CN112068429A - SFCS (sparse form-factor correction) algorithm-based sliding mode controller parameter setting method and device and storage medium - Google Patents
SFCS (sparse form-factor correction) algorithm-based sliding mode controller parameter setting method and device and storage medium Download PDFInfo
- Publication number
- CN112068429A CN112068429A CN202010906950.6A CN202010906950A CN112068429A CN 112068429 A CN112068429 A CN 112068429A CN 202010906950 A CN202010906950 A CN 202010906950A CN 112068429 A CN112068429 A CN 112068429A
- Authority
- CN
- China
- Prior art keywords
- algorithm
- bird nest
- sfcs
- mode controller
- parameter setting
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
Abstract
The invention discloses a sliding mode controller parameter setting method, a sliding mode controller parameter setting device and a storage medium based on an SFCS (sparse representation Circuit) algorithm, wherein an explosion operator with self-adaptive capacity and a reverse learning strategy based on the worst solution of a population are added in the traditional CS algorithm, so that the populations can communicate with each other, and the population can be further guided to evolve towards the direction of the optimal solution; in the convergence aspect, the influence of an explosion operator and Levier flight on the generation of a candidate solution of the algorithm is balanced by the population growth curve, so that the global exploration and local development capability of the algorithm is balanced; in addition, when the algorithm is in the later iteration stage, the explosion operator is dominant, and the convergence speed and precision of the algorithm are effectively improved; and the SFCS algorithm is applied to parameter setting of the sliding mode controller, so that the parameter optimization setting is well realized, the time for manually adjusting the parameters is saved, and a better control effect is obtained.
Description
Technical Field
The invention relates to the field of sliding mode controllers of bridge cranes, in particular to a sliding mode controller parameter setting method and device based on an SFCS (small form-factor circuit switching) algorithm and a storage medium.
Background
The bridge crane system is often used in the transportation of all kinds of heavy industry place and harbour goods, and the during operation relies on the lifting rope that the platform truck hung to pull the load to the assigned position, and whole process need be when transporting fast, and the load swing of minimizing to improve transport efficiency and safety level. The control difficulty is greatly increased due to the fact that the control quantity dimension of the bridge crane is less than the degree of freedom of the bridge crane and the requirement of accurate positioning of the trolley and suppression of load swing in the operation process. The sliding mode control has the characteristics of quick response, simple design, strong robustness and the like, and is widely applied to the control of an under-actuated system, but the parameter setting process of the sliding mode controller is complex, and the setting result has great influence on the control effect. The parameters of the sliding-mode controller of the bridge crane are mostly debugged manually, wherein the parameters of the controller have certain coupling, so that the parameter adjusting process is complicated, and the finally obtained parameters are not ideal in control effect. Part of scholars use the traditional optimization algorithm to carry out parameter setting on the sliding mode controller, but the traditional optimization algorithm has the problems of low convergence speed, low convergence precision and easy falling into the local optimal solution, so that the parameter setting result is poor.
Disclosure of Invention
The present invention is directed to solving at least one of the problems of the prior art.
Therefore, the invention provides the sliding mode controller parameter setting method based on the SFCS algorithm, overcomes the defects of low later convergence speed and low solving precision of the traditional CS algorithm, improves the searching precision of the algorithm and the later searching speed of the algorithm, and adopts the optimization algorithm to carry out optimization setting, thereby saving the time for manually adjusting the parameters and obtaining better control effect.
The invention also provides a sliding mode controller parameter setting device based on the SFCS algorithm, which applies the sliding mode controller parameter setting method based on the SFCS algorithm.
The invention also provides a computer readable storage medium applying the sliding mode controller parameter setting method based on the SFCS algorithm.
The SFCS algorithm-based sliding mode controller parameter setting method according to the first aspect of the invention comprises the following steps:
step S100, initializing parameters and population, and firstly setting an evolution algebra G, a maximum evolution algebra Gmax, a bird nest number n, a search space dimension D, a found probability Pa, a search space upper limit u and a search space lower limit l;
step S200, randomly generating n bird nest positions and calculating the fitness value of each bird nest position;
step S300, updating each bird nest position by using an inertial weight decreasing algorithm based on an S-shaped function, calculating the fitness of a new bird nest position, comparing the existing bird nest with the previous generation bird nest position, selecting a bird nest position with a better fitness value to replace a bird nest position with a poorer fitness value, and storing the bird nest with the worst fitness value;
step S400, generating a random number Pt of [0,1], if Pt is less than Pa, utilizing a reverse learning algorithm based on a worst population solution to carry out preference random walk to update the position of a bird nest, calculating the fitness value of the bird nest, comparing the position of the existing bird nest with the position of the original bird nest, and selecting the position of the bird nest with a better fitness value to replace the position of the bird nest with a poorer fitness value;
and step S500, if G is less than Gmax, returning to execute the step S300, otherwise, outputting an optimization result of the target function.
The sliding mode controller parameter setting method based on the SFCS algorithm, provided by the embodiment of the invention, has at least the following beneficial effects: the SFCS algorithm is an improved cuckoo search algorithm, and an explosion operator with self-adaptive capacity and a reverse learning strategy based on the worst solution of the population are added in the traditional CS algorithm, so that the populations can communicate with each other, and the population can be further guided to evolve towards the direction of the optimal solution; in the convergence aspect, the influence of an explosion operator and Levier flight on the generation of a candidate solution of the algorithm is balanced by the population growth curve, so that the global exploration and local development capability of the algorithm is balanced; in addition, when the algorithm is in the later iteration stage, the explosion operator is dominant, and the convergence speed and precision of the algorithm are effectively improved; and the SFCS algorithm is applied to parameter setting of the sliding mode controller, so that the parameter optimization setting is well realized, the time for manually adjusting the parameters is saved, and a better control effect is obtained.
According to some embodiments of the invention, the decreasing inertial weight algorithm based on the sigmoid function is represented as:
wherein Zi, j represents the j (j is 1,2, …, n) th candidate solution when the algorithm evolves to the ith generation; zi, g is the global optimal solution searched by the ith generation; η represents the initial search step of the algorithm, and is related to the dimension and complexity of the optimization problem, and is usually equal to 1;
representing point-to-point multiplication. L (lambda) represents the search path of the Laevir flight, w is the inertial weight, AjAnd carrying out the calculation value of the explosion operation on each bird nest for the explosion operator.
According to some embodiments of the invention, the population worst solution based inverse learning algorithm is represented as:
wherein, r and Pt are random numbers subject to uniform distribution, Heaviside (×) is a jump function, Pa is a discovery probability, and Pa is 0.25, Zi, j and Zi, w respectively represent two random feasible solutions of the algorithm evolution to the ith generation.
According to some embodiments of the invention, the objective function is represented as:
wherein F is the optimized driving force u (t), the position deviation signal ζ (t) and the swing angle signal x3(t) of the objective function, and w1, w2 and w3 are weight coefficients.
According to some embodiments of the invention, said AjThe calculation formula of (2) is as follows:
wherein the content of the first and second substances,
to limit the maximum amplitude, f (Zi, j) represents the fitness value of the jth candidate solution when the algorithm evolves to the ith generation, and is a very small constant to avoid the condition that the denominator is zero.
According to some embodiments of the invention, the factor limiting the maximum amplitude
The calculation formula of (2) is as follows:
wherein n is the population number of cuckoos and w is the inertial weight.
According to some embodiments of the present invention, if G < Gmax, returning to perform step S300, otherwise, outputting an optimization result of the objective function, including:
judging whether the G < Gmax condition is satisfied;
when the condition of G < Gmax is satisfied, returning to the step S300; and if the condition that G < Gmax is not satisfied, outputting an optimization result of the objective function.
According to the second aspect of the invention, the sliding mode controller parameter setting device based on the SFCS algorithm comprises:
the device comprises an initialization unit, a search unit and a search unit, wherein the initialization unit is used for initializing parameters and populations, and firstly setting an evolution algebra G, a maximum evolution algebra Gmax, a bird nest number n, a search space dimension D, a found probability Pa, a search space upper limit u and a search space lower limit l;
an operation unit for randomly generating n bird nest positions and calculating a fitness value of each bird nest position;
the first processing unit is used for updating each bird nest position by using an inertial weight decreasing algorithm based on an S-shaped function, calculating the fitness of a new bird nest position, comparing the existing bird nest with the previous generation bird nest position, selecting the bird nest position with a better fitness value to replace the bird nest position with a poorer fitness value, and storing the bird nest with the worst fitness value; the second processing unit is used for generating a random number Pt of [0,1], if Pt is less than Pa, utilizing a reverse learning algorithm based on a worst population solution to carry out preference random walk to update the position of the bird nest, calculating the adaptability value of the bird nest, comparing the position of the existing bird nest with the position of the original bird nest, and selecting the position of the bird nest with a better adaptability value to replace the position of the bird nest with a poorer adaptability value;
and the third processing unit is configured to return to execute step S300 if G < Gmax, and otherwise output an optimization result of the objective function.
According to some embodiments of the invention, the third processing unit comprises:
the judging unit is used for judging whether the G < Gmax condition is satisfied or not;
an execution unit, configured to return to execute step S300 when the G < Gmax condition is satisfied; and if the condition that G < Gmax is not satisfied, outputting an optimization result of the objective function.
The sliding mode controller parameter setting device based on the SFCS algorithm, provided by the embodiment of the invention, at least has the following beneficial effects: the SFCS algorithm is an improved cuckoo search algorithm, and an explosion operator with self-adaptive capacity and a reverse learning strategy based on the worst solution of the population are added in the traditional CS algorithm, so that the populations can communicate with each other, and the population can be further guided to evolve towards the direction of the optimal solution; in the convergence aspect, the influence of an explosion operator and Levier flight on the generation of a candidate solution of the algorithm is balanced by the population growth curve, so that the global exploration and local development capability of the algorithm is balanced; in addition, when the algorithm is in the later iteration stage, the explosion operator is dominant, and the convergence speed and precision of the algorithm are effectively improved; and the SFCS algorithm is applied to parameter setting of the sliding mode controller, so that the parameter optimization setting is well realized, the time for manually adjusting the parameters is saved, and a better control effect is obtained.
According to the computer-readable storage medium of the third aspect of the present invention, the SFCS algorithm-based sliding mode controller parameter tuning method according to the first aspect of the present invention can be applied.
The computer-readable storage medium according to the embodiment of the invention has at least the following advantages: the SFCS algorithm is an improved cuckoo search algorithm, and an explosion operator with self-adaptive capacity and a reverse learning strategy based on the worst solution of the population are added in the traditional CS algorithm, so that the populations can communicate with each other, and the population can be further guided to evolve towards the direction of the optimal solution; in the convergence aspect, the influence of an explosion operator and Levier flight on the generation of a candidate solution of the algorithm is balanced by the population growth curve, so that the global exploration and local development capability of the algorithm is balanced; in addition, when the algorithm is in the later iteration stage, the explosion operator is dominant, and the convergence speed and precision of the algorithm are effectively improved; and the SFCS algorithm is applied to parameter setting of the sliding mode controller, so that the parameter optimization setting is well realized, the time for manually adjusting the parameters is saved, and a better control effect is obtained.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a flowchart of a work flow of a sliding mode controller parameter setting method based on an SFCS algorithm according to a first embodiment of the present invention;
fig. 2 is a diagram of a crane system model in a sliding mode controller parameter setting method based on an SFCS algorithm according to a first embodiment of the present invention;
fig. 3 is a schematic structural diagram of a sliding mode controller parameter setting device based on an SFCS algorithm according to a second embodiment of the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention.
In the description of the present invention, unless otherwise explicitly defined, terms such as arrangement, connection and the like should be broadly construed, and those skilled in the art can reasonably determine the specific meanings of the above terms in the present invention in combination with the detailed contents of the technical solutions.
The traditional cuckoo search algorithm (CS) has the defects of low later-stage search speed and low search precision, and the exploration and development capabilities of the algorithm are unbalanced, however, the parameter setting of the bridge crane controller belongs to a complex optimization problem, and the algorithm is required to have stronger search capability, so that the traditional CS algorithm is improved.
The following is a related description of the cuckoo search algorithm:
cuckoo is a special bird that uses a unique host nest parasitic to breed offspring. During the breeding period of the cuckoo, a host similar to the hatching period and the chick period is searched, and before the host hatches eggs, the eggs which are hatched by the host are placed into nests of the host, so that the parasitic host birds hatch the cuckoo eggs; the host bird has some chance of detecting the foreign bird egg in its own nest and abandoning the nest. When the nest where cuckoo eggs live is abandoned, cuckoos randomly select a nest as a new parasitic nest again. Yang et al, Cambridge university, propose cuckoo search algorithm based on cuckoo incubation parasitic mechanism. The 1 parasitic nest of the algorithm represents 1 feasible solution. The cuckoo algorithm searches the position of a parasitic nest by adopting Levy flight, namely a mathematical expression of a feasible solution Zi, j:
wherein Zi, j represents the j (j ═ 1,2, …, n) th candidate solution when the algorithm evolves to the ith generation; zi, g is the global optimal solution searched by the ith generation; η represents the initial search step of the algorithm, and is related to the dimension and complexity of the optimization problem, and is usually equal to 1;
representing point-to-point multiplication. L (λ) represents a search path of the levy flight.
In an iterative process, the host bird finds the cuckoo hatchling with a certain probability Pt and discards the nest and generates the same number of new nests. This process can be expressed in the CS algorithm as updating the feasible solution with a preferred random walk.
Where, r, Pt is a random number subject to uniform distribution, Heaviside (×) is a jump function, Pa is a discovery probability, and Pa is 0.25, Zi, k and Zi, s respectively represent two random feasible solutions of the algorithm evolution to the ith generation.
The traditional cuckoo search algorithm (CS) has the defects of low later-stage search speed and low search precision, and the exploration and development capabilities of the algorithm are unbalanced, however, the parameter setting of the bridge crane controller belongs to a complex optimization problem, and the algorithm is required to have stronger search capability, so that the traditional CS algorithm needs to be improved.
Referring to fig. 2, a diagram of a model of a crane system, which is a main control object of a sliding mode controller described later, is shown, and the dynamic characteristics of the model can be described by the following dynamic equations:
in the formula, M and M respectively represent the trolley mass and the load mass, the trolley position and the load swing angle are respectively represented by x (t) and theta (t), l and g are the length of the lifting rope and the gravity acceleration, and u (t) is a control signal in the horizontal direction.
Considering the actual working conditions of the bridge crane system, reasonable assumptions are made about the bridge crane system as follows:
assume that 1: the trolley and the load are both solids with uniform mass distribution and can be regarded as particles;
assume 2: the model does not consider various friction forces, air resistance of load and the quality and elastic deformation of the lifting rope;
assume that 3: during transportation, the load swing angle theta (t) is always in the range of (-pi/2, pi/2), namely:
for a bridge crane system, the control objective is to bring the trolley quickly and accurately to the target position pd while substantially damping and eliminating the swinging of the load throughout the process. The target can be quantitatively described as follows:
q＝[x1(t),x2(t),x3(t),x4(t)]T→[pd 0 0 0]T
the dynamic model of the bridge crane system is processed before controller design and stability analysis is initiated. Defining a system state variable of a bridge crane as
The equation of state of the system dynamics model can be expressed as follows:
wherein the nonlinear functions f1(q), b1(q), f2(q) and b2(q) are defined as follows:
example one
Referring to fig. 1, an embodiment of the present invention provides a sliding mode controller parameter tuning method based on an SFCS algorithm, where an embodiment includes, but is not limited to, the following steps:
step S100, initializing parameters and population, and firstly setting an evolution algebra G, a maximum evolution algebra Gma x, a bird nest number n, a search space dimension D, a found probability Pa, a search space upper limit u and a search space lower limit l.
In this embodiment, in this step, initialization processing is first performed on the parameters and the population, and preparation is made for the setting optimization processing of the parameters.
In step S200, n bird nest positions are randomly generated and a fitness value of each bird nest position is calculated.
In this example, this step provides for the subsequent running of an improved cuckoo search algorithm.
And step S300, updating each bird nest position by using an inertial weight decreasing algorithm based on an S-shaped function, calculating the fitness of a new bird nest position, comparing the existing bird nest with the previous generation bird nest position, selecting the bird nest position with a better fitness value to replace the bird nest position with a poorer fitness value, and storing the bird nest with the worst fitness value.
In this embodiment, in order to improve the performance of the CS algorithm, in the embodiment of the present invention, an inertia weight is designed by using the characteristics that the S-type function increases gradually in the algorithm evolution process, increases faster in the middle period, and increases slowly in the early and late periods. The improved strategy enables the Levis flight of the CS algorithm to occupy a dominant position in the initial iteration stage, has stronger global exploration capability, balances the global exploration capability and the local development capability of the algorithm in the later iteration stage, and improves the convergence precision of the algorithm; and (3) an inertia weight decreasing strategy is used for balancing the influence of the Levis flight and explosion operator on the algorithm, namely the global exploration and local development capability of the algorithm is adjusted, and preparation is made for subsequent optimization processing.
And S400, generating a random number Pt of [0,1], if Pt is less than Pa, carrying out preference random walk to update the position of the bird nest by using a reverse learning algorithm based on the worst population solution, calculating the fitness value of the bird nest, comparing the position of the existing bird nest with the position of the original bird nest, and selecting the position of the bird nest with a better fitness value to replace the position of the bird nest with a poorer fitness value.
In this embodiment, in this step, after the nests of the conventional CS algorithm are found and discarded by the host bird, the cuckoo may randomly select a new nest as a new parasitic nest again, and information exchange between populations is lacked in this process, so that a reverse learning strategy based on a worst population solution is applied to a preference random walk process, and by using a guidance function of a candidate solution with a worst fitness value in the population, diversity of the nests can be improved in an early stage of the algorithm search, and the nests can be concentrated near the optimal solution for search in a later stage of the search, thereby improving convergence accuracy of the algorithm; the aim is to realize information interaction when the cuckoo selects the bird nest, namely the preference random walk process of the CS algorithm can perform reverse learning according to the solution with the worst fitness in the population, thereby improving the convergence performance of the algorithm.
And step S500, if G is less than Gmax, returning to execute the step S300, otherwise, outputting an optimization result of the target function.
In this embodiment, the step first determines the condition G < Gmax; when G < Gmax, continuing to return to execute step S300; and when the condition G < Gmax is not satisfied, the optimization result of the objective function is output, and the parameter optimization setting of the sliding mode controller is realized.
In some embodiments of the present invention, the decreasing inertial weight algorithm based on the sigmoid function is represented as:
wherein Zi, j represents the j (j is 1,2, …, n) th candidate solution when the algorithm evolves to the ith generation; zi, g is the global optimal solution searched by the ith generation; η represents the initial search step of the algorithm, and is related to the dimension and complexity of the optimization problem, and is usually equal to 1;
representing point-to-point multiplication. L (lambda) represents the search path of the Laevir flight, w is the inertial weight, AjAnd carrying out the calculation value of the explosion operation on each bird nest for the explosion operator. By utilizing the algorithm, the influence of the Levy flight and explosion operator on the cuckoo search algorithm can be balanced, namely the global search and local development capability of the algorithm is adjusted.
In some embodiments of the present invention, the population worst solution based inverse learning algorithm is represented as:
wherein, r and Pt are random numbers subject to uniform distribution, Heaviside (×) is a jump function, Pa is a discovery probability, and Pa is 0.25, Zi, j and Zi, w respectively represent two random feasible solutions of the algorithm evolution to the ith generation. The aim is to enable the cuckoo to carry out information interaction when the bird nest is selected, namely the preference random walk process of the CS algorithm can carry out reverse learning according to the solution with the worst (highest) fitness in the population, thereby improving the convergence performance of the algorithm.
In some embodiments of the invention, the objective function is expressed as:
wherein F is the optimized driving force u (t), the position deviation signal ζ (t) and the swing angle signal x3(t) of the objective function, and w1, w2 and w3 are weight coefficients. Due to the particularity of the working environment of the crane, the overshoot brings potential danger, and therefore larger penalty coefficients are given to the overshoot position error and the swing angle error. Because of the existence of the time item, the error with the same size appears at the later stage of the simulation, and the penalty factor for the algorithm is larger, so that the system can have a faster running speed and a faster stable speed. In addition, the smaller the value of the objective function F is, the closer the solution obtained by the optimization algorithm is to the global optimal solution is. And when the algorithm reaches the maximum iteration times, ending the optimization of the round, wherein the optimal solution in the population is the parameter value of the sliding mode controller. In order to ensure that a crane system has smaller oscillation performance and faster system response regulation speed in transient response, the embodiment of the invention considers trolley driving force u (t), position deviation signal zeta (t) and swing angle signal x3(t), and adopts control system performance evaluation indexes with good engineering practicability and selectivity: an Integrated Time and Absolute Error (ITAE) SFCS algorithm is used as an objective function for parameter setting of the sliding mode controller. The objective function starts from the actual working environment of the crane on the basis of ITAE, and considers the overshoot condition, so that the algorithm can search safer and more effective control parameters.
In some embodiments of the invention, AjThe calculation formula of (2) is as follows:
wherein the content of the first and second substances,
to limit the maximum amplitude, f (Zi, j) represents the fitness value of the jth candidate solution when the algorithm evolves to the ith generation, and is a very small constant to avoid the condition that the denominator is zero. Introducing an explosion operator to perform explosion operation on each bird nest Zi, j, and performing an explosion process to enter the bird nestThe optimization is further performed, and the diversity of bird nests can be further enhanced by the optimization method, and the searching speed and the optimization accuracy of the algorithm can be improved. To a certain extent, the evolution direction of the bird nest in the algorithm is guided; the explosion operator can search the bird nest with a large adaptability value in a large scale and search the bird nest with a small adaptability value in a small scale, wherein the large-scale search means that a candidate solution of the algorithm can be searched to a range farther than the self, and more possibilities can jump out the local optimum and evolve to a theoretical optimum solution of the adaptability function; the small-scale search means that the candidate solution of the algorithm can be searched near the candidate solution, because the self fitness value is small, the closer the candidate solution is to the theoretical optimal solution, the search is performed near the candidate solution, the convergence accuracy of the algorithm can be improved, and the closer the candidate solution is to the theoretical optimal solution. Therefore, the global exploration and local development capability of the traditional CS algorithm are balanced.
In some embodiments of the invention, the factor limiting the maximum amplitude
The calculation formula of (2) is as follows:
wherein n is the population number of cuckoos and w is the inertial weight. The bird nest is further optimized by executing the explosion process, and the optimization method not only can further enhance the diversity of the bird nest, but also can improve the searching speed and the optimization precision of the algorithm. To a certain extent, the evolution direction of bird nests in the algorithm is guided.
In step S500 of this embodiment, the following steps may be included, but not limited to:
s510, judging whether the G < Gmax condition is satisfied or not.
In this embodiment, the present step makes a condition determination as to whether or not the G < Gmax condition is satisfied.
S520, when the G < Gmax condition is satisfied, returning to execute the step S300; and if the condition that G < Gmax is not satisfied, outputting an optimization result of the objective function.
In this embodiment, when the condition G < Gmax is satisfied, the process returns to step S300 and continues the parameter optimization process; and if the condition that G < Gmax is not satisfied, outputting an optimized setting result of the objective function.
The method is used for improving and researching the defects that the late convergence speed of the cuckoo search algorithm is low, the solving precision is not high, the global searching capability and the local searching capability are unbalanced and the like. The improved mode is that an explosion operator with self-adaptive capacity is designed and is fused into Levis flight of a CS algorithm, and then an inertia weight decreasing strategy based on an S-shaped function is designed according to the rule of a population growth curve to distribute candidate solutions obtained by the Levis flight and the explosion operator, so that the global exploration and local development capacity of the algorithm is balanced; the purpose of this is that lewy flight is a flight mode combining long-time short-distance search and occasional long-distance search, and as the algorithm performs long-distance search, the algorithm has the capability of jumping out of local optimum, that is, the global exploration capability of the algorithm is ensured, but at the same time, as lewy flight has randomness, the candidate solution generated by the CS algorithm also has randomness, and the convergence accuracy of the algorithm is difficult to improve.
The designed explosion operator can well solve the problem, because the explosion range of the explosion operator is calculated based on the optimal solution in the population, the candidate solution of the algorithm can evolve towards the optimal solution, and because the search range of better individuals in the population is small, the convergence precision of the algorithm is favorably improved; the search range of the poorer individuals in the population is large, and the local optimal solution can be skipped out. Meanwhile, in order to enable the designed explosion operator to be better integrated into the CS algorithm, the Lavy flight and the explosion operator are in a competitive relationship under the enlightening of a population growth curve, the Lavy flight is dominant in the early stage of the algorithm, and the CS algorithm searches a global optimal solution under the action of the Lavy flight; along with the increase of the iteration times, the Levy flight is gradually positioned to the approximate position of the global optimal solution, the influence of an explosion operator is continuously enhanced at the moment, and a bird nest with a better position (lower fitness value) is searched nearby the bird nest so as to improve the convergence precision of the algorithm; the effect of the poorly positioned (poorly adapted) bird nest on the explosion operator still retains the ability to jump out of the locally optimal solution.
In addition, since the selection of the candidate solution in the preference random walk process of the cuckoo is random, the search efficiency of the algorithm is reduced, and for this reason, a candidate solution selection mechanism of preference random walk is improved. The specific way is to change the difference between two randomly generated candidate solutions into the difference between the current candidate solution and the worst solution of the population. The improvement has the advantage that each time the cuckoo carries out preference random walk, the worst solution of the current population can be used for guiding the population to evolve to the global optimal solution. The worst solution of the population in the early stage of the algorithm can increase the diversity of the population, and is beneficial to improving the overall development capability of the algorithm; in the later stage of the algorithm, the bird nest can be close to the optimal solution of the population under the action of the explosion operator, the difference value between the optimal solution and the worst solution of the population is not large, the search distance can be shortened based on the reverse learning mechanism of the worst solution of the population, the algorithm can perform refined search, and therefore the search precision of the algorithm is improved.
The crane sliding mode controller can realize basic control functions, but the controller has numerous parameters and is complex to set, a simulation experiment is mostly carried out by manually adjusting the parameters at present, and the process of debugging the parameters needs a lot of time.
The Optimization Algorithm of the invention is improved on the basis of Cuckoo Search Algorithm (CS), because the Cuckoo Search Algorithm has the problems of slow later convergence speed and low solving precision, the Search performance of the CS Algorithm needs to be improved in order to optimize better control parameters, so the invention provides an adaptive Cuckoo Search Algorithm (CSFOA) with an S function explosion operator, and the designed Algorithm can effectively improve the Search speed of the traditional CS Algorithm and balance the global exploration and local development capability of the Algorithm. The optimization algorithm provided by the invention is applied to parameter setting of the crane sliding mode controller, so that the controller can obtain better control performance, namely, the crane load positioning can be realized more quickly, the load swing angle is restrained more effectively, and the robustness is better.
According to the scheme, the SFCS algorithm is an improved cuckoo search algorithm, and an explosion operator with self-adaptive capacity and a reverse learning strategy based on the worst solution of the population are added in the traditional CS algorithm, so that the populations can communicate with each other, and the population can be further guided to evolve towards the optimal solution; in the convergence aspect, the influence of an explosion operator and Levier flight on the generation of a candidate solution of the algorithm is balanced by the population growth curve, so that the global exploration and local development capability of the algorithm is balanced; in addition, when the algorithm is in the later iteration stage, the explosion operator is dominant, and the convergence speed and precision of the algorithm are effectively improved; and the SFCS algorithm is applied to parameter setting of the sliding mode controller, so that the parameter optimization setting is well realized, the time for manually adjusting the parameters is saved, and a better control effect is obtained.
Example two
Referring to fig. 3, an embodiment of the present invention provides a sliding mode controller parameter setting apparatus 1000 based on an SFCS algorithm, including:
an initialization unit 1100, configured to initialize parameters and populations, and set an evolution algebra G, a maximum evolution algebra Gmax, a bird nest number n, a search space dimension D, a found probability Pa, a search space upper limit u, and a search space lower limit l;
an operation unit 1200, configured to randomly generate n bird nest locations and calculate a fitness value of each bird nest location;
the first processing unit is used for updating each bird nest position by using an inertial weight decreasing algorithm based on an S-shaped function, calculating the fitness of a new bird nest position, comparing the existing bird nest with the previous generation bird nest position, selecting the bird nest position with a better fitness value to replace the bird nest position with a poorer fitness value, and storing the bird nest with the worst fitness value;
a second processing unit 1300, configured to generate a random number Pt of [0,1], if Pt is less than Pa, perform preference random walk to update a nest position by using a reverse learning algorithm based on a worst population solution, calculate an adaptability value of a nest, compare an existing nest position with an original nest position, and select a nest position with a better adaptability value to replace a nest position with a poorer adaptability value;
the third processing unit 1400 is configured to return to perform step S300 if G < Gmax, otherwise output an optimization result of the objective function.
In some embodiments of the invention, the third processing unit 1400 comprises:
a determining unit 1410 configured to determine whether a G < Gmax condition is satisfied;
an executing unit 1420, configured to, when the G < Gmax condition is satisfied, return to executing step S300; and if the condition that G < Gmax is not satisfied, outputting an optimization result of the objective function.
It should be noted that, because the city illegal parking detection device based on the unmanned aerial vehicle in the embodiment is based on the same inventive concept as the city illegal parking detection method based on the unmanned aerial vehicle in the first embodiment, the corresponding content in the first method embodiment is also applicable to the embodiment of the system, and the detailed description is omitted here.
According to the scheme, the SFCS algorithm is an improved cuckoo search algorithm, and an explosion operator with self-adaptive capacity and a reverse learning strategy based on the worst solution of the population are added in the traditional CS algorithm, so that the populations can communicate with each other, and the population can be further guided to evolve towards the optimal solution; in the convergence aspect, the influence of an explosion operator and Levier flight on the generation of a candidate solution of the algorithm is balanced by the population growth curve, so that the global exploration and local development capability of the algorithm is balanced; in addition, when the algorithm is in the later iteration stage, the explosion operator is dominant, and the convergence speed and precision of the algorithm are effectively improved; and the SFCS algorithm is applied to parameter setting of the sliding mode controller, so that the parameter optimization setting is well realized, the time for manually adjusting the parameters is saved, and a better control effect is obtained.
EXAMPLE III
The third embodiment of the present invention further provides a computer-readable storage medium, where an executable instruction of a sliding mode controller parameter setting device based on an SFCS algorithm is stored in the computer-readable storage medium, and the executable instruction of the sliding mode controller parameter setting device based on the SFCS algorithm is used to enable the sliding mode controller parameter setting device based on the SFCS algorithm to execute the above sliding mode controller parameter setting method based on the SFCS algorithm, for example, execute the above-described method steps S100 to S500 in fig. 1, and implement the functions of the unit 1000 and 1420 in fig. 3.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an illustrative embodiment," "an example," "a specific example," or "some examples" or the like mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
While embodiments of the invention have been shown and described, it will be understood by those of ordinary skill in the art that: various changes, modifications, substitutions and alterations can be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.
Claims (10)
1. A sliding mode controller parameter setting method based on an SFCS algorithm is characterized by comprising the following steps:
step S100, initializing parameters and population, and firstly setting an evolution algebra G, a maximum evolution algebra Gmax, a bird nest number n, a search space dimension D, a found probability Pa, a search space upper limit u and a search space lower limit l;
step S200, randomly generating n bird nest positions and calculating the fitness value of each bird nest position;
step S300, updating each bird nest position by using an inertial weight decreasing algorithm based on an S-shaped function, calculating the fitness of a new bird nest position, comparing the existing bird nest with the previous generation bird nest position, selecting a bird nest position with a better fitness value to replace a bird nest position with a poorer fitness value, and storing the bird nest with the worst fitness value;
step S400, generating a random number Pt of [0,1], if Pt is less than Pa, utilizing a reverse learning algorithm based on a worst population solution to carry out preference random walk to update the position of a bird nest, calculating the fitness value of the bird nest, comparing the position of the existing bird nest with the position of the original bird nest, and selecting the position of the bird nest with a better fitness value to replace the position of the bird nest with a poorer fitness value;
and step S500, if G is less than Gmax, returning to execute the step S300, otherwise, outputting an optimization result of the target function.
2. The SFCS algorithm-based sliding-mode controller parameter setting method according to claim 1, wherein the S-shaped function-based inertial weight decreasing algorithm is represented as:
wherein Zi, j represents the j (j is 1,2, …, n) th candidate solution when the algorithm evolves to the ith generation; zi, g is the global optimal solution searched by the ith generation; η represents the initial search step of the algorithm, and is related to the dimension and complexity of the optimization problem, and is usually equal to 1;
representing point-to-point multiplication. L (lambda) represents the search path of the Laevir flight, w is the inertial weight, AjAnd carrying out the calculation value of the explosion operation on each bird nest for the explosion operator.
3. The SFCS algorithm-based sliding-mode controller parameter setting method according to claim 1, characterized in that the population worst solution-based inverse learning algorithm is expressed as:
wherein, r and Pt are random numbers subject to uniform distribution, Heaviside (×) is a jump function, Pa is a discovery probability, and Pa is 0.25, Zi, j and Zi, w respectively represent two random feasible solutions of the algorithm evolution to the ith generation.
4. The SFCS algorithm-based sliding mode controller parameter tuning method of claim 1, wherein the objective function is expressed as:
wherein F is the optimized driving force u (t), the position deviation signal ζ (t) and the swing angle signal x3(t) of the objective function, and w1, w2 and w3 are weight coefficients.
5. The SFCS algorithm-based sliding mode controller parameter setting method according to claim 2, wherein A isjThe calculation formula of (2) is as follows:
wherein the content of the first and second substances,
to limit the maximum amplitude, f (Zi, j) represents the fitness value of the jth candidate solution when the algorithm evolves to the ith generation, and is a very small constant to avoid the condition that the denominator is zero.
6. The SFCS algorithm-based sliding mode controller parameter setting method according to claim 5, wherein the method is characterized in thatFactors limiting the maximum amplitude
The calculation formula of (2) is as follows:
wherein n is the population number of cuckoos and w is the inertial weight.
7. The SFCS algorithm-based sliding-mode controller parameter setting method according to claim 1, wherein if G < Gmax, the step S300 is executed again, otherwise, the optimization result of the objective function is output, comprising:
judging whether the G < Gmax condition is satisfied;
when the condition of G < Gmax is satisfied, returning to the step S300; and if the condition that G < Gmax is not satisfied, outputting an optimization result of the objective function.
8. A sliding mode controller parameter setting device based on SFCS algorithm is characterized by comprising the following steps:
the device comprises an initialization unit, a search unit and a search unit, wherein the initialization unit is used for initializing parameters and populations, and firstly setting an evolution algebra G, a maximum evolution algebra Gmax, a bird nest number n, a search space dimension D, a found probability Pa, a search space upper limit u and a search space lower limit l;
an operation unit for randomly generating n bird nest positions and calculating a fitness value of each bird nest position;
the first processing unit is used for updating each bird nest position by using an inertial weight decreasing algorithm based on an S-shaped function, calculating the fitness of a new bird nest position, comparing the existing bird nest with the previous generation bird nest position, selecting the bird nest position with a better fitness value to replace the bird nest position with a poorer fitness value, and storing the bird nest with the worst fitness value;
the second processing unit is used for generating a random number Pt of [0,1], if Pt is less than Pa, utilizing a reverse learning algorithm based on a worst population solution to carry out preference random walk to update the position of the bird nest, calculating the adaptability value of the bird nest, comparing the position of the existing bird nest with the position of the original bird nest, and selecting the position of the bird nest with a better adaptability value to replace the position of the bird nest with a poorer adaptability value;
and the third processing unit is configured to return to execute step S300 if G < Gmax, and otherwise output an optimization result of the objective function.
9. The SFCS algorithm-based sliding mode controller parameter setting device according to claim 8, wherein the third processing unit comprises:
the judging unit is used for judging whether the G < Gmax condition is satisfied or not;
an execution unit, configured to return to execute step S300 when the G < Gmax condition is satisfied; and if the condition that G < Gmax is not satisfied, outputting an optimization result of the objective function.
10. A computer-readable storage medium characterized by: the computer-readable storage medium stores executable instructions of a sliding mode controller parameter setting device based on an SFCS algorithm, and the executable instructions of the sliding mode controller parameter setting device based on the SFCS algorithm are used for enabling the sliding mode controller parameter setting device based on the SFCS algorithm to execute the sliding mode controller parameter setting method based on the SFCS algorithm according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010906950.6A CN112068429B (en) | 2020-08-31 | 2020-08-31 | SFCS (sparse form-factor correction) algorithm-based sliding mode controller parameter setting method and device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010906950.6A CN112068429B (en) | 2020-08-31 | 2020-08-31 | SFCS (sparse form-factor correction) algorithm-based sliding mode controller parameter setting method and device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112068429A true CN112068429A (en) | 2020-12-11 |
| CN112068429B CN112068429B (en) | 2022-06-21 |
Family
ID=73665009
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010906950.6A Active CN112068429B (en) | 2020-08-31 | 2020-08-31 | SFCS (sparse form-factor correction) algorithm-based sliding mode controller parameter setting method and device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112068429B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116667325A (en) * | 2023-05-29 | 2023-08-29 | 江苏师范大学 | Micro-grid-connected operation optimization scheduling method based on improved cuckoo algorithm |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105152020A (en) * | 2015-09-30 | 2015-12-16 | 山东大学 | Bridge crane self-adaptation track controller with tracking error restraint and method |
| CN106127295A (en) * | 2016-06-21 | 2016-11-16 | 湘潭大学 | A kind of Optimal Design of Pressure Vessel method based on self adaptation cuckoo Yu fireworks hybrid algorithm |
| CN106249602A (en) * | 2016-09-30 | 2016-12-21 | 山东大学 | Overhead crane finite time contrail tracker and method for designing thereof |
| CN106315414A (en) * | 2016-03-30 | 2017-01-11 | 南开大学 | Bridge crane control method based on sliding mode surface |
| CN107024865A (en) * | 2017-05-02 | 2017-08-08 | 南开大学 | Drive lacking jib crane positions the pendulum nonlinear control method that disappears |
| CN107678277A (en) * | 2017-03-16 | 2018-02-09 | 南京工业大学 | A kind of sliding-mode control in double pendulum bridge crane nonlinear sliding mode face |
| CN108427280A (en) * | 2018-03-21 | 2018-08-21 | 南京邮电大学 | A kind of overhead crane anti-swing control method based on sliding mode control theory |
| CN108557664A (en) * | 2018-01-05 | 2018-09-21 | 山东大学 | Bridge type crane system enhances coupling nonlinear PD types sliding mode controller and method |
| CN109635915A (en) * | 2019-01-15 | 2019-04-16 | 华侨大学 | A kind of iterative learning control method based on balanced single evolution cuckoo algorithm |
| CN110589684A (en) * | 2019-09-30 | 2019-12-20 | 山东深川变频科技股份有限公司 | Anti-swing driving control method for crane |
-
2020
- 2020-08-31 CN CN202010906950.6A patent/CN112068429B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105152020A (en) * | 2015-09-30 | 2015-12-16 | 山东大学 | Bridge crane self-adaptation track controller with tracking error restraint and method |
| CN106315414A (en) * | 2016-03-30 | 2017-01-11 | 南开大学 | Bridge crane control method based on sliding mode surface |
| CN106127295A (en) * | 2016-06-21 | 2016-11-16 | 湘潭大学 | A kind of Optimal Design of Pressure Vessel method based on self adaptation cuckoo Yu fireworks hybrid algorithm |
| CN106249602A (en) * | 2016-09-30 | 2016-12-21 | 山东大学 | Overhead crane finite time contrail tracker and method for designing thereof |
| CN107678277A (en) * | 2017-03-16 | 2018-02-09 | 南京工业大学 | A kind of sliding-mode control in double pendulum bridge crane nonlinear sliding mode face |
| CN107024865A (en) * | 2017-05-02 | 2017-08-08 | 南开大学 | Drive lacking jib crane positions the pendulum nonlinear control method that disappears |
| CN108557664A (en) * | 2018-01-05 | 2018-09-21 | 山东大学 | Bridge type crane system enhances coupling nonlinear PD types sliding mode controller and method |
| CN108427280A (en) * | 2018-03-21 | 2018-08-21 | 南京邮电大学 | A kind of overhead crane anti-swing control method based on sliding mode control theory |
| CN109635915A (en) * | 2019-01-15 | 2019-04-16 | 华侨大学 | A kind of iterative learning control method based on balanced single evolution cuckoo algorithm |
| CN110589684A (en) * | 2019-09-30 | 2019-12-20 | 山东深川变频科技股份有限公司 | Anti-swing driving control method for crane |
Non-Patent Citations (3)
| Title |
|---|
| 向博: "桥式吊车的滑模变结构控制", 《控制工程》 * |
| 唐超等: "基于鸟群算法优化的桥式吊车线性自抗扰控制", 《高技术通讯》 * |
| 王天雷等: "基于ASCP-CS算法的桥式吊车滑模控制器设计", 《湖南大学学报(自然科学版)》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116667325A (en) * | 2023-05-29 | 2023-08-29 | 江苏师范大学 | Micro-grid-connected operation optimization scheduling method based on improved cuckoo algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112068429B (en) | 2022-06-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110321666B (en) | Multi-robot path planning method based on priori knowledge and DQN algorithm | |
| Hamza et al. | Cuckoo search algorithm based design of interval Type-2 Fuzzy PID Controller for Furuta pendulum system | |
| CN112068429B (en) | SFCS (sparse form-factor correction) algorithm-based sliding mode controller parameter setting method and device and storage medium | |
| CN110647038B (en) | Bridge crane sliding mode control parameter optimization method, device, equipment and storage medium | |
| US11887485B2 (en) | Control method and system for collaborative interception by multiple unmanned surface vessels | |
| CN110526124A (en) | A kind of anti-sway method, apparatus of bridge crane based on sliding-mode surface, equipment and storage medium | |
| Sun et al. | An adaptive DE algorithm based fuzzy logic anti-swing controller for overhead crane systems | |
| CN110658814A (en) | Self-adaptive ship motion modeling method applied to ship motion control | |
| Wan et al. | ME‐MADDPG: An efficient learning‐based motion planning method for multiple agents in complex environments | |
| Fan et al. | Model-based stochastic search for large scale optimization of multi-agent UAV swarms | |
| CN111240796B (en) | Load balancing scheduling method based on improved MMAS | |
| Mishra et al. | A Huber reward function-driven deep reinforcement learning solution for cart-pole balancing problem | |
| Liang et al. | Multi-UAV autonomous collision avoidance based on PPO-GIC algorithm with CNN–LSTM fusion network | |
| CN117270398A (en) | Robot gait planning algorithm based on neural network and reinforcement learning | |
| Ariza Ramirez et al. | Exploration of the applicability of probabilistic inference for learning control in underactuated autonomous underwater vehicles | |
| Shi et al. | The improved Q-Learning algorithm based on pheromone mechanism for swarm robot system | |
| Zhou et al. | Neural Q learning algorithm based UAV obstacle avoidance | |
| Pozzoli et al. | Tustin neural networks: a class of recurrent nets for adaptive MPC of mechanical systems | |
| Niu et al. | The adaptive control based on BP neural network identification for two-wheeled robot | |
| CN115357019A (en) | Unmanned ship multi-parameter particle swarm identification system and method | |
| Yin et al. | AUV 3D Path Planning Based on Improved Empire Competition Algorithm in Ocean Current Environment | |
| Sun et al. | Unmanned aerial vehicles control study using deep deterministic policy gradient | |
| Adolf et al. | Probabilistic roadmaps and ant colony optimization for UAV mission planning | |
| Hu et al. | Research on AUV global path planning based on multi-objective ant colony strategy | |
| Suh et al. | An automatic travel control of a container crane using neural network predictive PID control technique |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |