Flywheel energy storage converter control method

Abstract

The invention relates to the technical field of control methods, in particular to a flywheel energy storage converter control method, which comprises the steps of obtaining a parameter adjustment amount of a flywheel energy storage converter by using a fuzzy PID control algorithm, and adjusting parameters of the converter according to the parameter adjustment amount. According to the invention, fuzzification, fuzzy reasoning and clarification can be carried out on the data obtained in the PID control process, so that the control performance of the PID algorithm can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of control methods, and in particular to a flywheel energy storage converter control method. Background Art

[0002] A flywheel energy storage converter is a power electronic device used in a flywheel energy storage system. Its main function is to convert the direct current of the flywheel energy storage system into alternating current and manage the flow of electric energy. The flywheel energy storage system stores energy through a high-speed rotating flywheel and releases the energy to the grid or load when needed. When the flywheel energy storage converter is connected to the grid, it needs to provide a stable voltage and be able to respond quickly. There is an existing fast voltage stabilization algorithm based on PID control.

[0003] A Chinese patent document with announcement number CN106444357B discloses a variable domain fuzzy PID dual hydraulic cylinder electro-hydraulic servo synchronous control method, which includes the following steps: designing a domain expansion factor that can be adaptively adjusted according to the size of input and output variables; selecting different input variables for the free channel and the adjustment channel, and selecting ΔKP and ΔKI as output variables; fuzzifying the input variables; selecting a triangular membership function at the domain zero point and a Gaussian membership function near the fuzzy domain boundary; formulating a fuzzy rule table; performing fuzzy reasoning using the Mamdani rule; clarifying the fuzzy quantity, outputting it to the controlled object, and controlling it.

[0004] Through the above scheme, it can be known that the fuzzy PID method can be used to control the voltage of the flywheel energy storage inverter. In the process of fuzzy processing of data by the fuzzy PID method, the membership function is needed. The membership rule plays an important role in the result of fuzzy processing. The triangle membership function is selected in the above scheme, and the triangle membership function is not the best membership function during use. Different types of membership functions have different performances for the controller when performing voltage control. How to obtain a better membership function to optimize the fuzzy control results is a problem that needs to be solved now. Summary of the invention

[0005] In order to solve the problem of how to obtain a better membership function optimization fuzzy control result, the present invention provides a flywheel energy storage converter control method.

[0006] The present invention provides a flywheel energy storage converter control method, which adopts the following technical solution: The parameter adjustment amount of the flywheel energy storage converter is obtained by using the fuzzy PID control algorithm, and the converter parameters are adjusted according to the parameter adjustment amount; Among them, the construction method of the membership function curve in the fuzzy PID control algorithm is: Construct a rectangular coordinate system, with the horizontal axis of the coordinate system being the domain and the vertical axis being the membership value, and set the solution set of each language value in the coordinate system; The distribution area of ​​the data points in the solution set is divided into two sub-areas, the upper envelope and the lower envelope of the sub-areas are drawn, and the average of the upper envelope and the lower envelope is calculated to obtain the mean envelope; Calculate the priority of the data points in the solution set. The priority expression of the data points is:

[0007] In the formula, For the The solution set corresponding to the language value The priority of each data point, For the The discrete coefficient of the data points in the solution set region corresponding to the linguistic value is For the The solution set corresponding to the language value The update value coefficient of the data point, () represents an exponential function with base e; The priorities of the data points in the sub-region are clustered to obtain high-priority clusters, and the positions of the data points in the high-priority clusters are updated to obtain updated upper envelopes, lower envelopes and mean envelopes, and the updated mean envelope is used as the membership function curve.

[0008] The effect is that the membership function curve is obtained by updating the solution data, so that the data obtained in the PID control process can be fuzzified, fuzzy inferred and clarified through the membership function curve, thereby improving the control performance of the PID algorithm.

[0009] Preferably, the method for updating the position of the data point in the sub-region is: Construct the objective function, the objective function The expression is:

[0010] Where: Q is the overlap rate of the membership functions of two adjacent language values, is the maximum area between the upper and lower envelopes of the sub-region, It is the PID algorithm control performance index; The swarm intelligence optimization algorithm is used to solve the minimum value of the objective function to update the position of the data points in the high priority cluster; Until the area enclosed by the updated upper envelope and the lower envelope is smaller than a preset threshold, the updating of the position of the data point in the high priority cluster is stopped.

[0011] The effect is that, by updating the position of the data point, the area between the upper envelope and the lower envelope gradually converges, thereby obtaining a more accurate membership function curve.

[0012] Preferably, the control performance indicator is the sum of the rise time and the transition time.

[0013] Preferably, the swarm intelligence optimization algorithm is one of a genetic algorithm, an ant colony optimization algorithm, a particle swarm optimization algorithm, a firefly swarm optimization algorithm, and a cuckoo search algorithm.

[0014] Preferably, the method for obtaining the distribution area of ​​the data points in the solution set is: A two-dimensional normal distribution model of the data points in the solution set is constructed, and the intersection line of two adjacent two-dimensional normal distribution models is used as the boundary of the language value. The area between the two boundaries is the distribution area of ​​the data points in the solution set of the language value.

[0015] The effect is that the distribution area of ​​the data points in the solution set is obtained by determining the boundaries between the language values, making the language values ​​independent of each other.

[0016] Preferably, the expression of the update value coefficient is:

[0017] In the formula, For the The solution set corresponding to the language value The update value coefficient of the data point, is the hyperbolic tangent function, For the The solution set corresponding to the language value Data points and The slope value of the data point, For the The solution set corresponding to the language value Data points and The slope value of the data point, For the The solution set corresponding to the language value Data points and The Euclidean distance value of the data points is For the The solution set corresponding to the language value Data points and The Euclidean distance between the data points.

[0018] The effect is that by calculating the update value coefficient, it is easier to calculate the priority of the data point and improve the accuracy of the calculation.

[0019] Preferably, the expression of the update value coefficient is:

[0020] In the formula, For the The solution set corresponding to the language value The update value coefficient of the data point, is the hyperbolic tangent function, For the The solution set corresponding to the language value Data points and The slope value of the data point, For the The solution set corresponding to the language value Data points and The slope value of the data point, For the The solution set corresponding to the language value Data points and The Euclidean distance value of the data points is For the The solution set corresponding to the language value Data points and The Euclidean distance value of the data points is () represents an exponential function with base e.

[0021] Preferably, the method of dividing the distribution area of ​​the data points in the solution set into two sub-areas is: obtaining the data point with the maximum membership value in the solution set; and dividing the solution set into two sub-areas using the data point with the maximum membership value as a dividing point.

[0022] The effect is that the distribution area is divided into two sub-areas, and the membership function curves are drawn separately, thereby improving the accuracy of the membership function curve.

[0023] The present invention has the following technical effects: The envelope is obtained by constructing a solution set, and the swarm intelligence optimization algorithm is used to solve the objective function to update the position of the data points to reduce the area between the envelopes. When constructing the objective function, the overlap rate of the membership function of the language value and the control performance index are considered to obtain a better membership function, which can improve the control performance of the PID algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts.

[0025] Figure 1 The present invention is a method flow chart of a flywheel energy storage converter control method.

[0026] Figure 2 It is a flow chart of the method for constructing the membership function of each language value of the present invention.

[0027] Figure 3 It is a schematic diagram of a membership function generated by the present invention. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0029] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.

[0030] The embodiment of the present invention discloses a flywheel energy storage converter control method, referring to Figure 1 , including steps S1-S2, which are as follows: S1: Determine the membership function curve of each language value in the fuzzy PID control algorithm.

[0031] In the fuzzy PID control algorithm, the language values ​​include negative large NB, negative middle NM, negative small NS, zero ZO, positive small PS, positive middle PM, and positive large PB. The language values ​​are also called fuzzy subsets. The fuzzy algorithm in the fuzzy PID control algorithm is used to fuzzify, fuzzy reasoning and defuzzify the obtained data. This is an existing technology and the specific process will not be repeated here.

[0032] like Figure 2 As shown, the method for constructing the membership function of each language value is: S11: Determine the location of the domain where the language value is located.

[0033] Construct a rectangular coordinate system, the horizontal axis of the rectangular coordinate system is the domain, the vertical axis is the membership value, and arrange the language values ​​in the rectangular coordinate system in sequence.

[0034] S12: Setting a corresponding solution set for the language value, dividing the distribution area of ​​the data points in the solution set into a plurality of sub-areas, each of which contains a plurality of data points.

[0035] Corresponding solution sets are randomly constructed at the language values. The solution sets contain multiple data points. The connecting lines of the multiple data points are in a single-peak structure as a whole. The connecting curve at this time is the membership function that is initially set.

[0036] The method of dividing the distribution area of ​​the data points in the solution set into multiple sub-areas is as follows: obtaining the data point with the maximum membership in the solution set; using the data point with the maximum membership as the dividing point to divide the solution set into two sub-areas, specifically, drawing a vertical line along the data point with the maximum membership, the vertical line is the dividing line of the sub-areas in the solution set, and the dividing line divides the connecting curve into two left and right line segments. It should be noted that when there are multiple identical maximum values ​​in the solution set, the leftmost data point and the rightmost data point are used as the dividing line, and the data points in the middle are discarded to divide the solution set into two sub-areas.

[0037] The method for obtaining the distribution area of ​​the data points in the solution set is: calculate the priority of each data point in the solution set of each language value. The priority represents the importance of the data point. The priority expression of the data point is: , where For the The solution set corresponding to the language value The priority of each data point, For the The discrete coefficient of the data points in the solution set region corresponding to the linguistic value is For the The solution set corresponding to the language value The update value coefficient of the data point, () represents an exponential function with base e.

[0038] In one embodiment, , where For the The solution set corresponding to the language value The update value coefficient of the data point, is the hyperbolic tangent function, For the The solution set corresponding to the language value Data points and The slope value of the data point, For the The solution set corresponding to the language value Data points and The slope value of the data point, For the The solution set corresponding to the language value Data points and The Euclidean distance value of the data points is For the The solution set corresponding to the language value Data points and The Euclidean distance value of the data points is () represents an exponential function with base e.

[0039] In other embodiments, , where For the The solution set corresponding to the language value The update value coefficient of the data point, is the hyperbolic tangent function, For the The solution set corresponding to the language value Data points and The slope value of the data point, For the The solution set corresponding to the language value Data points and The slope value of the data point, For the The solution set corresponding to the language value Data points and The Euclidean distance value of the data points is For the The solution set corresponding to the language value Data points and The Euclidean distance value of the data points is () represents an exponential function with base e.

[0040] The larger the value of , the greater the deviation between the membership function curve and the ideal membership function curve, and the lower the priority of the data point on the corresponding membership curve. The calculation method of is as follows: obtain part of the curve in the current sub-region, obtain the mean envelope of the sub-region (the method of obtaining the mean envelope will be described below), obtain the cumulative distance matrix between the sub-region membership function curve and the mean envelope, and use the DTW (Dynamic Time Warping) algorithm to obtain the difference between each data point and the slope in the shortest path in the cumulative distance matrix. The difference obtained is , The larger it is, the lower the overlap rate between the sub-region membership function curve and the mean envelope curve, and the worse the similarity. The smaller it is, the data point at that location is not suitable for the membership function to converge, and the area near the data point cannot be used for membership function generation; The larger the value is, the more suitable the data point is for the membership function to converge, and the area near the data point can be used for membership function generation.

[0041] According to the priority of the data points, the priority mean of the solution set is obtained; according to the priority mean and the mean of the domain of the data points in the solution set, a two-dimensional normal distribution model of the corresponding language value is constructed, that is, the priority mean is the y mean in the two-dimensional normal distribution model, and the mean of the domain of the data points is the x mean in the two-dimensional normal distribution model. The specific construction process is the existing technology and will not be repeated here; according to the two-dimensional normal distribution model, the boundary between the language values ​​is determined, the interface between the two two-dimensional normal distribution models is the dividing interface, and the area between the two boundary surfaces is the data point distribution area in the solution set of the language value.

[0042] S13: Draw an upper envelope line and a lower envelope line for each sub-region, and calculate the area of ​​the region between the upper envelope line and the lower envelope line.

[0043] The method is: find extreme values, determine the local maximum and minimum values ​​of the data points, and use these points as candidate points of the envelope; draw the envelope, connect the extreme points to form the envelope, and use line segments or curves to connect the extreme points to form upper and lower envelopes.

[0044] S14: Iteratively updating the positions of the data points in the sub-region, and updating the corresponding upper envelope and lower envelope, until the area enclosed by the upper envelope and the lower envelope is smaller than a preset threshold.

[0045] The method for updating the position of the data points in the sub-region is: clustering the priorities of the data points in the solution set by the k-means algorithm to obtain two clusters, one of which is a high-priority cluster and the other is a low-priority cluster, discarding the high-priority cluster and retaining the low-priority cluster; constructing the objective function, the objective function The expression is: , where Q is the overlap rate of the membership functions of two adjacent language values, specifically, the ratio of the intersection length of the two membership functions to the minimum value of the membership function width (the width of the membership function with the smaller width), is the maximum area between the upper and lower envelopes of the sub-region, It is a PID algorithm control performance index, and the performance index includes rise time and transition time. Specifically, the control performance index is the sum of rise time and transition time. The swarm intelligence optimization algorithm is used to solve the minimum value of the objective function to update the position of the data point in the high priority cluster. At the same time, the upper envelope and the lower envelope are updated until the area surrounded by the upper envelope and the lower envelope is less than the preset threshold. The swarm intelligence optimization algorithm is one of the genetic algorithm, ant colony optimization algorithm, particle swarm optimization algorithm, firefly swarm optimization, and cuckoo search algorithm.

[0046] S15: Calculate the average of the upper envelope and the lower envelope to obtain a mean envelope, and connect multiple mean envelopes as a membership function curve of the language value.

[0047] For a language value, it corresponds to two sub-areas, and correspondingly has two mean envelopes. Connecting the two mean envelopes is the final membership function curve of the corresponding language value, such as Figure 3 shown.

[0048] S2: Obtain parameter adjustment values ​​of the flywheel energy storage converter by using a fuzzy PID control algorithm, and adjust the converter parameters according to the parameter adjustment values.

[0049] After obtaining the membership function curve, the obtained data is fuzzified, fuzzy inferred and clarified to obtain the PID algorithm parameter adjustment amount to control the parameters of the flywheel energy storage rheometer, including voltage and current.

[0050] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

[0051] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A flywheel energy storage converter control method, characterized in that: Includes steps: The parameter adjustment amount of the flywheel energy storage converter is obtained by using the fuzzy PID control algorithm, and the converter parameters are adjusted according to the parameter adjustment amount; Among them, the construction method of the membership function curve in the fuzzy PID control algorithm is: Construct a rectangular coordinate system, with the horizontal axis of the coordinate system being the domain and the vertical axis being the membership value, and set the solution set of each language value in the coordinate system; The distribution area of ​​the data points in the solution set is divided into two sub-areas, the upper envelope and the lower envelope of the sub-areas are drawn, and the average of the upper envelope and the lower envelope is calculated to obtain the mean envelope; Calculate the priority of the data points in the solution set. The priority expression of the data points is: In the formula, For the The solution set corresponding to the language value The priority of each data point, For the The discrete coefficient of the data points in the solution set region corresponding to the linguistic value is For the The solution set corresponding to the language value The update value coefficient of the data point, () represents an exponential function with base e; The priorities of the data points in the sub-region are clustered to obtain high-priority clusters, and the positions of the data points in the high-priority clusters are updated to obtain updated upper envelopes, lower envelopes and mean envelopes, and the updated mean envelope is used as the membership function curve.

2. A flywheel energy storage converter control method according to claim 1, characterized in that: The expression for updating the value coefficient is: In the formula, For the The solution set corresponding to the language value The update value coefficient of the data point, is the hyperbolic tangent function, For the The solution set corresponding to the language value Data points and The slope value of the data point, For the The solution set corresponding to the language value Data points and The slope value of the data point, For the The solution set corresponding to the language value Data points and The Euclidean distance value of the data points is For the The solution set corresponding to the language value Data points and The Euclidean distance between the data points.

3. A flywheel energy storage converter control method according to claim 1, characterized in that: The expression for updating the value coefficient is: In the formula, For the The solution set corresponding to the language value The update value coefficient of the data point, is the hyperbolic tangent function, For the The solution set corresponding to the language value Data points and The slope value of the data point, For the The solution set corresponding to the language value Data points and The slope value of the data point, For the The solution set corresponding to the language value Data points and The Euclidean distance value of the data points is For the The solution set corresponding to the language value Data points and The Euclidean distance value of the data points is () represents an exponential function with base e.

4. A flywheel energy storage converter control method according to claim 1, characterized in that: The method to update the position of data points in the high priority cluster is: Construct the objective function, the objective function The expression is: Where: Q is the overlap rate of the membership functions of two adjacent language values, is the maximum area between the upper and lower envelopes of the sub-region, It is the PID algorithm control performance index; The swarm intelligence optimization algorithm is used to solve the minimum value of the objective function to update the position of the data points in the high priority cluster; Until the area enclosed by the updated upper envelope and the lower envelope is smaller than a preset threshold, the updating of the position of the data point in the high priority cluster is stopped.

5. A flywheel energy storage converter control method according to claim 4, characterized in that: The control performance index is the sum of rise time and transition time.

6. A flywheel energy storage converter control method according to claim 4, characterized in that: The swarm intelligence optimization algorithm is one of the genetic algorithm, ant colony optimization algorithm, particle swarm optimization algorithm, firefly swarm optimization, and cuckoo search algorithm.

7. A flywheel energy storage converter control method according to claim 1, characterized in that: The method for determining the distribution area of ​​data points in the solution set is: A two-dimensional normal distribution model of the data points in the solution set is constructed, and the intersection line of two adjacent two-dimensional normal distribution models is used as the boundary of the language value. The area between the two boundaries is the distribution area of ​​the data points in the solution set of the language value.

8. A flywheel energy storage converter control method according to claim 1, characterized in that: The method of dividing the distribution area of ​​the data points in the solution set into two sub-areas is: obtaining the data point with the maximum membership value in the solution set; and dividing the solution set into two sub-areas by using the data point with the maximum membership value as the dividing point.

Images