Balance adjustment method of heating secondary network based on three-stage efficient collaborative algorithm

Abstract

The present invention relates to the technical field of intelligent optimization of the balance of the secondary network of a regional heating system, and specifically to a method for adjusting the balance of a heating secondary network based on a three-stage efficient collaborative algorithm. It includes initializing algorithm parameters, generating an initial solution, evaluating the newly generated initial solution and marking the optimal solution, and using the newly generated solution as the initial solution group of the algorithm; executing the search phase, the diffusion phase, and the reconnaissance phase. After each round of iteration, the newly generated solution in the current round of iteration is evaluated and marked to obtain a new optimal solution. If the number of algorithm termination iterations is met, the iteration is stopped and the current optimal solution is output. The present invention solves the problem of difficulty in regulating the balance of the secondary network due to hydraulic balance, and has the positive effect of improving production efficiency and reducing production costs.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent optimization of secondary network balance in a district heating system, and specifically to a heating secondary network balance adjustment method based on a three-stage efficient collaborative algorithm suitable for a modern new type of intelligent district heating system. Background Art

[0002] With the acceleration of urbanization and the increase in demand for energy conservation and environmental protection, heating companies are increasingly using intelligent technologies to optimize heating systems. The current intelligent heating system relies on the Internet of Things technology to achieve remote monitoring and control of the heating network, and uses advanced algorithms to optimize heat energy distribution, making the heating system more efficient and intelligent. Heating energy consumption accounts for about two-thirds of the total energy consumption of buildings, especially in China, where heating energy consumption is 2 to 3 times higher than other areas at the same latitude. Therefore, improving the energy utilization rate of the heating system through intelligent transformation has become a key goal of the industry.

[0003] Although the application of intelligent heating systems has achieved certain results, such as an increase of 15%-20% in energy utilization and a reduction of 10%-12% in energy consumption in some areas, the existing technology still has some shortcomings. For example, the traditional manual adjustment of heating leads to uneven heat distribution, affecting user experience. Although the current intelligent transformation achieves precise heat distribution through the automatic control of the secondary pipe network balancing valve, in a complex heating network, facing the fluctuation of heating demand and the delay of heat energy transmission, how to achieve the optimal adjustment of the balancing valve is still a major challenge.

[0004] Existing solutions, such as intermittent heating mode (IHM), sequential quadratic programming (SQP) algorithm, particle swarm optimization (PSO) and recursive deterministic policy gradient (RDPG) algorithm based on deep reinforcement learning, have solved the balancing valve adjustment problem to a certain extent, but there is still room for improvement, especially in the overall hydraulic balance control in complex networks. Therefore, it is crucial for heating companies to develop more efficient optimization methods. Summary of the invention

[0005] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a heating secondary network balancing adjustment method based on a three-stage efficient collaborative algorithm, which can be directly applied to the smart heating system of the current heating enterprises, solving the problem of difficult regulation in the secondary network balance due to hydraulic balance, and the problem of uneven heating and cooling at the far and near ends, avoiding unnecessary thermal energy expenditure and thus reducing production costs.

[0006] The present invention provides a heat supply secondary network balance adjustment method based on a three-stage efficient collaborative algorithm, which is characterized by comprising the following steps:

[0007] Step 1, digitizing the secondary network heating system according to the information of the secondary network heating system;

[0008] Step 2: Initialize the algorithm parameters, set the number of valves to be adjusted Vn, the number of initial solutions Vt, and the number of algorithm termination iterations , critical count of the reconnaissance operator ;

[0009] Step 3: Determine the number of users in the secondary network heating system , main pipe radius and main pipe flow rate , secondary pipe radius , the length of heating pipeline at each user end and heating area ;

[0010] Step 4: Based on the problem size and the number of users in the secondary network system , randomly generate multiple high-quality initial solutions within the constraints ;

[0011] Step 5: Calculate the total flow in the heating system , flow rate of secondary pipeline at each user end , the time it takes for each user end to complete a thermal cycle ;

[0012] Step 6: The initial solution generated in step 4 Evaluate and mark the optimal solution , and use the newly generated solution as the initial solution group of the algorithm;

[0013] Step 7: In the search phase, a spiral operator combining the acceleration strategy and the offset strategy is used to find the current optimal solution represented in step 4. As the center, other solutions in the solution group are used as starting points, and two different spiral lines are randomly selected as trajectories to move closer to the current optimal solution. The new solutions obtained in the search phase are evaluated, and the new solutions that are better than the previous solutions are replaced to obtain the updated solutions. , the rest of the solutions remain unchanged, and the solutions that failed to evolve in the search phase are recorded and cleared to update the solutions After each iteration, the solution in the solution group is re-evaluated and the new current optimal solution is selected. ;

[0014] Step 8: In the diffusion phase, the diffusion operator is updated with the solution obtained in step 7. Centered on updating solutions The distance to the random solution in the solution group is the radius Second local search, evaluate the new solution group obtained in the local search, if there is a solution that is better than the previous solution, replace it, otherwise keep the previous solution;

[0015] Step 9: In the reconnaissance phase, the solution set is tested, where the number of times it has not evolved exceeds the critical count of the reconnaissance operator The solution is initialized, the scout operator randomly generates a new solution to replace it, and resets the count of its number of times it has not evolved, where the critical count of the scout operator is It increases with the number of iterations;

[0016] Step 10: After each round of iteration, evaluate the newly generated solution in this round of iteration and mark the new optimal solution At the same time, check whether the termination condition is met. If it is met, stop the iteration and output the current optimal solution If not satisfied, return to step 7 to start a new round of iteration.

[0017] Furthermore, in step 4, the initial solution The generation criteria are as follows:

[0018] .

[0019] Furthermore, in step 4, the generated initial solution The following constraints are met, including:

[0020] The sum of the instantaneous heat value supplied by each user terminal does not exceed the instantaneous heat value of the heat exchange station, that is,

[0021] ;

[0022] The sum of the flow rates of the secondary pipes does not exceed the flow rate of the main pipe, that is,

[0023] ;

[0024] The heating value of each user end shall not be lower than the minimum heating value, that is,

[0025] ;

[0026] The return water temperature of each user end is the same, that is,

[0027] ;

[0028] in, The current client ID. is the valve opening position, is the instantaneous heating value at the user end, is the specific heat capacity of air, is the decision parameter of the scheme, is the instantaneous heating value of the heat exchange station, is the secondary pipeline flow rate, is the minimum instantaneous heating value, is the target room temperature, is the current room temperature of the user terminal, is the minimum energy consumption calculation parameter, It is the detection cycle of the secondary network thermal system. It is the return water temperature at the user end.

[0029] Furthermore, in step 5, the total flow of the heating system is calculated The formula is as follows,

[0030] .

[0031] Further, in step 5, the flow rate of the secondary pipeline at the user end The solution formula is as follows:

[0032]

[0033] in, represents the flow distribution coefficient, Indicates the current valve opening.

[0034] Furthermore, in step 5, the time for the user end to complete a thermal cycle is The calculation formula is,

[0035] .

[0036] Furthermore, in step 6, the solution newly generated in step 4 Evaluate and mark the optimal solution , where the initial solution evaluation formula is as follows,

[0037] ;

[0038] ;

[0039] in, It is the heating index of the user side. Heat load parameters, The instantaneous heating value of the entire heating system, is the instantaneous heating value at the user end, is the decision parameter of the scheme, The current client ID. is the valve opening position, It is the total gear of valve opening.

[0040] Furthermore, in step 7, the two spiral trajectory formulas randomly selected by the spiral operator are as follows:

[0041]

[0042] in, To update the solution, is the current optimal solution, , are all offset parameters, The current selected solution, , are random numbers perturbed between (0, 1) and (-1, 1), is a random number between (0, 1).

[0043] Furthermore, in step 8, the calculation formula for the local search performed by the diffusion operator in the diffusion stage is as follows:

[0044]

[0045] in, is a random solution in the solution set, is a perturbed random number.

[0046] Further, in step 9, the critical count of the reconnaissance operator The value of is determined by the following formula:

[0047]

[0048] in, is the current iteration number, is the target iteration number, is the perturbed random number, is the given constant.

[0049] Furthermore, in step 10, the termination condition is the current number of iterations. Whether the target number of iterations is met If satisfied, stop iterating and output the current optimal solution Otherwise, return to step 7 to start a new round of iteration.

[0050] The heat supply secondary network balance adjustment method based on the three-stage efficient collaborative algorithm provided by the present invention has the following obvious outstanding substantial features and significant advantages compared with the prior art:

[0051] 1. The present invention is a multi-operator efficient collaboration algorithm proposed based on the problem characteristics and problem scale;

[0052] 2. Aiming at the problem of balancing the secondary network, the present invention proposes a spiral operator and a local search rescheduling strategy that combines an acceleration strategy with an offset strategy. Compared with some existing traditional control algorithms, the present invention effectively solves the problems of uneven heating and cooling at the far and near ends and slow regulation response caused by hydraulic imbalance in the heating system. By optimizing the valve opening and dynamic rescheduling strategy, energy waste and overheating in the heating system are avoided, heating energy expenses are significantly reduced, and the overall operational efficiency of heating companies is improved, which has the positive effect of reducing heating costs and improving energy utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a flow chart for realizing the present invention;

[0054] Figure 2 A schematic diagram of the secondary network and user terminal deployment of the method of the present invention;

[0055] Figure 3 It is a schematic diagram of the comparison of the evolution curves of the method of the present invention and the existing method;

[0056] Figure 4 It is the optimal heating value convergence diagram of the method of the present invention. DETAILED DESCRIPTION

[0057] like Figure 1 and Figure 2 The present invention provides a heat supply secondary network balancing adjustment method based on a three-stage efficient collaborative algorithm, which is implemented through the following steps.

[0058] Step 1: digitize the secondary network heating system according to the information of the secondary network heating system.

[0059] Step 2: Initialize the algorithm parameters, set the number of valves to be adjusted Vn, the number of initial solutions Vt, and the number of algorithm termination iterations , critical count of the reconnaissance operator .

[0060] Step 3: Determine the number of users in the secondary network heating system , main pipe radius and main pipe flow rate , secondary pipe radius , the length of heating pipeline at each user end and heating area .

[0061] Step 4: Based on the problem size and the number of users in the secondary network system , randomly generate multiple high-quality initial solutions within the constraints Specifically, the initial solution The generation criteria are as follows:

[0062] .

[0063] The generated initial solution The following constraints are met, including:

[0064] The sum of the instantaneous heat value supplied by each user terminal does not exceed the instantaneous heat value of the heat exchange station, that is,

[0065] ;

[0066] The sum of the flow rates of the secondary pipes does not exceed the flow rate of the main pipe, that is,

[0067] ;

[0068] The heating value of each user end shall not be lower than the minimum heating value, that is,

[0069] ;

[0070] The return water temperature of each user end is the same, that is,

[0071] ;

[0072] in, The current client ID. is the valve opening position, is the instantaneous heating value at the user end, is the specific heat capacity of air, is the decision parameter of the scheme, is the instantaneous heating value of the heat exchange station, is the secondary pipeline flow rate, is the minimum instantaneous heating value, is the target room temperature, is the current room temperature of the user terminal, is the minimum energy consumption calculation parameter, It is the detection cycle of the secondary network thermal system. It is the return water temperature at the user end.

[0073] Step 5: Calculate the total flow in the heating system , flow rate of secondary pipeline at each user end , the time it takes for each user end to complete a thermal cycle Specifically, calculate the total flow of the heating system The formula is as follows,

[0074] ;

[0075] Flow rate of secondary pipeline at user end The solution formula is as follows:

[0076]

[0077] in, represents the flow distribution coefficient, Indicates the current valve opening;

[0078] The time it takes for the user to complete a thermal cycle The calculation formula is,

[0079] .

[0080] Step 6: The initial solution generated in step 4 Evaluate and mark the optimal solution , and use the newly generated solution as the initial solution group of the algorithm. Specifically, for the newly generated solution in step 4 Evaluate and mark the optimal solution , where the initial solution evaluation formula is as follows,

[0081] ;

[0082] ;

[0083] in, It is the heating index of the user side. Heat load parameters, The instantaneous heating value of the entire heating system.

[0084] Step 7: In the search phase, a spiral operator combining the acceleration strategy and the offset strategy is used to find the current optimal solution represented in step 4. As the center, other solutions in the solution group are used as starting points, and two different spiral lines are randomly selected as trajectories to move closer to the current optimal solution. The new solutions obtained in the search phase are evaluated, and the new solutions that are better than the previous solutions are replaced to obtain the updated solutions. , the rest of the solutions remain unchanged, and the solutions that failed to evolve in the search phase are recorded and cleared to update the solutions After each iteration, the solution in the solution group is re-evaluated and the new current optimal solution is selected. Specifically, the two spiral trajectory formulas randomly selected by the spiral operator are as follows:

[0085]

[0086] in, To update the solution, is the current optimal solution, , are all offset parameters, The current selected solution, , are random numbers perturbed between (0, 1) and (-1, 1), is a random number between (0, 1).

[0087] Step 8: In the diffusion phase, the diffusion operator is updated with the solution obtained in step 7. Centered on updating solutions The distance to the random solution in the solution group is the radius The local search is performed again, and the new solution group obtained in the local search is evaluated. If there is a solution that is better than the previous solution, it is replaced, otherwise the previous solution is retained. Specifically, the calculation formula for the local search performed by the diffusion operator in the diffusion stage is as follows:

[0088]

[0089] in, is a random solution in the solution set, is a perturbed random number.

[0090] Step 9: In the reconnaissance phase, the solution set is tested, where the number of times it has not evolved exceeds the critical count of the reconnaissance operator The solution is initialized, the scout operator randomly generates a new solution to replace it, and resets the count of its number of times it has not evolved, where the critical count of the scout operator is As the number of iterations increases, it increases. Specifically, the critical count of the reconnaissance operator The value of is determined by the following formula:

[0091]

[0092] in, is the current iteration number, is the target iteration number, is the perturbed random number, is the given constant.

[0093] Step 10: After each round of iteration, evaluate the newly generated solution in this round of iteration and mark the new optimal solution At the same time, check whether the termination condition is met. If it is met, stop the iteration and output the current optimal solution If not, return to step 7 to start a new round of iteration. In step 10, the termination condition is the current number of iterations. Whether the target number of iterations is met If satisfied, stop iterating and output the current optimal solution Otherwise, return to step 7 to start a new round of iteration.

[0094] The present invention is further described below by listing a specific embodiment of the present invention.

[0095] This example is applicable to a smart heating system in which a single heat exchange station connects to multiple users. Solve a scheduling solution that can reduce the total energy consumption cost, and the heating system can reduce the instantaneous heating value as much as possible within the constraints. The initial flow rate of the pipeline of the heating system is 2m / s, and the diameters of the primary and secondary pipelines are 0.65m and 0.2m respectively. The initial parameters of the embodiment of each user unit module number, water inlet pipeline flow rate, water inlet temperature, user-side pipeline length, heating area, heating heat index, and current room temperature are shown in Table 1.

[0096] Table 1 Initial parameters of the application example of the present invention

[0097]

[0098] At the same time, set the parameters required for algorithm initialization, and adjust the number of valves Vn, the number of initialization solutions Vt, and the target number of iterations , critical count of the reconnaissance operator . Where Vn=10, Vt=10, , SN=5.

[0099] The experiment was repeated ten times under the above conditions, and the experimental results are shown in Table 2.

[0100] Table 2 Experimental results of the application examples of the present invention

[0101]

[0102] In ten repeated experiments under the conditions of this embodiment, the solution obtained by the three-stage efficient collaborative combination algorithm (SDS-CA) of the present invention is consistent with the exact solution obtained by MATLAB. The SDS-CA algorithm can search for the target solution we need in the solution space. At the same time, in ten repeated experiments of the same example, the RPI of the SDS-CA algorithm is significantly smaller than that of other algorithms. Therefore, the algorithm performance of the three-stage efficient collaborative combination algorithm (SDS-CA) of the present invention is significantly better than that of other algorithms in the table.

[0103] like Figure 3 and Figure 4 As shown, under the condition of large-scale examples, the three-stage efficient collaborative combination algorithm (SDS-CA) of the present invention is superior to the compared algorithms in solving problems compared to the binary cat colony optimization (SBCSO) algorithm, the binary bee colony (NBABC) algorithm, the discrete whale optimization algorithm (DWOA) and the hybrid rice optimization (HRO) algorithm. In order to reduce the error of the experiment and make the calculation results more effective and general, each example is repeated 30 times to generate statistical results. The present invention uses relative percentage increase (RPI) as a performance evaluation indicator, and the calculation formula of RPI is:

[0104]

[0105] in, represents the optimal solution obtained by a specific algorithm when solving a case, and It indicates the optimal solution of the five algorithms. Obviously, the smaller the RPI value, the better the solution found by the algorithm and the better the performance.

[0106] In summary, the three-stage efficient collaborative combination algorithm (SDS-CA) of the present invention can be directly applied to the current intelligent heating system of heating enterprises, and can effectively solve the problems of uneven heating and cooling at the far and near ends and slow control response caused by hydraulic imbalance in the heating system. By optimizing the valve opening and dynamic rescheduling strategy, energy waste and overheating in the heating system are avoided, the heating energy expenditure is significantly reduced, the overall operating efficiency of the heating enterprise is improved, and it helps to reduce heating costs and improve energy utilization. Compared with the binary cat colony optimization (SBCSO) algorithm, the binary bee colony (NBABC) algorithm, the discrete whale optimization algorithm (DWOA) and the hybrid rice optimization (HRO) algorithm, the three-stage efficient collaborative combination algorithm (SDS-CA) of the present invention can obtain a more outstanding solution (i.e., the optimal solution), so the application of the present invention can effectively improve the production efficiency of the heating system and reduce production costs.

Claims

1. A heat supply secondary network balance adjustment method based on a three-stage efficient collaborative algorithm, characterized in that: The following steps are included: Step 1, digitizing the secondary network heating system according to the information of the secondary network heating system; Step 2: Initialize the algorithm parameters, set the number of valves to be adjusted Vn, the number of initial solutions Vt, and the number of algorithm termination iterations , critical count of the reconnaissance operator ; Step 3: Determine the number of users in the secondary network heating system , main pipe radius and main pipe flow rate , secondary pipe radius , the length of heating pipeline at each user end and heating area ; Step 4: Based on the problem size and the number of users in the secondary network system , randomly generate multiple high-quality initial solutions within the constraints ; Step 5: Calculate the total flow in the heating system , flow rate of secondary pipeline at each user end , the time it takes for each user end to complete a thermal cycle ; Step 6: The initial solution generated in step 4 Evaluate and mark the optimal solution , and the newly generated solution is used as the initial solution group of the algorithm; the initial solution evaluation formula is as follows, ； ； in, It is the heating index of the user side. Heat load parameters, The instantaneous heating value of the entire heating system, is the instantaneous heating value at the user end, is the decision parameter of the scheme, The current client ID. is the valve opening position, is the total gear position of the valve opening; Step 7: In the search phase, a spiral operator combining the acceleration strategy and the offset strategy is used to find the current optimal solution represented in step 4. As the center, other solutions in the solution group are used as starting points, and two different spiral lines are randomly selected as trajectories to move closer to the current optimal solution. The new solutions obtained in the search phase are evaluated, and the new solutions that are better than the previous solutions are replaced to obtain the updated solutions. , the rest of the solutions remain unchanged, and the solutions that failed to evolve in the search phase are recorded and cleared to update the solutions After each iteration, the solution in the solution group is re-evaluated and the new current optimal solution is selected. ; Step 8: In the diffusion phase, the diffusion operator is updated with the solution obtained in step 7. Centered on updating solutions The distance to the random solution in the solution group is the radius Second local search, evaluate the new solution group obtained in the local search, if there is a solution that is better than the previous solution, replace it, otherwise keep the previous solution; Step 9: In the reconnaissance phase, the solution set is tested, where the number of times it has not evolved exceeds the critical count of the reconnaissance operator The solution is initialized, the scout operator randomly generates a new solution to replace it, and resets the count of its number of times it has not evolved, where the critical count of the scout operator is It increases with the number of iterations; Step 10: After each round of iteration, evaluate the newly generated solution in this round of iteration and mark the new optimal solution At the same time, check whether the termination condition is met. If it is met, stop the iteration and output the current optimal solution If not satisfied, return to step 7 to start a new round of iteration.

2. According to claim 1, a heat supply secondary network balance adjustment method based on a three-stage efficient collaborative algorithm is characterized in that: In step 4, the initial solution The generation criteria are as follows: 。 3. According to claim 1, a method for balancing and regulating the secondary heating network based on a three-stage efficient collaborative algorithm is characterized in that: In step 4, the generated initial solution The following constraints are met, including: The sum of the instantaneous heat value supplied by each user terminal does not exceed the instantaneous heat value of the heat exchange station, that is, ； The sum of the flow rates of the secondary pipes does not exceed the flow rate of the main pipe, that is, ； The heating value of each user end shall not be lower than the minimum heating value, that is, ； The return water temperature of each user end is the same, that is, ； in, The current client ID. is the valve opening position, is the instantaneous heating value at the user end, is the specific heat capacity of air, is the decision parameter of the scheme, is the instantaneous heating value of the heat exchange station, is the secondary pipeline flow rate, is the minimum instantaneous heating value, is the target room temperature, is the current room temperature of the user terminal, is the minimum energy consumption calculation parameter, It is the detection cycle of the secondary network thermal system. It is the return water temperature at the user end.

4. According to claim 1, a heat supply secondary network balance adjustment method based on a three-stage efficient collaborative algorithm is characterized in that: In step 5, calculate the total flow of the heating system The formula is as follows, 。 5. According to claim 1, a method for balancing and regulating the secondary heating network based on a three-stage efficient collaborative algorithm is characterized in that: In step 5, the flow rate of the secondary pipeline at the user end The solution formula is as follows: in, represents the flow distribution coefficient, Indicates the current valve opening.

6. The method for balancing and regulating the secondary heating network based on a three-stage efficient collaborative algorithm according to claim 1 is characterized in that: In step 5, the time it takes for the user to complete a thermal cycle The calculation formula is, 。 7. The method for balancing and regulating the secondary heating network based on a three-stage efficient collaborative algorithm according to claim 1 is characterized in that: In step 7, the two spiral trajectory formulas randomly selected by the spiral operator are as follows: in, To update the solution, is the current optimal solution, , are all offset parameters, The current selected solution, , are random numbers perturbed between (0, 1) and (-1, 1), is a random number between (0, 1).

8. A heat supply secondary network balance adjustment method based on a three-stage efficient collaborative algorithm according to claim 7, characterized in that: In step 8, the calculation formula for the local search performed by the diffusion operator in the diffusion stage is as follows, in, is a random solution in the solution set, is a perturbed random number.

9. The method for balancing and regulating the secondary heating network based on a three-stage efficient collaborative algorithm according to claim 1 is characterized in that: In step 9, the critical count of the scout operator The value of is determined by the following formula: in, is the current iteration number, is the target iteration number, is the perturbed random number, is the given constant.

Images