Parallel water pump load distribution method, system, device and medium based on mayfly algorithm

Abstract

The present invention provides a parallel pump load distribution method, system, device and medium based on the mayfly algorithm. The basic parameters of the mayfly algorithm are initialized based on the operating parameters of the parallel pumps. The mayfly population is divided into multiple subgroups, the fitness value of the population is obtained, and the initial pump energy consumption is obtained. The female mayflies and male mayflies in multiple mayfly subgroups adjust their positions according to their own and their neighbors' experiences respectively. The speed of the mayflies is updated according to the individual optimal position and the group optimal position of the mayfly population. The updated position is substituted into the pump energy consumption calculation formula to obtain a new fitness value. At the same time, during the process of updating the mayfly positions, the evolution state of each subgroup is detected, the stagnant subgroups are activated and the optimal population is excluded. The iteration end condition is satisfied and the optimal solution is output. The present application proposes a multi-subgroup cooperation strategy to detect the subgroup state, activate the stopped subgroups and exclude the optimal population, which improves the search ability of the algorithm, significantly improves the energy-saving effect, and significantly improves the convergence accuracy, convergence speed and algorithm stability.

Description

Technical Field

[0001] The present invention belongs to the technical field of energy conservation, and particularly relates to a parallel water pump load distribution method, system, device and medium based on the mayfly algorithm. Background Technique

[0002] With the improvement of people's living standards and the increasing demand for indoor comfort, central air conditioners have become essential air conditioning equipment for large public buildings. As the main energy-consuming equipment of central air conditioners, parallel water pumps account for about 25% - 30% of the electrical load of central air conditioners, and there is a large space for energy conservation and consumption reduction. Since different types of water pumps have different equipment parameters, the start-stop state and speed of the water pumps can be adjusted to meet different pipe network flow requirements. Therefore, how to reduce the operating energy consumption of water pumps and improve the operating efficiency of parallel water pumps has become an important research content in building energy conservation in recent years.

[0003] In the prior art, the mathematical programming solution method and the meta-heuristic algorithm have good effects in solving the optimization configuration problem of the speed ratio and the number of operating units of parallel variable-frequency water pumps. However, the mathematical programming solution method cannot fully consider the actual situation of the operation of the central air-conditioning system, and there are problems such as low solution accuracy; most meta-heuristic algorithms have problems such as too slow convergence speed and low search efficiency. Summary of the Invention

[0004] Aiming at the problems existing in the prior art, the present invention provides a parallel water pump load distribution method, system, device and medium based on the mayfly algorithm, which can significantly improve the energy-saving effect, and the convergence accuracy, convergence speed and algorithm stability are significantly improved.

[0005] The present invention is realized through the following technical solutions:

[0006] The parallel water pump load distribution method based on the mayfly algorithm is characterized by including the following steps:

[0007] S1: Initialize the basic parameters of the mayfly algorithm based on the operating parameters of the parallel water pumps, divide the mayfly population into multiple subgroups, obtain the population fitness value, and obtain the initial water pump energy consumption;

[0008] S2: Adjust the positions of the female mayflies and male mayflies in multiple mayfly subgroups according to their own and neighbors' experiences, update the speeds of the mayflies according to the individual optimal position and the group optimal position of the mayfly population, and substitute the updated positions into the water pump energy consumption calculation formula to obtain a new fitness value;

[0009] S3: If the new fitness value is better than the individual optimal, update the individual optimal. If the new fitness value is better than the global optimal, update the global optimal. During the process of updating the mayfly positions, detect the evolutionary state of each subgroup, activate the stagnant subgroups and exclude the optimal population;

[0010] S4: Determine whether the iteration end condition is met. If it is met, exit the loop and output the optimal solution to obtain the optimal optimization plan; otherwise, return to step S2.

[0011] Further, when initializing the basic parameters of the mayfly algorithm based on the operating parameters of the parallel pumps in step S1, it is necessary to represent the position of each mayfly individual by the sequence combination of the speed ratios of each pump in the pump unit, that is, ω i = x i,j , and each mayfly individual represents a feasible solution to the load distribution problem of the parallel pumps. At the same time, it is necessary to meet the predefined range of the pump speed ratio. The obtained basic parameters of the mayfly algorithm are:

[0012] x i,j ∈C{(0.6, 1), 0}.

[0013] Further, the initial pump energy consumption in step S1 is:

[0014]

[0015] where P is the output power of the entire pump unit, n is the number of parallel pumps, and H is the flow rate of the pump unit.

[0016] Further, the expression for the female mayfly to adjust its position according to its own and its neighbors' experience in step S2 is:

[0017]

[0018] where x i t is the current position of mayfly i in the search space at time step t, and v i t+1 is the velocity added at the current position;

[0019] The expression for the male mayfly to adjust its position according to its own and its neighbors' experience is:

[0020]

[0021] where is mayfly i at time t, and v i t+1 is the velocity added at the current position.

[0022] Further, the velocity of the male mayfly is updated according to the individual optimal position and the group optimal position of the mayfly population in step S2 as:

[0023]

[0024] where is the velocity of mayfly i in the j - dimension at time t, is the position at time t, a1 and a2 are the positive attraction coefficients of social effects, pbest is the historical best position of the mayfly, gbest is the best position of the mayfly, β is the visibility coefficient of the mayfly, r p is the distance between the current position and pbest, r g is the distance between the current position and gbest;

[0025] The velocity of female mayflies is updated according to the individual optimal position and the group optimal position of the mayfly population as follows:

[0026]

[0027] Among them, is the velocity, is the position of the mayfly, a2 is a positive coefficient, β is a fixed visibility coefficient, r mf is the distance of the female mayfly from the position of the male mayfly, fl is a random walk coefficient, and r is a random number in the range [-1, 1].

[0028] Furthermore, detecting the evolutionary state of each subgroup in step S3 includes the following steps:

[0029] If the best mayfly sbest of each subgroup j cannot gradually optimize the solution with the increase of the iteration number and stagnation occurs, then a stagnation counter θ is introduced. When setting the subgroup stagnation times threshold Θ, when the subgroup stagnation times exceed the threshold Θ, an activation sample will be constructed through the information interaction between the stagnant subgroup and the optimal subgroup. When sbest j is in a continuous update state, the subgroup does not need to be intervened, and the stagnation detection counter is closed;

[0030] Among them, the best mayfly sbest of each subgroup j (j = 1, 2,..., H, where H is the number of subgroups) is the state of the j-th subgroup, and the sbest of the optimal subgroup is the group optimal value, that is, sbest j = gbest;

[0031] The construction of the activation sample includes the following steps:

[0032] Use the crossover mutation operator to interact the group optimal solution (gbest) with the stagnant subgroup:

[0033]

[0034] Among them, r d is a random mutation factor in the range [0, 1], is the activation sample of subgroup j in dimension d;

[0035] The optimal subgroup fuses some of its beneficial information with the stopping subgroup to help the stopping subgroup jump out of the local optimal solution;

[0036] After the stagnant subgroup is activated, it uses the stagnation detector to detect the activation effect again. If the sbest of the new samples of this subgroup is better than the old sbest, the stagnation detector θ is set to zero; otherwise, this subgroup will continue to be activated in the next iteration.

[0037] Furthermore, the exclusive optimal population in step S3 includes the following steps:

[0038] As the number of algorithm iterations increases, other subgroups will gather together prematurely under the guidance of the optimal solution; when the mayflies of other subgroups approach the optimal solution, the mayfly will be repelled; the repulsion expression is:

[0039]

[0040] where r d is a random number in the range of [0, 1], and the upper and lower bounds x d max and x d min .

[0041] A parallel pump load distribution system based on the mayfly algorithm, comprising:

[0042] An initial pump energy consumption module, configured to initialize the basic parameters of the mayfly algorithm based on the operating parameters of the parallel pumps, divide the mayfly population into multiple subgroups, obtain the population fitness value, and obtain the initial pump energy consumption;

[0043] A new fitness value module, configured to adjust the positions of the female and male mayflies in multiple mayfly subgroups according to their own and neighbors' experiences respectively, update the velocity of the mayflies according to the individual optimal position and the group optimal position of the mayfly population, and substitute the updated position into the pump energy consumption calculation formula to obtain a new fitness value;

[0044] An update module, configured to update the individual optimal if the new fitness value is better than the individual optimal, and update the global optimal if the new fitness value is better than the global optimal, detect the evolutionary state of each subgroup during the mayfly position update process, activate the stagnant subgroup and exclude the optimal population;

[0045] A judgment module, configured to judge whether the iteration end condition is reached. If satisfied, exit the loop and output the optimal solution to obtain the optimal optimization scheme; otherwise, continue to update.

[0046] A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the parallel pump load distribution method based on the mayfly algorithm are implemented.

[0047] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the steps of a parallel pump load distribution method based on the mayfly algorithm.

[0048] Compared with the prior art, the present invention has the following beneficial technical effects:

[0049] The present invention provides a parallel pump load distribution method, system, device and medium based on the mayfly algorithm. Initialize the basic parameters of the mayfly algorithm based on the operating parameters of the parallel pumps, divide the mayfly population into multiple subgroups, obtain the population fitness value, and obtain the initial pump energy consumption; adjust the positions of the female mayflies and male mayflies in multiple mayfly subgroups respectively according to their own and neighbors' experiences, update the speed of the mayflies according to the individual optimal position and the group optimal position of the mayfly population, and substitute the updated position into the pump energy consumption calculation formula to obtain a new fitness value; if the new fitness value is better than the individual optimal, update the individual optimal, and if the new fitness value is better than the global optimal, update the global optimal. Detect the evolution state of each subgroup during the mayfly position update process, activate the stagnant subgroup and exclude the optimal population; determine whether the iteration end condition is met, if satisfied, exit the loop and output the optimal solution to obtain the optimal optimization plan; this application solves with the speed ratio of each pump as the optimization variable, proposes a multi-subgroup cooperation strategy, detects the subgroup state, activates the stopped subgroup and excludes the optimal population, improves the search ability of the algorithm, and thus can significantly improve the energy-saving effect, and the convergence accuracy, convergence speed and algorithm stability are significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a flowchart of the parallel pump load distribution method based on the mayfly algorithm of the present invention;

[0051] Figure 2 It is a schematic diagram of the update of the inertia weight of a group of female mayflies with a population size of 200 at the initial stage of iteration in the specific implementation of the present invention

[0052] Figure 3 It is a flowchart for constructing an activation sample of the present invention;

[0053] Figure 4 It is a schematic diagram of the multi-subgroup cooperation strategy structure of the present invention;

[0054] Figure 5 It is a graph of an exponential decay function of distance introduced when updating the mayfly position of the exclusion strategy of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0055] The following further describes the present invention in detail with specific embodiments, which is an explanation rather than a limitation of the present invention.

[0056] To enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0057] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0058] This application provides a parallel pump load distribution method based on the mayfly algorithm, as Figure 1 shown, including the following steps:

[0059] S1: Initialize the basic parameters of the mayfly algorithm based on the operating parameters of the parallel pumps, divide the mayfly population into multiple subgroups, obtain the population fitness value, and obtain the initial pump energy consumption;

[0060] S2: Adjust the positions of the female mayflies and male mayflies in multiple mayfly subgroups according to their own and their neighbors' experiences respectively, update the speed of the mayflies according to the individual optimal position and the group optimal position of the mayfly population, and substitute the updated position into the pump energy consumption calculation formula to obtain a new fitness value;

[0061] S3: If the new fitness value is better than the individual optimal, update the individual optimal. If the new fitness value is better than the global optimal, update the global optimal. During the process of updating the mayfly positions, detect the evolutionary state of each subgroup, activate the stagnant subgroups and exclude the optimal population;

[0062] S4: Determine whether the iteration end condition is reached. If it is satisfied, exit the loop and output the optimal solution to obtain the optimal optimization plan. Otherwise, return to step S2.

[0063] Preferably, when initializing the basic parameters of the mayfly algorithm based on the operating parameters of the parallel pumps in step S1, it is necessary to represent the position of each mayfly individual by the sequence combination of the speed ratios of each pump in the pump unit, that is, ω i = xi,j , each individual mayfly represents a feasible solution to the load distribution problem of the parallel pumps, while satisfying the predefined range of the pump speed ratio. The basic parameters of the mayfly algorithm obtained are as follows:

[0064] x i,j ∈C{(0.6,1),0}.

[0065] Preferably, the initial pump energy consumption in step S1 is:

[0066]

[0067] Where P is the output power of the entire pump unit, n is the number of parallel pumps, and H is the flow rate of the pump unit.

[0068] Specifically, when the parallel variable-frequency pumps adopt constant differential pressure control, when the differential pressure H at both ends of the pump is not equal to the differential pressure set value H set , it is necessary to adjust the pump speed to meet the required flow rate Q at the end S . The relationship between the flow rate and the head in the pipe network system when the parallel pumps are working is:

[0069]

[0070] In the formula, S is the resistance of the pipe network, Q S is the total required flow rate of the system, m 3 / h; Q is the actual flow rate of the pump.

[0071] When the pump operates at the rated speed, its head-flow curve and efficiency-flow can be expressed in the following quadratic curve form:

[0072]

[0073] Where a, b, c, j, k, l are the performance parameters of the pump, and Q0, H0, η0 are the flow rate, head, and efficiency of the pump under the rated condition, respectively.

[0074] When the pump speed changes, according to the similarity of the pump, its performance parameters change as follows:

[0075]

[0076] Where Q i , H i , η i , n i represent the actual flow rate, head, efficiency, and speed of each pump respectively, and n0 is the rated speed of the pump.

[0077] The speed ratio of the i-th pump is represented by ω i , and is expressed as:

[0078]

[0079] When the water pump operates at any rotational speed:

[0080]

[0081]

[0082] The power of the water pump unit can be determined by the head and speed ratio of each water pump:

[0083]

[0084] where P is the output power of the entire water pump unit, in kW; and n is the number of parallel water pumps.

[0085] The head-flow curve of the water pump at a certain speed ratio is a parabola opening downward. To make the quadratic equation have real solutions, that is, the water pump meets the flow-head requirements at the current speed, the following constraints are proposed, otherwise the water pump is shut down:

[0086]

[0087] For the problem of optimizing the configuration of parallel water pumps, it can be described as the combination of water pump speeds that minimizes the energy consumption of the water pump unit under the condition of meeting the required flow Q of the system S . This optimization problem can be sorted into the mathematical description of Equation (9) as follows:

[0088] Min(P)

[0089] s.t. Q min < Q i < Q max

[0090]

[0091] Δ=(bω i ) 2 -4a(cω i 2 -H i )>0

[0092] ω min < ω i ≤1.

[0093] Preferably, in step S2, the female mayfly adjusts its position expression according to its own and its neighbors' experience as:

[0094]

[0095] where x it is the current position of mayfly \(i\) in the search space at time step \(t\), \(v\) i t+1 is the velocity added to the current position;

[0096] Male mayflies adjust their position expressions based on their own and their neighbors' experiences as follows:

[0097]

[0098] where, is mayfly \(i\) at time \(t\), \(v\) i t+1 is the velocity added to the current position.

[0099] Preferably, in step S2, male mayflies update their velocity according to the individual optimal position and the group optimal position of the mayfly population as follows:

[0100]

[0101] where, is the velocity of mayfly \(i\) in the \(j\) - dimension at time \(t\), is the position at time \(t\), \(a_1\) and \(a_2\) are positive social attraction coefficients, pbest is the historical best position of the mayfly, gbest is the best mayfly position, \(\beta\) is the visibility coefficient of the mayfly, \(r\) p is the distance between the current position and pbest, \(r\) g is the distance between the current position and gbest;

[0102] Female mayflies update their velocity according to the individual optimal position and the group optimal position of the mayfly population as follows:

[0103]

[0104] where, is the velocity, is the position of the mayfly, \(a_2\) is a positive coefficient, \(\beta\) is a fixed visibility coefficient, \(r\) mf is the distance of the female mayfly from the position of the male mayfly, \(fl\) is a random walk coefficient, \(r\) is a random number in the range \([-1,1]\).

[0105] Furthermore, for the operation of the algorithm, it is important that the best mayflies in the group continue to perform their characteristic up - and - down dance. Therefore, the best mayflies must constantly change their velocity, and in this case, the calculation is as follows:

[0106]

[0107] where \(d\) is the dance coefficient, \(r\) is a random number between \([-1,1]\). This up - and - down movement introduces a random element into the algorithm.

[0108] Furthermore, the crossover operator represents the mating process of two mayflies: one parent is selected from the male population and one from the female population. The way of selecting parents is the same as the way male mayflies attract females. In particular, the selection can be random or based on their fitness function. In the latter case, the best females reproduce with the best males, and the second-best females reproduce with the second-best males. The result of crossover is two offspring, which are generated as follows:

[0109]

[0110] where male is the male parent, female is the female parent, and offspring is the offspring mayfly. L is a random factor in [-1, 1].

[0111] The MA algorithm introduces Gaussian mutation, and the offspring mayflies mutate in a randomly selected dimension. If the number of mayfly population is N and the mutation probability m is defined, the number of mutated individuals in the offspring is The mutation formula is:

[0112]

[0113] where, offspring n is the nth dimension of the mutated mayfly, σ is the standard deviation of the normal distribution, and N n (0, 1) is the standard normal distribution with a mean of 0 and a variance of 1.

[0114] Furthermore, in a certain iteration cycle of the inertia weight coefficient adjustment strategy, since the inertia weight is only related to the number of iterations, the inertia weights of mayfly individuals are equal in one iteration, which reduces the diversity of the mayfly population and causes the algorithm to not converge quickly. This application uses a dynamic inertia weight method to update the individual optimal position and the group optimal position of the mayfly population and update the speed of male mayflies;

[0115] The following modifications are made to the speed update of male mayflies

[18] :

[0116]

[0117] And the speed update formula of female mayflies is modified to:

[0118]

[0119] where, g max and g min represent the maximum and minimum values that the inertia weight can take respectively, f i represents the fitness value of the current mayfly individual, and f p represents the historical optimal fitness value of the current mayfly individual. Represents the average fitness value of the mayfly population.

[0120] The dynamic inertia weight method adopted in this application introduces the fitness value of mayfly individuals and the optimal fitness value, solving the problems that different mayflies have the same inertia weight and the inertia weight only decreases monotonically with the number of iterations in one iteration cycle. In addition, due to the strong correlation between the inertia weight and the fitness value, when the mayfly position is within the range of the optimal solution, the inertia weight is adjusted to a very small value according to its update formula, thereby strengthening the local development ability of mayfly individuals and achieving fast and accurate optimization. In addition, in the same iteration cycle, different mayfly individuals will have different inertia weights according to their different fitness values, thereby enhancing the diversity of mayfly speed updates and improving the optimization efficiency of the algorithm. If a mayfly individual approaches the optimal solution in the middle of the iteration, the improved inertia weight method avoids the problem that the larger inertia weight of the traditional method causes the mayfly to deviate from the range of the optimal solution limit by assigning a smaller inertia weight value, speeds up the search time of the mayfly, and enables the algorithm to converge quickly. As Figure 2 shown, it is a schematic diagram of the update of the inertia weight of a group of female mayflies with a population size of 200 at the initial stage of iteration. It can be seen from the figure that the inertia weight update strategy proposed in this paper has a strong correlation with the fitness value, and the update trend of the inertia weight coefficient has a good tracking relationship with its fitness value. Moreover, in the same iteration number, different mayflies have their own independent inertia weight coefficients, and the fitness values of different mayflies determine whether their tasks are global exploration or local development, which greatly improves the search efficiency of the algorithm.

[0121] Preferably, to further improve the convergence speed and convergence accuracy of the algorithm, avoid the algorithm falling into local optimal solutions, and enhance the global search ability of the algorithm, this paper introduces a multi-subgroup cooperation strategy, as Figure 4 shown, that is, the entire mayfly population is divided into 4 subgroups, and the best mayfly sbest of each subgroup is used to reflect the evolutionary state of the subgroup. If sbest can gradually approach the optimal solution as the number of iterations increases, the subgroup will evolve independently without communicating with other subgroups. A stop detector is introduced. If the subgroup shows an evolutionary stagnation phenomenon, activation samples will be generated through information interaction between multiple groups to prompt the stagnant subgroup to continue optimizing. In addition, to avoid the problem of premature convergence of the algorithm caused by premature aggregation of subgroups, a subgroup repulsion mechanism is introduced; as Figure 3 shown, detecting the evolutionary state of each subgroup in step S3 includes the following steps:

[0122] If the best mayfly sbest of each subgroup j cannot gradually approach the optimal solution as the number of iterations increases and shows a stagnation phenomenon, a stagnation counter θ is introduced. When setting the subgroup stagnation times threshold Θ, when the subgroup stagnation times exceed the threshold Θ, activation samples will be constructed through information interaction between the stagnant subgroup and the optimal subgroup. When sbest jIn a state of continuous update, the subgroup requires no intervention, and the stagnation detection counter is turned off.

[0123] Among them, the best mayfly sbest of each subgroup j (j = 1, 2,..., H, where H is the number of subgroups) is the state of the j-th subgroup, and the sbest of the optimal subgroup is the global optimal value, that is, sbest j = gbest.

[0124] Preferably, the construction of the activation sample includes the following steps:

[0125] Use the crossover mutation operator to perform information interaction between the global optimal solution (gbest) and the stagnant subgroup:

[0126]

[0127] Among them, r d is a random mutation factor within the range of [0, 1], is the activation sample of subgroup j in dimension d;

[0128] The optimal subgroup fuses some of its favorable information with the stagnant subgroup to help the stagnant subgroup jump out of the local optimal solution;

[0129] After the stagnant subgroup is activated, it uses the stagnation detector to detect the activation effect again. If the sbest of the new sample of this subgroup is better than the old sbest, the stagnation detector θ is set to zero; otherwise, this subgroup will continue to be activated in the next iteration.

[0130] Preferably, the exclusion of the optimal population in step S3 includes the following steps:

[0131] As the number of algorithm iterations increases, other subgroups will gather together prematurely under the guidance of the optimal solution; when the mayflies of other subgroups approach the optimal solution, the mayfly will be excluded; the exclusion expression is:

[0132]

[0133] Among them, r d is a random number within the range of [0, 1], and the upper and lower bounds x d max and x d min of the solution in dimension d are introduced. Specifically, due to the strong correlation between the repulsive force and the mayfly distance, an exponential decay function regarding the distance is introduced when updating the mayfly position in the exclusion strategy, as Figure 5 shown. It can be seen from the figure that the closer the mayflies of other subgroups are to the optimal solution, the stronger the repulsive force, and the mayflies far from the optimal solution are hardly affected by the repulsive force in terms of their positions.

[0134] The present invention provides a parallel pump load distribution system based on the mayfly algorithm, including:

[0135] An initial pump energy consumption module, which is used to initialize the basic parameters of the mayfly algorithm based on the operating parameters of the parallel pumps, divide the mayfly population into multiple subgroups, obtain the population fitness value, and obtain the initial pump energy consumption;

[0136] A new fitness value module, which is used to adjust the positions of female and male mayflies in multiple mayfly subgroups according to their own and neighbors' experiences respectively, update the speed of the mayflies according to the individual optimal position and the group optimal position of the mayfly population, and substitute the updated position into the pump energy consumption calculation formula to obtain a new fitness value;

[0137] An update module, which is used to update the individual optimum if the new fitness value is better than the individual optimum, and update the global optimum if the new fitness value is better than the global optimum. During the process of updating the mayfly positions, detect the evolutionary state of each subgroup, activate the stagnant subgroups and exclude the optimal population;

[0138] A judgment module, which is used to judge whether the iteration end condition is reached. If it is satisfied, exit the loop and output the optimal solution to obtain the optimal optimization scheme, otherwise continue to update.

[0139] A preferred embodiment provided by this application is:

[0140] Select the load distribution problems of 2 typical air-conditioning chilled water circulation pump systems as research objects to verify the performance of the IMA algorithm. The system of Case 1 is composed of four variable-frequency pumps with a rated flow of 200 m 3 / h in parallel, the total designed flow of the system is 783 m 3 / h, and the designed head is 25 m; the system of Case 2 is composed of 3 variable-frequency pumps with a rated flow of 2080 m 3 / h and 1 variable-frequency pump with 1050 m 3 / h in parallel, the total designed flow of the system is 5783 m 3 / h, and the designed head is 50 m. The two cases are respectively composed of four identical pumps in parallel and three large and one small pumps in parallel, and the rated flow of the two groups of pumps has a large difference. The purpose is to test the ability of the IMA algorithm of this application to solve the parallel pump load distribution problem in different parallel pump systems and different working conditions. The circulating pump units all adopt constant pressure control. The operating parameters of the same model pumps in the system are the same, but due to the differences in the actual rotation speed and flow of the pumps during long-term operation, the actual operating curves of each pump are different. Therefore, the performance parameters of this pump group obtained from actual tests are shown in Table 1.

[0141] Table 1 List of characteristic parameters of the pump

[0142]

[0143] To verify the feasibility of the IMA algorithm of this application in solving the load distribution problem of parallel pumps, the improved and unimproved mayfly algorithms are first used for experiments under different working conditions, and the experimental results are compared with the results of the optimization algorithms of GA, PSO, and DE. In the experiment of Case 1, first, the MA algorithm is compared with the GA and PSO algorithms to verify the feasibility of the MA algorithm in solving the load distribution problem of parallel pumps. To verify the effectiveness of the IMA algorithm of this application proposed in this paper in solving the load distribution problem of parallel pumps, the optimization results of the IMA and MA algorithms of this application are also compared. The results are shown in Table 2.

[0144] Table 2 Comparison of the results of GA, PSO, MA, and IMA algorithms in Case 1

[0145]

[0146] As can be seen from Table 2, when the MA algorithm is not improved, its optimization ability is comparable to that of GA and PSO. However, when the system is at 80% and 50% flow demands, the search results of the GA and MA algorithms are not the optimal solutions in the search space, and the algorithms fall into local optimal values under this working condition. After the MA algorithm is improved, the IMA algorithm of this application has been significantly improved in terms of optimization ability. Compared with the GA algorithm, the energy-saving effect of the IMA algorithm of this application can reach 0.72% - 8.68%; compared with the PSO algorithm, the energy-saving effect of the IMA algorithm of this application can reach 0.72% - 1.88%; compared with the unimproved MA algorithm, the energy-saving effect of the IMA algorithm of this application can reach 0.02% - 4.51%. This proves that the IMA algorithm of this application can achieve relatively good energy-saving effects under different flow demands.

[0147] To further verify the effectiveness of the improved mayfly algorithm of this paper in solving the load distribution problem of parallel pumps, the optimization results of the IMA algorithm of this application are compared with those of GA, DE, and MA algorithms. The experimental results are shown in Table 3.

[0148] Table 3 Comparison of the results of GA, DE, MA, and IMA algorithms in Case 2

[0149]

[0150] In the optimization results of Table 3, the optimization results of the IMA of the present application are compared with the optimization results of algorithms such as GA, DE, and MA. Under the same traffic demand, the optimization results of the IMA algorithm of the present application are 1.86% to 7.66% more energy-efficient than GA and 0.01% to 7.15% more energy-efficient than the DE algorithm. Especially when the traffic demand is 90%, both the GA and DE algorithms fall into local optima, while the operating strategies obtained by the IMA algorithm of the present application are 7.66% and 7.15% more energy-efficient than the two respectively. Compared with the MA algorithm, when the traffic demand is less than 80%, the energy-saving effect of the algorithm is equivalent, and it is 1% energy-saving when the traffic demand is 90%. The overall energy-saving effect is relatively ideal. Therefore, when solving complex optimization problems with different pump combinations and large traffic demands, the IMA algorithm of the present application shows a good optimization effect and has obvious advantages compared with other optimization algorithms, indicating the feasibility of the IMA algorithm of the present application in solving the load distribution problem of parallel pumps.

[0151] In another embodiment of the present invention, a computer device is provided. The computer device includes a processor and a memory. The memory is used to store a computer program. The computer program includes program instructions. The processor is used to execute the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, and is suitable for implementing one or more instructions. Specifically, it is suitable for loading and executing one or more instructions in the computer storage medium to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of the parallel pump load distribution method based on the mayfly algorithm.

[0152] In another embodiment of the present invention, the present invention further provides a storage medium, specifically a computer-readable storage medium (Memory). The computer-readable storage medium is a memory device in a computer device and is used to store programs and data. It can be understood that the computer-readable storage medium here can include both the built-in storage medium in the computer device and, of course, the extended storage medium supported by the computer device. The computer-readable storage medium provides a storage space, and the operating system of the terminal is stored in this storage space. Moreover, one or more instructions suitable for being loaded and executed by the processor are stored in this storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. One or more instructions stored in the computer-readable storage medium can be loaded and executed by the processor to implement the corresponding steps of the parallel water pump load distribution method based on the mayfly algorithm in the above embodiment.

[0153] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0154] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the flows and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0155] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the functions in Figure 1 one flow or multiple flows and / or blocksFigure 1 The functions specified in one or more boxes.

[0156] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide for implementing the steps of the functions specified in one or more processes and / or boxes Figure 1 One or more processes and / or boxes Figure 1 The steps of the functions specified in one or more boxes.

[0157] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A parallel pump load distribution method based on the mayfly algorithm, characterized in that, It includes the following steps: S1: Initialize the basic parameters of the mayfly algorithm based on the operating parameters of the parallel pumps, divide the mayfly population into multiple subgroups, obtain the population fitness value, and get the initial pump energy consumption; The initial pump energy consumption is: where P is the output power of the entire pump unit, n is the number of parallel pumps, and H is the flow rate of the pump unit; S2: Adjust the positions of the female mayflies and male mayflies in multiple mayfly subgroups respectively according to their own and their neighbors' experiences, update the velocities of the mayflies according to the individual optimal position and the group optimal position of the mayfly population, and substitute the updated positions into the pump energy consumption calculation formula to obtain a new fitness value; S3: If the new fitness value is better than the individual optimal value, update the individual optimal value. If the new fitness value is better than the global optimal value, update the global optimal value. During the process of updating the mayfly positions, detect the evolutionary states of each subgroup, activate the stagnant subgroups and repel the optimal population; S4: Judge whether the iteration end condition is reached. If it is satisfied, exit the loop and output the optimal solution to obtain the optimal optimization scheme. Otherwise, return to step S2.

2. The parallel pump load distribution method based on the mayfly algorithm according to claim 1, wherein When initializing the basic parameters of the mayfly algorithm based on the operating parameters of the parallel pumps in step S1, it is necessary to represent the position of each mayfly individual by the sequence combination of the speed ratios of each pump in the pump unit, that is, ω i = x i,j , each mayfly individual represents a feasible solution to the load distribution problem of the parallel pumps. At the same time, it is necessary to meet the predefined range of the pump speed ratio. The obtained basic parameters of the mayfly algorithm are as follows: x i,j ∈C{(0.6, 1), 0}。 3. The parallel pump load distribution method based on the mayfly algorithm according to claim 1, characterized in that The expression for the female mayfly to adjust its position according to its own and its neighbors' experiences in step S2 is: where x i t is the current position of mayfly i in the search space at time step t, and v i t+1 is the velocity added to the current position; The expression for the male mayfly to adjust its position according to its own and its neighbors' experiences is: Among them, is the mayfly i at time t, v i t+1 is to add speed at the current position.

4. The method for parallel pump load distribution based on the mayfly algorithm according to claim 1, characterized in that The expression for the male mayfly to update its velocity according to the individual optimal position and the group optimal position of the mayfly population in step S2 is: Among them, is the velocity of mayfly i at the j - dimension and the t - moment, is the position at the t - moment, a1 and a2 are the positive attraction coefficients of social interaction, pbest is the historical best position of the mayfly, gbest is the best position of the mayfly, β is the visibility coefficient of the mayfly, r p is the distance between the current position and pbest, r g is the distance between the current position and gbest; The expression for the female mayfly to update its velocity according to the individual optimal position and the group optimal position of the mayfly population is: Among them, is the speed, is the position of the mayfly, a2 is a positive coefficient, β is a fixed visibility coefficient, r mf is the position of the female mayfly relative to the male mayfly, fl is a random walk coefficient, and r is a random number in the range [-1, 1].

5. The parallel pump load distribution method based on the mayfly algorithm according to claim 1, characterized in that The detection of the evolutionary states of each subgroup in step S3 includes the following steps: If the best mayfly sbest of each subgroup j cannot gradually optimize the solution as the number of iterations increases and the optimal solution stagnates, then a stagnation counter θ is introduced. When setting the threshold Θ for the stagnation times of the subgroup, when the stagnation times of the subgroup exceed the threshold Θ, an activation sample will be constructed through the information interaction between the stagnant subgroup and the optimal subgroup. When sbest j is in a continuous update state, the subgroup does not need to be intervened, and the stagnation detection counter is turned off; Among them, the best mayfly sbest of each subgroup j (j = 1, 2,..., H, where H is the number of subgroups) is the state of the j-th subgroup, and the sbest of the optimal subgroup is the global optimal value of the population, that is, sbest j = gbest; The construction of the activation sample includes the following steps: Use the crossover mutation operator to interact the information between the group optimal solution (gbest) and the stagnant subgroup; where r d is a random mutation factor within the range of [0, 1], is the activation sample of subgroup j in the d dimension; The optimal subgroup fuses some of its beneficial information with the stagnant subgroup to help the stagnant subgroup jump out of the local optimal solution; After the stagnant subgroup is activated, use the stagnation detector to detect the activation effect again. If the sbest of the new sample of this subgroup is better than the old sbest, set the stagnation detector θ to zero. Otherwise, this subgroup will continue to be activated in the next iteration.

6. The parallel pump load distribution method based on the mayfly algorithm according to claim 1, characterized in that The repelling of the optimal population in step S3 includes the following steps: As the number of algorithm iterations increases, other subgroups will gather together prematurely under the guidance of the optimal solution; when the mayflies in other subgroups approach the optimal solution, this mayfly will be repelled; the repelling expression is: where r d is a random number in the range of [0, 1], and the upper and lower bounds x d max and x d min .

7. The parallel pump load distribution system based on the mayfly algorithm is characterized in that The parallel pump load distribution method based on the mayfly algorithm according to any one of claims 1-6 includes: An initial pump energy consumption module, which is used to initialize the basic parameters of the mayfly algorithm based on the operating parameters of the parallel pumps, divide the mayfly population into multiple subgroups, obtain the population fitness value, and get the initial pump energy consumption; A new fitness value module, which is used to adjust the positions of the female mayflies and male mayflies in multiple mayfly subgroups respectively according to their own and their neighbors' experiences, update the velocities of the mayflies according to the individual optimal position and the group optimal position of the mayfly population, and substitute the updated positions into the pump energy consumption calculation formula to obtain a new fitness value; An update module, which is used to update the individual optimal value if the new fitness value is better than the individual optimal value, and update the global optimal value if the new fitness value is better than the global optimal value. During the process of updating the mayfly positions, detect the evolutionary states of each subgroup, activate the stagnant subgroups and repel the optimal population; A judgment module, configured to judge whether an iteration end condition is met. If so, exit the loop and output an optimal solution to obtain an optimal optimization scheme; otherwise, continue to update.

8. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, the steps of the parallel pump load distribution method based on the mayfly algorithm according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, the steps of the parallel pump load distribution method based on the mayfly algorithm according to any one of claims 1 to 6 are implemented.

Images