Pump-valve combined pump group scheduling method based on improved artificial bee colony algorithm
By improving the artificial bee colony algorithm and pump and valve scheduling model, the problems of low automation level and waste of pressure in the water supply pump group are solved, and the speed and reliability of joint pump and valve regulation are achieved, and energy consumption is reduced.
Patent Information
- Application Number
- CN202410956289.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-17
- Publication Date
- 2024-11-15
AI Technical Summary
The existing water supply pump set operation methods have problems such as low automation level, waste of pressure and low scheduling efficiency, especially in adjusting the number of pump sets and valve opening control.
The pump and valve joint scheduling method based on improved artificial bee colony algorithm is adopted. By adjusting the speed regulation ratio and valve opening of the inverter, the test sample data of flow, head and shaft power are obtained, and the pump and valve scheduling model with the lowest energy consumption is established. Implant it into the upper computer of the centrifugal pump system to achieve real-time regulation.
The speed and reliability of pump and valve regulation are improved, and the centrifugal pump system can be directly controlled, achieving better energy consumption performance and more efficient joint pump and valve regulation.
Smart Images

Figure CN118965606A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of pump group scheduling, and in particular to a pump-valve combined pump group scheduling method based on an improved artificial bee colony algorithm. Background Art
[0002] Pump groups are an important part of the normal operation of cities. With the continuous increase in urban scale and building water demand, it is increasingly important to achieve better energy consumption performance in existing water supply pump groups. At present, it is generally proven that the energy consumption of the same type of pump groups is optimal when running at the same speed, but few studies have studied when to adjust the number of pump groups. Most of the control methods are to put a second pump into operation when one pump group reaches the maximum operating frequency. The continuous development of intelligent algorithms and mechanical automation provides room for improvement in this control method.
[0003] The existing water supply pump group operation method has the following disadvantages: 1. The traditional parallel pump group operation method adopts manual or single frequency conversion control method, and the automation level is low; 2. The commonly used variable frequency constant pressure control method adopts a fixed valve opening control strategy, resulting in unnecessary pressure waste; 3. Due to the complexity of the control method, the scheduling method is difficult to implant into the interconnected components of the hardware equipment for collaborative work, and the execution efficiency is low. Summary of the invention
[0004] In order to overcome the shortcomings in the operation of the current parallel pump group system, the present application provides a pump-valve combined pump group scheduling method based on an improved artificial bee colony algorithm with the characteristics of fast control speed and high reliability.
[0005] The present application provides a pump-valve combined pump group scheduling method based on an improved artificial bee colony algorithm, which adopts the following technical solutions: A pump-valve combined pump group scheduling method based on an improved artificial bee colony algorithm comprises the following steps: Step S100, adjusting the speed ratio k of the frequency converter and the valve opening L to obtain test sample data of flow rate Q, head H and shaft power P under different valve openings L; Step S200, drawing and fitting the valve opening-shaft power diagram under different speed ratios k, and obtaining the P(L, k) relationship through nonlinear regression; Step S300, establishing a pump valve scheduling model with the goal of minimizing energy consumption; Step S400, improving the artificial bee colony algorithm according to the centrifugal pump system constraints; Step S500: implanting a pump-valve scheduling model based on the improved artificial bee colony algorithm into a host computer of a centrifugal pump system, and realizing pump-valve joint control of the centrifugal pump system based on real-time measurement data.
[0006] Preferably, the speed ratio k of the frequency converter is adjusted by adjusting the output frequency of the frequency converter, and the speed ratio k=output frequency / 50.
[0007] Preferably, in step S100, after obtaining test sample data of flow rate Q, head H and shaft power P under different valve openings L, the collected test sample data are preprocessed for data rationality screening and elimination, and the experiment is repeated multiple times to obtain the final test sample data.
[0008] Preferably, in step S200, a valve opening-shaft power diagram under different speed ratios k is drawn and fitted, and a P(L, k) relationship is obtained by nonlinear regression, and the following steps are also included: Step S210, using valve opening L as the horizontal coordinate and shaft power P as the vertical coordinate, a valve opening-shaft power diagram under different speed ratios k is drawn; Step S220, based on the flow-shaft power characteristic equation and the valve opening-flow characteristic equation, the valve opening-shaft power relationship under different speed ratios k is obtained by nonlinear regression. P(L, k) = c0·(Q max ·R L-1 ) 2 +c1·(Q max ·R L-1 )·k+c2·k 2 ; Among them, the flow-shaft power characteristic equation is as follows: P=c0·Q 2 +c1·Q·k+c2·k 2 ; In the flow-shaft power characteristic equation, P is the shaft power, Q is the flow, k is the speed ratio, and c0, c1 and c2 are fitting coefficients; Among them, the valve opening-flow characteristic equation is as follows: Q=Q max ·R L-1 ; Q in the valve opening-flow characteristic equation max is the maximum flow rate that the valve can adjust, R is the valve adjustable ratio, and L is the valve opening.
[0009] Preferably, in step S300, The demand flow Q r and required head H r Given the conditions of the centrifugal pump system, the pump-valve scheduling model is established using the valve opening L and the speed ratio k as the determining variables; The formula of the pump-valve scheduling model is as follows: Where n is the pump serial number; m is the total number of pumps; Q r and H r is the required flow and required head; Q n is the flow rate of pump n; k n is the speed ratio of pump n.
[0010] Preferably, step S400 includes the following steps: Step S410, according to the centrifugal pump system constraints, the artificial bee colony algorithm is improved to optimize the initial population distribution and individual generation method, and the Latin hypercube population initialization method is used to optimize the initial population distribution. The formula of the Latin hypercube population initialization method is as follows: x ij =UB i +rand(0,1)·Δx i +(i-1)·Δx i ; Where: x ij is the parameter value of the i-th individual in the j-th dimension, also known as the honey source, i is the serial number of the individual in the initial population of the algorithm, and j is the serial number of the optimization parameter dimension; UB i and LB i are the upper and lower limits of each parameter; t is the number of subintervals divided by the Latin hypercube method; Δx i is the length of the parameter i subinterval; Step S420, adopting a method of equally selecting an original search method and an elite guided search method for the individual generation method; The equation of the original search method is: The equation for the elite guided search method is: in: is the new parameter value generated by the i-th individual in the j-th dimension, also known as the new nectar source; a is the acceleration coefficient, usually 1; is a random number uniformly distributed between [-1, 1]; w is a parameter different from i; x Bj is the optimal nectar source in the jth dimension.
[0011] Preferably, step S500 includes the following steps: Step S510, implanting the pump valve scheduling model based on the improved artificial bee colony algorithm into the host computer of the centrifugal pump system; Step S520: collect the demand flow Q at the end of the pipe network in real time r and required head H r , as the input of the pump-valve scheduling model; Step S530, the valve opening L and the speed ratio k are used as two parameters of the individuals in the improved artificial bee colony algorithm, and the power P tot As the optimization goal of improving the artificial bee colony algorithm; Step S540: When the improved artificial bee colony algorithm converges or reaches a predetermined number of iterations, the individual with the lowest fitness value is selected as the optimal solution. The optimal solution is the power P in the pump valve scheduling model. tot Minimum valve opening L and speed ratio k; Step S550: Apply the valve opening L and the speed ratio k calculated by the improved artificial bee colony algorithm to the centrifugal pump system for regulation.
[0012] In summary, the present application includes at least one of the following beneficial technical effects: 1. This application uses valve opening as the independent variable, overcoming the indirectness of the traditional method of using flow as the independent variable. The pump-valve joint control increases the direct controllability of the centrifugal pump system; 2. This application will implant the improved artificial bee colony algorithm and pump-valve scheduling model into the host computer of the centrifugal pump system for calculation and control, so as to realize real-time detection of the operating status of the centrifugal pump system and precise joint control of the pump and valve, with fast control speed and high reliability. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] Figure 1 It is a flow chart of the pump-valve combined pump group scheduling method based on the improved artificial bee colony algorithm.
[0014] Figure 2 It is a schematic diagram of the valve opening-shaft power diagram.
[0015] Figure 3 is a schematic diagram of a centrifugal pump system. DETAILED DESCRIPTION
[0016] The following is combined with Figure 1-3 This application is described in further detail.
[0017] A pump-valve combined pump group scheduling method based on improved artificial bee colony algorithm, referring to Figure 1 As shown, the following steps are included: Step S100, adjusting the speed ratio k of the frequency converter and the valve opening L to obtain test sample data of flow rate Q, head H and shaft power P under different valve openings L.
[0018] According to the technical solution defined in step S100, specifically, the frequency converter is set in the centrifugal pump system, the frequency converter is used to control the motor of the centrifugal pump and the motor that controls the opening of the centrifugal pump valve, and the centrifugal pump is regulated by adjusting the frequency converter. In this application, there is a relationship between the speed ratio k and the output frequency of the frequency converter, and the speed ratio k can be obtained by converting the output frequency. The speed ratio k of the frequency converter can be adjusted by adjusting the output frequency of the frequency converter, and the speed ratio k = output frequency / 50.
[0019] Among them, after obtaining the test sample data of flow rate Q, head H and shaft power P under different valve openings L, the collected test sample data will be pre-processed for data rationality screening and elimination, and the experiment will be repeated many times to obtain the final test sample data.
[0020] The final test sample data is shown in Table 1 below: Table 1 The test sample data of the shaft power P under different speed ratios k and different valve openings L can be obtained from the above Table 1.
[0021] Step S200, drawing and fitting the valve opening-shaft power diagram under different speed ratios k, and obtaining the P(L, k) relationship through nonlinear regression.
[0022] According to the technical solution defined in step S200, specifically, it is known that when a centrifugal pump runs at rated speed, there is a flow-shaft power relationship, and the flow-shaft power characteristic equation is as follows: P0=a0·Q 2 +a1·Q+a2; Among them, P0 is the shaft power when running at rated speed, Q is the flow rate, and a0, a1 and a2 are fitting coefficients.
[0023] It can also be known that there is an equal percentage characteristic between the flow rate and the valve opening of the centrifugal pump. It can be obtained that there is a relationship between the flow rate Q, the adjustable ratio R and the valve opening L, and the valve opening-flow characteristic equation can be obtained. The above adjustable ratio R is the valve adjustable rate. The valve opening-flow characteristic equation is as follows: Q=Q max ·R L-1 ; Among them, Q max is the maximum flow rate that the valve can adjust, R is the valve adjustable ratio, and L is the valve opening.
[0024] Then, the test data of different speed ratios k are integrated into the same curve chart, and a valve opening-shaft power curve chart is drawn with valve opening L as the horizontal coordinate and shaft power P as the vertical coordinate. Among them, the different speed ratios k are speed ratio k1=1, speed ratio k2=0.9, speed ratio k3=0.8, speed ratio k4=0.7, and speed ratio k5=0.6. The valve opening-shaft power curve chart after fitting is as follows: Figure 2 shown.
[0025] According to the flow-shaft power characteristic equation and the valve opening-flow characteristic equation, the points under different speed ratios k are used with nonlinear regression method to obtain the P(L, k) relationship. The fitted curve expression is as follows: k1: P=-0.22441*30^(2L-2)+0.5771*30^(L-1)+0.01921; k2: P=-0.22607*30^(2L-2)+0.52071*30^(L-1)+0.01551; k3: P=-0.22743*30^(2L-2)+0.4643*30^(L-1)+0.01204; k4: P=-0.22448*30^(2L-2)+0.40389*30^(L-1)+0.00954; k5: P=-0.22725*30^(2L-2)+0.34853*30^(L-1)+0.00682; Among them, P is the shaft power and L is the valve opening.
[0026] Therefore, based on the fact that there are flow-shaft power characteristic equations and valve opening-flow characteristic equations when the centrifugal pump runs at rated speed, and according to the pump similarity law, it can be obtained that the centrifugal pump has the following flow-shaft power characteristic equation when the speed is variable (i.e., at different speed ratios k): P=c0·Q 2 +c1·Q·k+c2·k 2 ; Among them, P is the shaft power, Q is the flow rate, k is the speed ratio, and c0, c1 and c2 are fitting coefficients.
[0027] And based on the above valve opening-flow characteristic equation: Q=Q max ·R L-1 ; The valve opening-shaft power equation can be further obtained: P(L,k)=c0·(Q max ·R L-1 ) 2 +c1·(Q max·R L-1 )·k+c2·k 2 ; Among them, Q max is the maximum flow rate that the valve can adjust, R is the valve adjustable ratio, and L is the valve opening. It is worth noting that Q used in this application max =7m 3 / h, the valve adjustable ratio R can be used as a constant value, the valve adjustable ratio R = 30.
[0028] Therefore, based on the curve expression of P(L, k) fitting under the above speed ratio k1, the curve expression of P(L, k) fitting under the speed ratio k2, the curve expression of P(L, k) fitting under the speed ratio k3, the curve expression of P(L, k) fitting under the speed ratio k4, and the curve expression of P(L, k) fitting under the speed ratio k5, and further fitting based on the relationship of the speed ratio k, a specific curve expression of valve opening-shaft power can be obtained: P(L,k)=-0.0025·30 2(L-1) +4.0474·30 L-1 ·k+0.01916·k 2 .
[0029] Step S300, establishing a pump valve scheduling model with the goal of minimizing energy consumption.
[0030] According to the technical solution defined in step S300, specifically, the required flow rate Q r and required head H r Given the conditions of the centrifugal pump system, the pump-valve scheduling model is established by using the valve opening L and the speed ratio k as the determining variables; The formula of the pump-valve scheduling model is as follows: Where n is the pump serial number; m is the total number of pumps; Q r and H r is the required flow and required head; Q n is the flow rate of pump n; k n is the speed ratio of pump n.
[0031] Step S400, improving the artificial bee colony algorithm according to the centrifugal pump system constraints.
[0032] According to the technical solution defined in step S400, specifically, the following steps are also included: Step S410, according to the centrifugal pump system constraints, the artificial bee colony algorithm is improved to optimize the initial population distribution and individual generation method. Since the random population generation method may bring many infeasible solutions, in order to improve the quality of the initial population, the present invention adopts the Latin hypercube population initialization method to optimize the initial population distribution. The formula of the Latin hypercube population initialization method is as follows: x ij =UB i +rand(0,1)·Δx i +(i-1)·Δx i ; Where: x ij is the parameter value of the i-th individual in the j-th dimension, also known as the nectar source, x i is a two-dimensional parameter vector; i is the serial number of the individual in the initial population of the algorithm, and the individual contains two variables, valve opening L and adjustable ratio k; j is the serial number of the optimization parameter dimension. Since it only contains two variables, valve opening L and adjustable ratio k, in this embodiment, j={1, 2}, where j represents the valve opening L and the adjustable ratio k in the two dimensions, respectively. When j=1 represents the valve opening L, j=2 represents the adjustable ratio k.
[0033] It is worth noting that the artificial bee colony algorithm will generate N individuals during initialization, and the serial number of each individual is i. In one embodiment, the artificial bee colony algorithm will generate 30 individuals during initialization. Relative to the solutions for 30 valve openings L and adjustable ratios k, the serial number of each solution is i.
[0034] UB i and LB i are the upper and lower limits of each parameter; t is the number of subintervals divided by the Latin hypercube method; Δx i is the length of the parameter i subinterval.
[0035] Step S420, in order to speed up the convergence speed, the present invention adopts a method of equally selecting the original search method and the elite guided search method for the individual generation method; The equation of the original search method is: The equation for the elite guided search method is: in: is the new parameter value generated by the i-th individual in the j-th dimension, also known as the new nectar source; a is the acceleration coefficient, usually 1; is a random number uniformly distributed between [-1, 1], which determines the degree of disturbance of the old nectar source on the generation of the new nectar source; w is a parameter different from i; xBj is the optimal nectar source in the jth dimension.
[0036] Step S500: implanting a pump-valve scheduling model based on the improved artificial bee colony algorithm into a host computer of a centrifugal pump system, and realizing pump-valve joint control of the centrifugal pump system based on real-time measurement data.
[0037] According to the technical solution defined in step S500, specifically, the following steps are also included: Step S510, implanting the pump valve scheduling model based on the improved artificial bee colony algorithm into the host computer of the centrifugal pump system; Step S520: collect the demand flow Q at the end of the pipe network in real time r and required head H r , as the input of the pump-valve scheduling model; Step S530, the valve opening L and the speed ratio k are used as two parameters of the individuals in the improved artificial bee colony algorithm, and the power P tot As the optimization goal of improving the artificial bee colony algorithm; Step S540: When the improved artificial bee colony algorithm converges or reaches a predetermined number of iterations, the individual with the lowest fitness value is selected as the optimal solution. The optimal solution is the power P in the pump valve scheduling model. tot Minimum valve opening L and speed ratio k; Step S550: Apply the valve opening L and the speed ratio k calculated by the improved artificial bee colony algorithm to the centrifugal pump system for regulation.
[0038] Reference Figure 3 As shown, the relationship between the improved artificial bee colony algorithm and the pump-valve scheduling model is that the improved artificial bee colony algorithm is used to optimize the decision variables (valve opening L and adjustable ratio k) in the pump-valve scheduling model to minimize the energy consumption of the centrifugal pump system.
[0039] Therefore, in the improved artificial bee colony algorithm, there are the following process modes: 1. Initialize the population: generate the initial population according to the Latin hypercube population initialization method, and each individual in the population represents a combination of valve opening L and speed ratio k; 2. Calculate the fitness value: for each individual in the population, use the pump-valve scheduling model to calculate the total energy consumption (fitness value); 3. Select and generate new individuals: select high-quality individuals (i.e. individuals with relatively low total energy consumption) based on the fitness value, and generate new individuals through the original search method and the new search method with elite guidance; 4. Iterative search: continuously iterate the above steps, update the individuals in the population, and gradually approach the optimal solution; 5. Application of the optimal solution: when the algorithm reaches the predetermined number of iterations, select the individual with the lowest fitness value as the optimal solution, which is the valve opening L and speed ratio k with the lowest energy consumption in the centrifugal pump system; 6. Real-time control: apply the optimal solution to the actual centrifugal pump system, and adjust the operation of the pump and valve in real time through the PLC and host computer program to achieve precise joint control of the pump and valve.
[0040] In this application, the programmable logic controller PLC in the centrifugal pump system collects the required flow rate Q at the end of the pipe network. r and required head H r , and the demand flow Q r and required head H r As a constraint of the pump-valve scheduling model, the improved artificial bee colony algorithm in the upper computer is used to calculate the valve opening L and speed ratio scheme when the energy consumption is lowest. The valve opening L and speed ratio k obtained by optimization calculation are applied to the actual centrifugal pump system to adjust the operation of the pump and valve to achieve precise joint control of the pump and valve of the centrifugal pump system.
[0041] It is worth noting that the improved artificial bee colony algorithm is implanted into the upper computer program of the centrifugal pump system. The improved artificial bee colony algorithm is also called the improved ABC algorithm. The software in the upper computer is written in LabVIEW language. The PLC lower computer of SIMATICS7-200 SMART controls data collection, pump speed and valve allocation. According to the demand flow Q measured at the end of the pipe network, r and required head H r The energy consumption of control scheme a obtained by the traditional method is compared with that of control scheme b obtained by optimizing the improved artificial bee colony algorithm, as shown in Table 2 below.
[0042] It can be seen from Table 2 that the energy consumption of control scheme b obtained by the improved artificial bee colony algorithm under different working conditions is always better than that of control scheme a obtained by the traditional method, and the method proposed in this application can give clear speed ratio k and valve opening L parameters to realize the automatic control of relevant operating parameters of the centrifugal pump system.
[0043] Table 2 Finally, the pump-valve joint scheduling method based on the improved artificial bee colony algorithm was used to stably and effectively obtain the optimal speed ratio k and valve opening L control scheme that meets the current operating conditions, and a better variable frequency control method was achieved. This can further provide solution guidance for variable operating conditions and reduce the energy consumption of the centrifugal pump system.
[0044] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.
Claims
1. A pump-valve combined pump group scheduling method based on an improved artificial bee colony algorithm, characterized in that: The steps include: Step S100, adjusting the speed ratio k of the frequency converter and the valve opening L to obtain test sample data of flow rate Q, head H and shaft power P under different valve openings L; Step S200, drawing and fitting the valve opening-shaft power diagram under different speed ratios k, and obtaining the P(L, k) relationship through nonlinear regression; Step S300, establishing a pump valve scheduling model with the goal of minimizing energy consumption; Step S400, improving the artificial bee colony algorithm according to the centrifugal pump system constraints; Step S500: implanting a pump-valve scheduling model based on the improved artificial bee colony algorithm into a host computer of a centrifugal pump system, and realizing pump-valve joint control of the centrifugal pump system based on real-time measurement data.
2. A pump-valve combined pump group scheduling method based on improved artificial bee colony algorithm according to claim 1, characterized in that: The speed ratio k of the inverter is adjusted by adjusting the output frequency of the inverter, and the speed ratio k = output frequency / 50.
3. The method for pump-valve combined pump group scheduling based on improved artificial bee colony algorithm according to claim 1 is characterized in that: In step S100, after obtaining the test sample data of flow rate Q, head H and shaft power P under different valve openings L, the collected test sample data is preprocessed for data rationality screening and elimination, and the experiment is repeated multiple times to obtain the final test sample data.
4. The method for pump-valve combined pump group scheduling based on improved artificial bee colony algorithm according to claim 1 is characterized in that: In step S200, a valve opening-shaft power diagram under different speed ratios k is drawn and fitted, and a P(L, k) relationship is obtained by nonlinear regression, which also includes the following steps: Step S210, using valve opening L as the horizontal coordinate and shaft power P as the vertical coordinate, a valve opening-shaft power diagram under different speed ratios k is drawn; Step S220, based on the flow-shaft power characteristic equation and the valve opening-flow characteristic equation, the valve opening-shaft power relationship under different speed ratios k is obtained by nonlinear regression. P(L,k)=c0·(Q max ·R L-1 ) 2 +c1·(Q max ·R L-1 )·k+c2·k 2 ; Among them, the flow-shaft power characteristic equation is as follows: P=c0·Q 2 +c1·Q·k+c2·k 2 ; In the flow-shaft power characteristic equation, P is the shaft power, Q is the flow, k is the speed ratio, and c0, c1 and c2 are fitting coefficients; Among them, the valve opening-flow characteristic equation is as follows: Q=Q max ·R L-1 ; Q in the valve opening-flow characteristic equation max is the maximum flow rate that the valve can adjust, R is the valve adjustable ratio, and L is the valve opening.
5. The method for pump-valve combined pump group scheduling based on improved artificial bee colony algorithm according to claim 1 is characterized in that: In step S300, The demand flow Q r and required head H r Given the conditions of the centrifugal pump system, the pump-valve scheduling model is established using the valve opening L and the speed ratio k as the determining variables; The formula of the pump-valve scheduling model is as follows: Where n is the pump serial number; m is the total number of pumps; Q r and H r is the required flow and required head; Q n is the flow rate of pump n; k n is the speed ratio of pump n.
6. A pump-valve combined pump group scheduling method based on improved artificial bee colony algorithm according to claim 5, characterized in that: In step S400, the following steps are included: Step S410, according to the centrifugal pump system constraints, the artificial bee colony algorithm is improved to optimize the initial population distribution and individual generation method, and the Latin hypercube population initialization method is used to optimize the initial population distribution. The formula of the Latin hypercube population initialization method is as follows: x ij =UB i +rand(0,1)·Δx i +(i-1)·Δx i ; Where: x ij is the parameter value of the i-th individual in the j-th dimension, also known as the honey source, i is the serial number of the individual in the initial population of the algorithm, and j is the serial number of the optimization parameter dimension; UB i and LB i are the upper and lower limits of each parameter; t is the number of subintervals divided by the Latin hypercube method; Δx i is the length of the parameter i subinterval; Step S420, adopting a method of equally selecting an original search method and an elite guided search method for the individual generation method; The equation of the original search method is: The equation for the elite guided search method is: in: is the new parameter value generated by the i-th individual in the j-th dimension, also known as the new nectar source; a is the acceleration coefficient, usually 1; is a random number uniformly distributed between [-1, 1]; w is a parameter different from i; x Bj is the optimal nectar source in the jth dimension.
7. The method for pump-valve combined pump group scheduling based on improved artificial bee colony algorithm according to claim 6 is characterized in that: In step S500, the following steps are included: Step S510, implanting a pump-valve scheduling model based on an improved artificial bee colony algorithm into a host computer of a centrifugal pump system; Step S520: collect the demand flow Q at the end of the pipe network in real time r and required head H r , as the input of the pump-valve scheduling model; Step S530, the valve opening L and the speed ratio k are used as two parameters of the individuals in the improved artificial bee colony algorithm, and the power P tot As the optimization goal of improving the artificial bee colony algorithm; Step S540: When the improved artificial bee colony algorithm converges or reaches a predetermined number of iterations, the individual with the lowest fitness value is selected as the optimal solution. The optimal solution is the power P in the pump valve scheduling model. tot Minimum valve opening L and speed ratio k; Step S550: Apply the valve opening L and the speed ratio k calculated by the improved artificial bee colony algorithm to the centrifugal pump system for regulation.
