset evolution iteration number is too large. And if the current evolution iteration times are less than the preset evolution iteration times, performing cross and mutation treatment on the last initialization population. The purpose of crossover and mutation processing is to create new individuals, increasing the probability of the best solution occurring. When a new individual is generated after once crossing and mutation processing, recalculating the objective function value of the individual in the current initialized population, and judging whether the initialized population needs to be evolved again; if the evolved initialization population has individuals better than the initialization population before evolution (the last initialization population), replacing the optimal solution and the corresponding combination thereof reserved in the previous step to update the initialization population to obtain the current initialization population; otherwise, returning to cross and mutation and carrying out new evolution. And if the current evolution iteration times are more than or equal to the preset evolution iteration times, determining the initialized population as the current population.
In some examples, individuals with poor filtering (e.g., low objective function values) may be selected for crossover and mutation processing. Wherein the cross-over treatment can be the gene exchange sequence of the individual, and can be the element exchange position in the array, for example, the array A [1,2,5,1,6] becomes A' [5,2,1,1,6] after the cross-over treatment; mutation treatment may be the sudden change of a certain gene of an individual to another type, and in the array may be the change of a certain element to another value, for example, the array B [2,7,5,9,3] becomes B' [2,7,6,9,3] after mutation treatment.
In one embodiment, the step of processing the iterative optimal solution of the current population by using a greedy algorithm to obtain the locally optimal individual comprises the following steps:
processing the iterative optimal solution of the current population by adopting a greedy algorithm to obtain a local optimal solution;
determining individuals corresponding to the local optimal solution meeting the preset precision as local optimal individuals;
if the local optimal solution does not meet the preset precision, eliminating the individuals with low fitness and high concentration of the current population by adopting an immune algorithm; and increasing the preset population evolution iteration times according to the eliminated number of individuals until a local optimal solution meeting the preset precision is obtained.
Specifically, the traditional genetic algorithm has low convergence speed, and a greedy algorithm is introduced, namely, the global optimum is constructed by trying to construct a local optimum solution without considering the overall optimum, and the global optimum solution is approached step by step. The newly obtained objective function value of the individual can be compared with a preset precision, and if the objective function value of the individual is within the range of the preset precision, the individual corresponding to the local optimal solution meeting the preset precision is determined as a local optimal individual; if the population is not in the preset precision range, the current population is processed by adopting an immune algorithm, namely the probability that the individuals are replaced is calculated according to the fitness and the concentration of the individuals of the current population, so that the individuals with low fitness and high concentration of the current population are eliminated, and meanwhile, the preset population evolution iteration times are increased according to the number of the eliminated individuals, so that the initialized population can continue to evolve, and the current population is updated.
In some examples, the greedy algorithm may be to take an iterative optimal solution of the current population, randomly extract a single gene of its individual to perform random transformation, and then calculate an objective function value of the transformed individual, if the objective function value of the transformed individual is greater than the objective function value before transformation, the transformed gene value is retained, otherwise, the gene value before transformation is restored, and if the transformation is repeatedly extracted for several times (e.g., 3 to 5 times), it may be considered to complete one local optimization (obtaining a local optimal solution). The immune algorithm can find out the individual with low fitness and high concentration for replacement by selectively eliminating the individual, namely by using the working process of an immune system of an organism. The fitness is a factor for measuring the fit degree of a feasible solution and a problem to be solved, the objective function or the reciprocal value of the objective function is generally used for judging, the concentration is a factor for measuring the diversity of the population, and the individual similarity, the population quantity and other parameters are needed to be used for calculation. Can be realized by setting a probability threshold value P k And calculating the probability P of each individual being replaced by concentration and fitness i . The calculated probability P of each body being replaced i With a probability threshold value P k Making a comparison if P i ≤P k Then continuing evolution iteration; if P i ＞P k If the new individual is generated, the probability of local convergence is effectively reduced, and a basis is provided for finding a more suitable internal parameter of the filter to be optimized.
By introducing the immune algorithm and the local optimization with the greedy algorithm thinking, the embodiment of the application can avoid the defect of error caused by using mathematical preconditions which may not be satisfied in reality in the traditional analysis design method, and overcomes the defects that the conventional genetic algorithm is easy to fall into local convergence and has low convergence speed.
In one embodiment, the objective function value of an individual is obtained by filtering each harmonic amplitude value through linear normalization processing; and each subharmonic amplitude is obtained through a simulation result of a power grid simulation model accessed to the filter to be optimized.
Specifically, the filtering effect of the filter to be optimized may be determined by an objective function that comprehensively considers the total harmonic wave after filtering and the harmonic wave with the maximum amplitude, for example, the objective function of the filtering effect of the filter to be optimized may be determined by processing the amplitude of each harmonic wave after filtering through linear normalization. Each individual (namely, a filter to be optimized) is accessed into a power grid to calculate an objective function value, and the optimal objective function value and the corresponding individual parameter thereof need to be stored after each calculation is finished.
In some examples, the individual objective function values may be obtained using the following formula:
in the formula, F i An objective function value for the ith individual; alpha is a weighting coefficient; mapminmax is linear normalization processing; u shape 1 Fundamental amplitude that is a harmonic; n is the harmonic frequency; u shape j Represents the amplitude of the jth harmonic; u shape max The harmonic amplitude with the maximum amplitude;
for measuring the total harmonic amplitude after filtering,
for weighing the harmonic with the highest amplitude after filtering. Since it is desirable that the harmonic component is as small as possible, it is possible to cope with F i Calculating the reciprocal as the value of the objective function, F i Greater generation of reciprocal valueThe better the filter will work.
The fitness of each individual of the locally optimal cluster can be obtained by adopting the following formula:
in the formula, R i The fitness of the ith individual; f ij The objective function value of the ith individual under the operation mode of the jth type power grid is obtained; m is the total number of the operation modes of the power grid simulation model; the individual with the smallest fitness value in the locally optimal cluster can be determined as the globally optimal individual.
According to the method and the device, the artificial intelligence algorithm is adopted, when the individual objective function is set, the overall harmonic component and the prominent individual harmonic component are comprehensively considered, and therefore the overall harmonic component can be considered to filter out the full-band harmonic component as far as possible; in addition, the genetic algorithm is combined with the immune algorithm, the situation of falling into local optimum can be avoided as much as possible, and meanwhile, the local optimization by introducing the thought of the greedy algorithm is adopted, so that rapid convergence is realized.
It should be understood that, although the steps in the flowcharts related to the embodiments as described above are sequentially displayed as indicated by arrows, the steps are not necessarily performed sequentially as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a part of the steps in the flowcharts related to the embodiments described above may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the execution order of the steps or stages is not necessarily sequential, but may be rotated or alternated with other steps or at least a part of the steps or stages in other steps.
Based on the same inventive concept, the embodiment of the present application further provides a filter parameter design apparatus for implementing the above-mentioned filter parameter design method. The implementation scheme for solving the problem provided by the apparatus is similar to the implementation scheme described in the above method, so the specific limitations in one or more embodiments of the filter parameter design apparatus provided below may refer to the limitations on the filter parameter design method in the foregoing, and are not described herein again.
In one embodiment, as shown in fig. 5, a filter parameter design apparatus is provided. The device comprises:
the population obtaining module 510 is configured to generate an initialization population according to external parameters of a filter to be optimized based on a current operation mode of the power grid simulation model; the external parameters comprise a filter placement position and a filter structure determined according to a power grid simulation model in a full-wiring mode;
an individual obtaining module 520, configured to process the initialized population by using an artificial intelligence model, so as to obtain a locally optimal individual corresponding to the current operation mode;
a cluster obtaining module 530, configured to switch the current operation mode until each operation mode of the power grid simulation model is traversed, obtain locally optimal individuals respectively corresponding to each operation mode, and obtain a locally optimal cluster of each locally optimal individual;
a parameter obtaining module 540, configured to obtain a global optimal individual based on the objective function values of the individuals of the local optimal cluster; the globally optimal individuals are used to characterize the internal parameters of the filter to be optimized.
In one embodiment, the individual acquisition module is further configured to iteratively update the initialized population by using a genetic algorithm to obtain a current population; processing the iterative optimal solution of the current population by adopting a greedy algorithm to obtain a local optimal individual; the iterative optimal solution is determined according to the objective function values of the individuals of the current population.
In one embodiment, the individual obtaining module is further configured to perform intersection and mutation processing on the last initialization population until a new individual with a target function value better than the last initialization population is obtained, so as to update the initialization population to obtain the current initialization population, if the current evolution iteration number is smaller than the preset evolution iteration number.
In one embodiment, the individual acquisition module is further configured to process an iterative optimal solution of the current population by using a greedy algorithm to obtain a local optimal solution; determining individuals corresponding to the local optimal solution meeting the preset precision as local optimal individuals; if the local optimal solution does not meet the preset precision, eliminating the individuals with low fitness and high concentration of the current population by adopting an immune algorithm; and increasing the preset population evolution iteration times according to the eliminated number of individuals until a local optimal solution meeting the preset precision is obtained.
The modules in the filter parameter design device can be wholly or partially implemented by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided. The computer device comprises a memory storing a computer program and a processor implementing the steps of the method described above when executing the computer program.
In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 6. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a filter parameter design method. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
Those skilled in the art will appreciate that the architecture shown in fig. 6 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer-readable storage medium is provided. The computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method described above.
In one embodiment, a computer program product is provided. The computer program product comprises a computer program which, when being executed by a processor, carries out the steps of the above-mentioned method.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, databases, or other media used in the embodiments provided herein can include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high-density embedded nonvolatile Memory, resistive Random Access Memory (ReRAM), Magnetic Random Access Memory (MRAM), Ferroelectric Random Access Memory (FRAM), Phase Change Memory (PCM), graphene Memory, and the like. Volatile Memory can include Random Access Memory (RAM), external cache Memory, and the like. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others. The databases referred to in various embodiments provided herein may include at least one of relational and non-relational databases. The non-relational database may include, but is not limited to, a block chain based distributed database, and the like. The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, quantum computing based data processing logic devices, etc., without limitation.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.
Claims (10)
1. A method for filter parameter design, the method comprising:
generating an initialization population according to external parameters of a filter to be optimized based on the current operation mode of the power grid simulation model; the external parameters comprise a filter placement position and a filter structure determined by the power grid simulation model in a full-wiring mode;
processing the initialization population by adopting an artificial intelligence model to obtain a local optimal individual corresponding to the current operation mode;
switching the current operation mode until traversing each operation mode of the power grid simulation model to obtain the local optimal individuals respectively corresponding to each operation mode, and acquiring a local optimal cluster of each local optimal individual;
acquiring a global optimal individual based on the objective function value of each individual of the local optimal cluster; the global optimal individual is used for representing internal parameters of the filter to be optimized.
2. The method of claim 1, wherein the step of processing the initialization population using an artificial intelligence model to obtain locally optimal individuals for the current operating mode comprises:
iteratively updating the initialized population by adopting a genetic algorithm to obtain a current population; processing the iterative optimal solution of the current population by adopting a greedy algorithm to obtain the local optimal individual; the iterative optimal solution is determined according to the objective function values of the individuals of the current population.
3. The method according to claim 1 or 2, characterized in that the filter placement position is determined by processing the total harmonic distortion of each bus voltage with a threshold to be filtered; the total harmonic distortion is obtained through a simulation result of a power grid simulation model in a full-wiring mode;
the filter structure is determined according to the number of the harmonic ranges by dividing the harmonic to be filtered in the total harmonic distortion into a plurality of harmonic ranges.
4. The method of claim 2, wherein the step of iteratively updating the initialization population using a genetic algorithm to obtain the current population comprises:
and if the current evolution iteration number is less than the preset evolution iteration number, performing cross and mutation processing on the last initialization population until a new individual with a target function value superior to that of the last initialization population is obtained, so as to update the initialization population to obtain the current initialization population.
5. The method of claim 2, wherein the step of processing the iterative optimal solution of the current population using a greedy algorithm to obtain the locally optimal individual comprises:
processing the iterative optimal solution of the current population by adopting a greedy algorithm to obtain a local optimal solution;
determining individuals corresponding to the local optimal solution meeting the preset precision as local optimal individuals;
if the local optimal solution does not meet the preset precision, eliminating the individuals with low fitness and high concentration of the current population by adopting an immune algorithm; and increasing the preset population evolution iteration times according to the eliminated number of individuals until the local optimal solution meeting the preset precision is obtained.
6. The method of any one of claims 1 to 5, wherein the objective function values of the individuals are obtained for respective harmonic amplitudes filtered by a linear normalization process; and the amplitude of each subharmonic is obtained through a simulation result of the power grid simulation model accessed to the filter to be optimized.
7. An apparatus for filter parameter design, the apparatus comprising:
the population acquisition module is used for generating an initialization population according to external parameters of the filter to be optimized based on the current operation mode of the power grid simulation model; the external parameters comprise a filter placement position and a filter structure determined by the power grid simulation model in a full-wiring mode;
the individual acquisition module is used for processing the initialization population by adopting an artificial intelligence model to obtain a local optimal individual corresponding to the current operation mode;
the cluster acquisition module is used for switching the current operation mode until traversing each operation mode of the power grid simulation model to obtain the local optimal individuals respectively corresponding to each operation mode and obtain the local optimal cluster of each local optimal individual;
the parameter acquisition module is used for acquiring a global optimal individual based on the objective function value of each individual of the local optimal cluster; the global optimal individual is used for representing internal parameters of the filter to be optimized.
8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 6.
9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.
10. A computer program product comprising a computer program, characterized in that the computer program realizes the steps of the method of any one of claims 1 to 6 when executed by a processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210536043.6A CN114818509A (en) | 2022-05-17 | 2022-05-17 | Filter parameter design method, filter parameter design device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210536043.6A CN114818509A (en) | 2022-05-17 | 2022-05-17 | Filter parameter design method, filter parameter design device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114818509A true CN114818509A (en) | 2022-07-29 |
Family
ID=82515209
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210536043.6A Pending CN114818509A (en) | 2022-05-17 | 2022-05-17 | Filter parameter design method, filter parameter design device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114818509A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116579222A (en) * | 2023-07-12 | 2023-08-11 | 深圳飞骧科技股份有限公司 | Optimization method, system and related equipment for parameters of surface acoustic wave filter |
| CN117252136A (en) * | 2023-11-14 | 2023-12-19 | 高拓讯达(北京)微电子股份有限公司 | Data processing method and device for filter parameters, electronic equipment and storage medium |
-
2022
- 2022-05-17 CN CN202210536043.6A patent/CN114818509A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116579222A (en) * | 2023-07-12 | 2023-08-11 | 深圳飞骧科技股份有限公司 | Optimization method, system and related equipment for parameters of surface acoustic wave filter |
| CN116579222B (en) * | 2023-07-12 | 2024-02-06 | 深圳飞骧科技股份有限公司 | Optimization method, system and related equipment for parameters of surface acoustic wave filter |
| CN117252136A (en) * | 2023-11-14 | 2023-12-19 | 高拓讯达(北京)微电子股份有限公司 | Data processing method and device for filter parameters, electronic equipment and storage medium |
| CN117252136B (en) * | 2023-11-14 | 2024-02-27 | 高拓讯达(北京)微电子股份有限公司 | Data processing method and device for filter parameters, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114818509A (en) | Filter parameter design method, filter parameter design device, computer equipment and storage medium | |
| Wang et al. | A Pareto-based estimation of distribution algorithm for the multi-objective flexible job-shop scheduling problem | |
| Yu et al. | Zinb-based graph embedding autoencoder for single-cell rna-seq interpretations | |
| Hu | Forecasting tourism demand using fractional grey prediction models with Fourier series | |
| Zhang et al. | A multi-objective memetic algorithm based on decomposition for big optimization problems | |
| Naik et al. | Genetic algorithm-aided dynamic fuzzy rule interpolation | |
| Saoud et al. | Household energy consumption prediction using the stationary wavelet transform and transformers | |
| Zhang et al. | G-CoS: GNN-accelerator co-search towards both better accuracy and efficiency | |
| CN114266421B (en) | New energy power prediction method based on composite meteorological feature construction and selection | |
| Zhang et al. | Load prediction based on hybrid model of VMD-mRMR-BPNN-LSSVM | |
| Wang et al. | An improved hybrid algorithm based on biogeography/complex and metropolis for many-objective optimization | |
| Gan et al. | Automated synthesis of passive analog filters using graph representation | |
| Gao et al. | Online dynamic ensemble deep random vector functional link neural network for forecasting | |
| Yi et al. | Intergroup cascade broad learning system with optimized parameters for chaotic time series prediction | |
| Guo et al. | Drift analysis of mutation operations for biogeography-based optimization | |
| Moradi Far et al. | Cost-effective optimal allocation and sizing of active power filters using a new fuzzy-MABICA method | |
| Mofrad et al. | Efficient distributed graph analytics using triply compressed sparse format | |
| Pourtaheri et al. | Stability investigation of multi-objective heuristic ensemble classifiers | |
| Awad et al. | Multi-objective differential evolution based on normalization and improved mutation strategy | |
| Chai et al. | A node-priority based large-scale overlapping community detection using evolutionary multi-objective optimization | |
| Jiang et al. | Multi-objective optimization for community detection in multilayer networks | |
| Xu et al. | A fast two-objective differential evolution for the two-objective coverage problem of WSNs | |
| CN108256694A (en) | Based on Fuzzy time sequence forecasting system, the method and device for repeating genetic algorithm | |
| Barmada et al. | An evolutionary algorithm for global optimization based on self-organizing maps | |
| CN114118608A (en) | Power grid business matching optimization method and device, terminal and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |