Multi-constraint intelligent scheduling method, system and equipment for electric power metering equipment and medium

Abstract

The invention discloses an electric power metering equipment multi-constraint intelligent scheduling method, system, equipment and medium, and belongs to the technical field of electric power metering, and the method comprises the steps: building a metering equipment demand point, supply point, transportation time and safety intelligent scheduling model, and calculating the demand point importance index of the metering equipment based on the demand point importance index and the metering equipment demand rate; a demand priority is calculated through weighted double-target proximity; a particle swarm is initialized and divided into a main swarm and multiple dynamic subswarms, the multiple dynamic subswarms are independently evolved and periodically recombined, the main swarm updates the speed through a dynamic jitter strategy and learns optimal particles of the subswarms, and a progressive confinement strategy is adopted to adaptively adjust and constrain a feasible region range; and screening supply points meeting the latest arrival time limit, generating a transportation queue according to a time urgency descending order, and outputting a scheduling scheme through multi-subgroup collaborative optimization. According to the invention, an efficient, reliable and flexible solution is provided in the production and transportation scheduling of the electric power metering equipment, and the overall operation efficiency and the service quality can be obviously improved.

Description

Technical Field

[0001] This invention relates to the field of power metering technology, specifically to a multi-constraint intelligent scheduling method, system, equipment, and medium for power metering equipment. Background Technology

[0002] In modern power systems, electricity metering equipment plays a crucial role. These devices are used not only to measure electricity consumption but also to monitor the operational status of the power system, ensuring the effective allocation and rational use of power resources. With the continuous growth of electricity demand and the gradual opening of the electricity market, the demand for electricity metering equipment is also constantly increasing. Against this backdrop, how to efficiently produce and dispatch electricity metering equipment has become an urgent problem to be solved.

[0003] Traditional methods of dispatching power metering equipment typically rely on experience and manual operation. This approach is not only inefficient but also prone to errors, leading to equipment delays in reaching demand locations and impacting the normal operation of the power system. With the development of information technology and intelligent algorithms, more and more researchers are focusing on how to utilize modern technologies to optimize the dispatching process of power metering equipment.

[0004] The introduction of intelligent scheduling models has significantly improved the efficiency of power metering equipment scheduling. By establishing mathematical models and combining them with optimization algorithms, optimal scheduling schemes can be found under multiple constraints. These constraints include equipment demand, supply, transportation time, and transportation safety. The goal of intelligent scheduling models is to maximize the number of safe arrivals of equipment while minimizing transportation time and costs.

[0005] In intelligent scheduling models, transportation time and safety are two crucial considerations. Transportation time directly impacts whether equipment can reach its destination within the stipulated time, while transportation safety relates to the risk of damage during transport. Therefore, both factors must be considered comprehensively during scheduling to ensure the safe and timely arrival of equipment.

[0006] Furthermore, the dispatching of power metering equipment is also affected by various uncertainties, such as weather conditions and traffic conditions. These uncertainties bring additional challenges to the dispatching process. Therefore, intelligent dispatching models need to incorporate a certain degree of randomness and flexibility to cope with the impact of these uncertainties.

[0007] To improve the efficiency and accuracy of scheduling models, researchers have introduced various optimization algorithms, such as genetic algorithms, particle swarm optimization (PSO), and ant colony optimization (ACO). These algorithms, by simulating natural evolutionary processes or group behavior, can effectively search for the optimal solution to the problem. In the scheduling of power metering equipment, PSO has been widely used due to its advantages such as fast convergence speed and simple implementation.

[0008] However, traditional particle swarm optimization algorithms often face a trade-off between convergence and diversity when dealing with multi-constraint optimization problems. To address this issue, researchers have proposed various improvement strategies, such as dynamic subswarming strategies and asymptotically restricted domain strategies. These strategies can effectively improve the algorithm's performance by adjusting the search space and parameters during the optimization process.

[0009] The dynamic subswarm strategy divides the particle swarm into multiple subswarms, allowing each subswarm to search in different regions, thereby improving the algorithm's diversity and global search capability. The progressive confinement strategy, on the other hand, sets a larger search range in the early stages of optimization and gradually reduces the feasible region, allowing the search process to gradually focus on solutions that better meet the constraints, thus improving search accuracy.

[0010] In practical applications, the scheduling model for power metering equipment needs to be customized to meet specific application scenarios and requirements. For example, in some emergency situations, priority may need to be given to transportation time, while in other situations, priority may need to be given to transportation cost and safety. Therefore, the design of the scheduling model needs to be flexible and adaptable to meet the needs of different application scenarios.

[0011] In summary, the dispatching of power metering equipment is a complex multi-constraint optimization problem involving numerous factors and uncertainties. By introducing intelligent dispatching models and optimization algorithms, the efficiency and accuracy of the dispatching process can be effectively improved, ensuring the timely and safe arrival of power metering equipment. This not only helps improve the operational efficiency of the power system but also provides strong support for the healthy development of the power market. In the future, with continuous technological advancements and evolving application needs, power metering equipment dispatching models will continue to develop and improve, bringing more innovation and breakthroughs to the power industry. Summary of the Invention

[0012] In view of the above-mentioned problems, the present invention is proposed.

[0013] Therefore, this invention aims to solve the following key technical problems faced in the dispatching process of power metering equipment:

[0014] Optimal scheduling under multiple constraints: The scheduling of power metering equipment involves various constraints, including equipment demand, supply, transportation time, and transportation safety. These constraints interact, complicating the scheduling problem. Traditional methods struggle to find the optimal scheduling solution within limited time and resources. This invention, by constructing an intelligent scheduling model, can optimize equipment scheduling under multiple constraints.

[0015] Improving the safety and timeliness of equipment transportation: Equipment may be affected by various factors during transportation, such as weather and traffic conditions, leading to extended transportation time or equipment damage. This invention introduces transportation safety and transportation time as optimization objectives to ensure that equipment can arrive at the required location safely and on time.

[0016] Addressing Uncertainties in the Scheduling Process: The scheduling process involves numerous uncertainties, such as sudden changes in demand and supply chain fluctuations. These uncertainties increase the complexity and difficulty of scheduling. This invention improves the adaptability of the scheduling model to uncertainties by introducing stochastic and flexible strategies.

[0017] Balancing Convergence and Diversity in Optimization Algorithms: When using optimization algorithms (such as particle swarm optimization) for scheduling, a trade-off between convergence and diversity is often encountered. Traditional algorithms may become overly focused on local optima, making it difficult to escape them. This invention balances convergence and diversity through a dynamic subgroup strategy and a progressive domain constraint strategy, thereby improving global search capability and accuracy.

[0018] Improving the flexibility and adaptability of the scheduling model: Scheduling requirements may differ across application scenarios. For example, in some cases, priority may be given to transportation time, while in others, priority may be given to transportation cost. The scheduling model design of this invention is flexible and adaptable, capable of being adjusted according to specific needs to meet the requirements of different scenarios.

[0019] By solving the above technical problems, the present invention can significantly improve the efficiency and accuracy of power metering equipment dispatch, ensure that the equipment arrives safely at the required location within the specified time, and support the efficient operation of the power system.

[0020] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a multi-constraint intelligent scheduling method for power metering equipment, comprising,

[0021] An intelligent scheduling model is established to consider the demand points, supply points, transportation time, and safety of metering equipment. Based on the importance index of demand points and the demand rate of metering equipment, demand priority is calculated through weighted bi-objective proximity. A particle swarm is initialized and divided into a main group and multiple dynamic subgroups. The multiple dynamic subgroups evolve independently and reorganize periodically. The main group updates its speed through a dynamic jitter strategy and learns the optimal particles of the subgroups. A progressive domain constraint strategy is used to adaptively adjust the range of the feasible domain. Supply points that meet the latest arrival time limit are selected, and transportation queues are generated in descending order of time urgency. The scheduling scheme is output through multi-subgroup collaborative optimization.

[0022] As a preferred embodiment of the multi-constraint intelligent scheduling method for power metering equipment described in this invention, the intelligent scheduling model includes: setting the storage areas for metering equipment into X1, X2, X3, ..., X...M The number of measuring devices needed for the warehouse is x. i (i=1,2,3,…,M),Y1,Y2,Y3,…,Y N The location from which metering equipment is supplied is given, with a supply quantity of y. j (j = 1, 2, 3, ..., N), y ij Represents Y j To X i The quantity of metering equipment supplied;

[0023] Make X i The optimal scheduling model for obtaining an accurate number of metering devices within a specified time is expressed as follows:

[0024]

[0025] Among them, s ji (0≤s ji ≤1) indicates from Y j To X i Safety, φ is the scheduling scheme, ω is the distance from the transport metering equipment to X i The set of all scheduling schemes, A i (φ) is the distance to X i The possible number of measuring devices, T(φ) is from Y j (j=1,2,3,…,Z) Transport metering equipment to X i The time set, X i Belongs to X1, X2, X3, ..., X M Y j Belongs to Y1, Y2, Y3, ..., Y N .

[0026] As a preferred embodiment of the multi-constraint intelligent scheduling method for power metering equipment described in this invention, the calculation of demand priority includes sorting the demand priorities according to the importance index of the metering equipment demand location and the metering equipment demand rate.

[0027] The importance indicators include the urgency of transporting the metering equipment, the difficulty of supplying it, time constraints for transportation, route conditions, and weather conditions. The importance of the locations where the metering equipment is needed is expressed as follows:

[0028]

[0029] Where, α i This represents the importance index of the location where the i-th metering device is needed, k is the variable index, and δ k ε is the weighting coefficient of the k-th indicator. k This represents the original evaluation value of the k-th indicator;

[0030] The demand rate for metering equipment is expressed as follows:

[0031]

[0032] Where, β i Location X i The demand rate for metering equipment, x i It is X i The demand for meters, where M is the total number of warehouses storing metering equipment.

[0033] As a preferred embodiment of the multi-constraint intelligent scheduling method for power metering equipment described in this invention, the calculation of demand priority further includes balancing the importance index value and the metering equipment demand rate by weighted dual-objective proximity, based on the fact that the importance index value is directly proportional to the priority and the demand rate is inversely proportional to the priority.

[0034] Will affect X i Priority importance index value and the closeness μ of the metering equipment demand rate to the optimal value i1 Perform balancing:

[0035]

[0036] Similarly, μ is obtained. i2 :

[0037]

[0038] X i Relative closeness μ to the optimal value i for:

[0039]

[0040] Where ξ1 is the weight of the importance of the demand location, and ξ2 is the weight of the demand rate of metering equipment, with ξ1 and ξ2 being 0.7 and 0.3 respectively; α i1 α i2 These are the maximum and minimum values ​​of the location importance index, respectively; β i1 β i2 These are the minimum and maximum values ​​of the demand rate for metering equipment, respectively.

[0041] As a preferred embodiment of the multi-constraint intelligent scheduling method for power metering equipment described in this invention, the intelligent scheduling model includes calculating X, considering the time constraints of metering equipment supply. i The optimized values ​​of all scheduling schemes are used to construct the scheduling scheme with the shortest execution time, as follows:

[0042]

[0043] Where ζ1 and ζ2 are two weighting indicators, with weights of 0.6 and 0.4 respectively; t 1i ,t 2i ,t 3i ,…,t pi ,…,t Ni From Y j (j=1,2,3,…,N) to X i The transportation time, T i It is the transport meter to X i The latest time.

[0044] As a preferred embodiment of the multi-constraint intelligent scheduling method for power metering equipment described in this invention, the initialization of the particle swarm and its division into a main group and multiple dynamic subgroups includes improving the constrained multi-objective particle swarm optimization with multi-subgroup cooperation through multi-objective optimization methods and constraint processing techniques.

[0045] The particles are initialized using the CMOPSO-IMSC algorithm. N particles are randomly selected to form the main group, and the remaining N particles are divided into multiple subgroups (where S is the number of particles in each subgroup).

[0046] For subgroups, a cyclical dynamic recombination strategy is used to iteratively update using the standard PSO update formula, randomly redistributing subgroups at a preset frequency, and providing the main group with the best subgroup particles as learning objects.

[0047] For the main group, a dynamic dithering update strategy is used to help it escape local optima. The main group learns by relying on the best particles in the subgroups. The velocity update formula for particle i in the main group is as follows:

[0048]

[0049]

[0050] Where ω is the inertia weight, c1 and c2 are learning factors, r1 and r2 are random numbers in the interval (0,1), d is the dimension, and subp d It is the optimal particle in all subgroups. It is the particle's own historical best.

[0051] As a preferred embodiment of the multi-constraint intelligent scheduling method for power metering equipment described in this invention, the dynamic jitter update strategy includes introducing a random jitter term into the speed update formula. Through the random jitter term, the particle speed update is subject to random interference, thus limiting the jitter range.

[0052] randfactor=unifrnd(-0.1,0.1,N,D)

[0053] Update speed based on the limited jitter range:

[0054]

[0055] The proposed method employs a progressive confinement strategy to adaptively adjust the feasible region range. This strategy sets an initial search range in the early stages of optimization, allowing particles to search within this range and increasing the likelihood of finding the global optimum. As the number of iterations increases, the value of ε further decreases, resulting in the feasible region range of:

[0056]

[0057] The value of ε is approximately equal to 0, meaning that the feasible region of ε gradually becomes equivalent to the real feasible region.

[0058] Another objective of this invention is to provide a multi-constraint intelligent scheduling system for power metering equipment.

[0059] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a multi-constraint intelligent scheduling system for power metering equipment, comprising: a model building module, a multi-subgroup collaborative optimization module, and a scheduling module;

[0060] The model building module establishes an intelligent scheduling model for the demand points, supply points, transportation time, and safety of metering equipment. Based on the importance index of the demand points and the demand rate of metering equipment, the demand priority is calculated through weighted dual-objective proximity.

[0061] The multi-subgroup collaborative optimization module initializes the particle swarm and divides it into a main group and multiple dynamic subgroups. The multiple dynamic subgroups evolve independently and reorganize periodically. The main group updates its speed through a dynamic jittering strategy and learns the optimal particles of the subgroups. It adopts a progressive domain constraint strategy to adaptively adjust the range of the constraint feasible domain.

[0062] The scheduling module filters supply points that meet the latest arrival time limit, generates transportation queues in descending order of time urgency, and outputs a scheduling scheme through multi-subgroup collaborative optimization.

[0063] The present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, characterized in that the processor executes the computer program to implement the steps of the aforementioned multi-constraint intelligent scheduling method for power metering equipment.

[0064] The present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, implements the steps of the aforementioned intelligent scheduling method for multi-constraint power metering equipment.

[0065] The beneficial effects of this invention are as follows: By establishing a multi-constraint intelligent scheduling model, this invention can quickly calculate the optimal equipment scheduling scheme, which significantly improves scheduling efficiency and reduces manual intervention and decision-making time.

[0066] This invention achieves optimized resource allocation of metering equipment by comprehensively considering factors such as transportation time, safety, and demand priority, ensuring that transportation costs and time are minimized while meeting demand. The introduction of dynamic jitter update strategy and progressive domain restriction strategy enables the algorithm to escape local optima when searching for the optimal solution, enhancing the robustness and adaptability of the system and helping to maintain efficient scheduling performance in dynamically changing environments.

[0067] By introducing and optimizing safety indicators, the safe arrival of metering equipment during transportation is ensured, reducing the risk of equipment damage and loss and improving supply reliability. The method of this invention can adaptively adjust the scheduling scheme according to actual needs and environmental changes, providing highly flexible scheduling capabilities and applicable to application scenarios of different scales and complexities. Through a multi-objective optimization model, it supports decision-makers to make optimization choices under various constraints, helping them to make more scientific and reasonable scheduling decisions.

[0068] Through these beneficial effects, the present invention provides an efficient, reliable and flexible solution for the production and transportation scheduling of power metering equipment, which can significantly improve overall operational efficiency and service quality. Attached Figure Description

[0069] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0070] Figure 1 This is a flowchart illustrating a multi-constraint intelligent scheduling method for power metering equipment, provided as an embodiment of the present invention.

[0071] Figure 2 This is a particle swarm reconstruction process diagram of a multi-constraint intelligent scheduling method for power metering equipment provided in one embodiment of the present invention.

[0072] Figure 3 The final population of CMOPSO-IMSC based on different RFs on MW9 is provided as an embodiment of the present invention for a multi-constraint intelligent scheduling method for power metering equipment.

[0073] Figure 4The test results of CMOPSO-IMSC and other algorithms for a multi-constraint intelligent scheduling method for power metering equipment provided in one embodiment of the present invention are shown.

[0074] Figure 5 The results of HV tests on the CMOPSO-IMSC and other algorithms of a multi-constraint intelligent scheduling method for power metering equipment provided in one embodiment of the present invention are shown.

[0075] Figure 6 The IGD test results of CMOPSO-IMSC and CMOPSO-IMSC2 for a multi-constraint intelligent scheduling method for power metering equipment provided in an embodiment of the present invention.

[0076] Figure 7 The IGD test results of CMOPSO-IMSC based on different S are provided for a multi-constraint intelligent scheduling method for power metering equipment according to an embodiment of the present invention.

[0077] Figure 8 This invention provides a solution process for a scheduling program in a multi-constraint intelligent scheduling method for power metering equipment, as an embodiment of the present invention. Detailed Implementation

[0078] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0079] Example 1, referring to Figures 1 to 7 This is one embodiment of the present invention, which provides a multi-constraint intelligent scheduling method for power metering equipment, including:

[0080] S1. Establish an intelligent scheduling model for the demand points, supply points, transportation time, and safety of metering equipment. Based on the importance index of demand points and the demand rate of metering equipment, calculate the demand priority through weighted dual-objective proximity.

[0081] Specifically, the warehouse for metering equipment is divided into X1, X2, X3, ..., X M The number of measuring devices they need is x. i (i = 1, 2, 3, ..., M). During this period, Y1, Y2, Y3, ..., Y N Let y be the locations that can supply metering equipment to external entities. j (j = 1, 2, 3, ..., N). t ji Indicates from Yj To X i The transportation time, s ji (0≤s ji ≤1) indicates from Y j To X i Security. ij Represents Y j To X i The quantity of metering equipment supplied. A i (i = 1, 2, 3, ..., M) represents the number of people who can safely reach X. i The number of measuring devices corresponds to the number of meters on the transportation route. ji and y ji The sum of the products of .

[0082] In practice, there are many scheduling schemes, such as φ={(Y1,y 1i ,t 1i ,s 1i ),(Y2,y 2i ,t 2i ,s 2i ),(Y3,y 3i ,t 3i ,s 3i ),…,(Y Z ,y Zi ,t Zi ,s Zi )}, where Z represents the destination X i The number of locations. Zi From Y Z Transport to X i The number of metering devices, y Z It is Y Z The number of metering devices that can be dispatched externally. During the dispatching process of metering devices, y Z It must be greater than or equal to y Zi .

[0083] The scheduling model established in this invention aims to enable decision-makers to select an optimal scheduling scheme so that X... i Obtain the accurate number of metering devices within the specified time. In the scheduling scheme φ, T(φ) is from = Y j (j=1,2,3,…,Z) Transport metering equipment to =X i The time set, A i (φ) is the distance to X i The possible number of metering devices, where ω is the number of metering devices transported to X. i The set of all scheduling schemes. Based on the above description, we can express the optimal scheduling model as follows:

[0084]

[0085] Among them, s ji (0≤s ji ≤1) indicates from Y j To X i Safety, φ is the scheduling scheme, ω is the distance from the transport metering equipment to X i The set of all scheduling schemes, A i (φ) is the distance to X i The possible number of measuring devices, T(φ) is from Y j (j=1,2,3,…,Z) Transport metering equipment to X i The time set, X i Belongs to X1, X2, X3, ..., X M Y j Belongs to Y1, Y2, Y3, ..., Y N .

[0086] It should be noted that the order of priority for demand ranking is based on the importance of the locations requiring metering equipment and the demand rate of metering equipment; the demand priority of metering equipment is the order of transportation operations. This priority allows us to determine which locations require a large number of meters. Based on the actual supply of metering equipment in practice, we focus on demand priority based on the importance of the locations requiring metering equipment and the demand rate of metering equipment.

[0087] The importance indicators include the urgency of transporting the metering equipment, the difficulty of supplying it, time constraints for transportation, route conditions, and weather conditions. The importance of the locations where the metering equipment is needed is expressed as follows:

[0088]

[0089] Where, α i This represents the importance index of the location where the i-th metering device is needed, k is the variable index, and δ k Let ε be the weighting coefficient of the k-th indicator. k This represents the original evaluation value of the k-th indicator;

[0090] The demand rate for metering equipment is expressed as follows:

[0091]

[0092] Where, β i Location X i The demand rate for metering equipment, x i It is X i The demand for meters, where M is the total number of warehouses storing metering equipment.

[0093] Based on the principle that importance index values ​​are directly proportional to priority and demand rates are inversely proportional to priority, the importance index values ​​and the demand rate of metering equipment are balanced by weighted dual-objective proximity.

[0094] Generally, the more important the location where metering equipment is needed, the more priority we should give to transporting meters there during the supply process. Similarly, the lower the demand rate at a location, the more priority we should give to meeting that metering equipment demand during transportation. From these two perspectives, the priority of metering equipment demand is positively correlated with the importance of the demand location and negatively correlated with the demand rate of the metering equipment; therefore, the prioritization process is a dual-objective selection process. To quantify priority, many researchers often convert the influencing factors into optimal and worst-case values. According to the above analysis, the higher the importance index value, the higher the priority; therefore, α... i The maximum value of is its optimal value, and the minimum value is its worst value. However, the smaller the demand rate, the higher its priority; therefore, the minimum value is its optimal value, and the maximum value is its worst value. If α i1 α i2 These are the maximum and minimum values ​​of the demand location importance index, respectively, while β i1 ,β i2 These are the minimum and maximum values ​​of the demand rate for metering equipment, respectively.

[0095] Will affect X i Priority importance index value and the closeness μ of the metering equipment demand rate to the optimal value i1 Perform balancing:

[0096]

[0097] Similarly, μ is obtained. i2 :

[0098]

[0099] X i Relative closeness μ to the optimal value i for:

[0100]

[0101] Where ξ1 is the weight of the importance of the demand location, and ξ2 is the weight of the demand rate of metering equipment, with ξ1 and ξ2 being 0.7 and 0.3 respectively; α i1 α i2 These are the maximum and minimum values ​​of the location importance index, respectively; β i1 β i2 These are the minimum and maximum values ​​of the demand rate for metering equipment, respectively.

[0102] It should also be noted that time constraints in the supply of metering equipment are becoming increasingly prominent. During the transportation of metering equipment, there is typically a latest arrival time T at the destination. If the demand meter cannot reach the designated location within time T, then the transportation process is meaningless.

[0103] To facilitate the analysis of the solution process of the scheduling model, if t 1i ,t 2i ,t 3i ,…,t pi ,…,t Ni From Y j (j=1,2,3,…,N) to X i The transportation time, and T i It is the transport meter to X i The latest time. To make the transportation of metering equipment meaningful, it typically requires t. ji (j = 1, 2, 3, ..., N) is less than or equal to T i .

[0104] If there are n supply locations that can meet the latest time requirement, i.e., t ji ≤T i (j = 1, 2, 3, ..., n). Then, these n supply locations are sorted in descending order to obtain a transportation queue Q. i ={(Y1,y 1i ,t 1i ,s 1i ),(Y2,y 2i ,t 2i ,s 2i ),(Y3,y 3i ,t 3i ,s 3i ),...,(Y n ,y ni ,t ni ,s ni )} used for transporting meters to X i .

[0105] In the queue, if it appears A scheduling scheme φ can be obtained. 1i Based on the above principles, we can also obtain the scheduling scheme with the shortest execution time φ. 2i We can also obtain other scheduling schemes. The optimized values ​​for all scheduling schemes are as follows:

[0106]

[0107] Where ζ1 and ζ2 are two weighting indicators, with weights of 0.6 and 0.4 respectively; t 1i,t 2i ,t 3i ,…,t pi ,…,t Ni From Y j (j=1,2,3,…,N) to X i The transportation time, T i It is the transport meter to X i The latest time.

[0108] S2. Initialize the particle swarm and divide it into a main swarm and multiple dynamic subgroups. The multiple dynamic subgroups evolve independently and reorganize periodically. The main swarm updates its velocity through a dynamic jittering strategy and learns the optimal particles of the subgroups. It adopts a progressive domain constraint strategy to adaptively adjust the range of the constraint feasible domain.

[0109] Specifically, this invention learns the idea of ​​multi-subgroup strategy, combines multi-objective optimization methods and constraint handling techniques, and proposes an improved constrained multi-objective particle swarm optimization algorithm for multi-subgroup cooperation. The aim is to balance the constraint, convergence and diversity of the algorithm to solve CMOP more effectively.

[0110] The particles are initialized using the CMOPSO-IMSC algorithm. N particles are randomly selected to form the main group, and the remaining N particles are divided into multiple subgroups (where S is the number of particles in each subgroup).

[0111] For subgroups, a cyclical dynamic recombination strategy is used to iteratively update using the standard PSO update formula, randomly redistributing subgroups at a preset frequency, and providing the main group with the best subgroup particles as learning objects.

[0112] For the main group, a dynamic jittering update strategy is used to prevent it from escaping local optima. A distribution diversification enhancement strategy is designed to ensure a uniform distribution of the solution set and improve diversity. A progressive domain constraint strategy is designed to guarantee both constraint and convergence. Cooperation and information exchange between subgroups and between subgroups and the main group, as well as different update strategies, enable the algorithm to achieve a balance between diversity, convergence, and constraint, thereby improving the quality of the final solution set. The population is divided into a main group and multiple subgroups.

[0113] The main group learns by relying on the best particles in the subgroups. The velocity update formula for particle i in the main group is as follows:

[0114]

[0115]

[0116] Where ω is the inertia weight, c1 and c2 are learning factors, r1 and r2 are random numbers in the interval (0,1), d is the dimension, and subp d It is the optimal particle in all subgroups. It is the particle's own historical best.

[0117] like Figure 2 As shown, PSO with smaller population sizes generally performs better on complex problems, achieving satisfactory results with only 3 to 5 particles. Therefore, we set the size S of each subpopulation to 5, i.e., containing 5 particles. To avoid subpopulations getting trapped in local optima, we introduce a cyclic dynamic reconstruction strategy. This strategy dynamically changes the neighborhood structure of the subpopulations. After every R iterations, all particles from all subpopulations are gathered together and randomly recombined to form new subpopulations for a new round of search. It is important to note that if R is too small, frequent recombinations may lead to insufficient subpopulation development capabilities and affect stability, reducing convergence speed. If R is too large, the global exploration capability of the subpopulations is insufficient, making it prone to getting trapped in local optima. Therefore, considering the balance between the algorithm's exploration and development capabilities, stability, and convergence speed, we experimentally verified that we set the maximum number of iterations to 200,000 and R to 2,000. Through this strategy, different subpopulations can exchange information, improving diversity while ensuring convergence. Figure 2 The process of cyclic dynamic reconstruction is demonstrated. It can be observed that the initial population is divided into 3 subgroups, each containing 3 particles; after R iterations, the subgroups are shuffled and reconstructed, the particles in the 3 subgroups are redistributed, and the topology of the subgroups is changed accordingly.

[0118] It should be noted that, to escape local optima, a dynamic jittering update strategy is proposed for the main swarm particles to increase the algorithm's diversity and exploration capability. A random jittering term is introduced into the velocity update formula. Through this random jittering term, the particle velocity update is subject to random disturbances, thus limiting the jittering range.

[0119] randfactor=unifrnd(-0.1,0.1,N,D)

[0120] Update speed based on the limited jitter range:

[0121]

[0122] To address constraints, directly employing a feasibility-driven constraint handling strategy often leads to algorithms overemphasizing feasible solutions, failing to balance convergence, diversity, and feasibility optimization. To resolve this issue, we propose a progressively narrowed domain strategy. This strategy initially sets a large search range, allowing particles to search within a relatively broad space, increasing the likelihood of finding the global optimum. As the number of iterations increases, the feasible domain range is gradually reduced. This allows particles to gradually focus on solutions that better meet the constraints, developing superior solutions within the feasible domain and improving search accuracy.

[0123]

[0124] According to the formula above, in the early stages of the algorithm's search, the value of ε is close to 1, meaning that almost all solutions in the target space are ε-feasible solutions. Based on this, the algorithm can successfully explore the entire target space without being constrained. As the number of iterations increases, the value of ε gradually decreases, and the focus of the algorithm's search gradually shifts to the feasible region. In the later stages of the algorithm, the value of ε approaches 0, meaning that the ε-feasible region gradually becomes equivalent to the true feasible region. The progressive constraint strategy achieves a balance between exploring the feasible and infeasible regions by adaptively adjusting the size of the ε-feasible region during the iteration process.

[0125] The IGD test results of CMOPSO-IMSC and other algorithms are as follows: Figure 4 As shown.

[0126] S3. Select supply points that meet the latest arrival time limit, generate transportation queues in descending order of time urgency, and output a scheduling scheme through multi-subgroup collaborative optimization.

[0127] Each set of plans is evaluated, mainly considering the following aspects: whether the actual arrival volume at the demand points meets the demand, especially for high-priority warehouses; whether the total transportation time is the shortest; whether transportation safety is guaranteed, and how to minimize losses during transit; and whether the load at the supply points is balanced, avoiding excessive pressure on any one supply point.

[0128] An improved multi-subgroup particle swarm optimization algorithm is used for optimization. For schemes where the allocation of supply points exceeds their own available quantity or the allocation of demand points is insufficient, the algorithm automatically adjusts the process, either by reverting the excess to the critical value or by allocating unmet demand to other supply points with surplus.

[0129] The results of experimental verification of the algorithm according to the present invention and the comparison algorithm

[0130] like Figure 5 As shown, among the 14 test cases, CMOPSO-IMSC achieved the best IGD value in 6 test cases. The Wilcoxon rank-sum test results show that CMOPSO-IMSC significantly outperformed ToP, PPS, and MOEA / DD in 10, 8, and 8 test cases, respectively.

[0131] like Figure 6 As shown, among the 14 test cases, CMOPSO-IMSC achieved the best HV value in 5 of them. The Wilcoxon rank-sum test results show that CMOPSO-IMSC significantly outperformed ToP, PPS, and MOEA / DD in 10, 7, and 8 test cases, respectively.

[0132] To verify the effectiveness of the dynamic jitter update strategy, a comparative experiment was conducted on MW between the CMOPSO-IMSC algorithm and one that did not employ the dynamic jitter update strategy (denoted as CMOPSO-IMSC2). The parameter settings of the two comparative algorithms were the same. Figure 7 The results of IGD tests for the CMOPSO-IMSC and CMOPSO-IMSC2 algorithms are shown. CMOPSO-IMSC achieved the best IGD values ​​in all 14 test cases. The Wilcoxon rank-sum test results show that CMOPSO-IMSC significantly outperforms CMOPSO-IMSC2 in 13 test cases.

[0133] Two important parameters exist in CMOPSO-IMSC: the subgroup size S and the jitter term range RF. Therefore, parameter analysis is performed on the MW test function. For S, we consider three values: 3, 4, and 5. The IGD comparison results of CMOPSO-IMSC based on different S values ​​are shown in Table 4. In 14 test cases, when S=5, CMOPSO-IMSC achieved 9 optimal IGD values; when S=4, it achieved 5 optimal IGD values; and when S=3, CMOPSO-IMSC could not find a solution set in 6 test cases. From the Wilcoxon rank-sum test results, when S=5, CMOPSO-IMSC significantly outperforms the algorithm when S=3 in 7 test cases; its performance is similar to that when S=4.

[0134] Example 2, refer to Figure 8 This is one embodiment of the present invention, which provides a method for improving the efficiency of short-term high-frequency energy storage. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through experiments.

[0135] ① Case Analysis and Model Implementation

[0136] At a certain moment, the metering equipment transportation department received an urgent task. The specifics of this task are as follows: There are eight locations in a certain area, namely X1, X2, X3, ..., X8. The required quantities of materials for these eight locations are 39, 50, 46, 73, 66, 57, 85, and 59 respectively, with transportation time limits of 80, 75, 85, 90, 70, 75, 85, and 90 seconds respectively. During this period, there are four metering equipment warehouses, namely Y1, Y2, Y3, and Y4. These warehouses can transport the required materials to the above eight locations, with storage capacities of 112, 146, 107, and 117 respectively. Simultaneously, the importance assessment data for the eight metering equipment demand locations is also provided. The transportation time and security from the four metering equipment warehouses to the eight demand locations are shown in Table 1.

[0137] Table 1. Data for Judging Importance

[0138] <![CDATA[δ1,δ2,δ3,δ4,δ5]]> <![CDATA[ε1,ε2,ε3,ε4,ε5 <!-- 10 -->]]> <![CDATA[X1]]> 0.6，0.7，0.7，0.6，0.5 0.28，0.19，0.16，0.23，0.14 <![CDATA[X2]]> 0.7，0.8，0.8，0.5，0.7 0.28，0.19，0.16，0.23，0.14 <![CDATA[X3]]> 0.6，0.9，0.8，0.6，0.7 0.28，0.19，0.16，0.23，0.14 <![CDATA[X4]]> 0.8，0.6，0.7，0.4，0.6 0.28，0.19，0.16，0.23，0.14 <![CDATA[X5]]> 0.9，0.7，0.6，0.5，0.8 0.28，0.19，0.16，0.23，0.14 <![CDATA[X6]]> 0.5，0.9，0.8，0.7，0.6 0.28，0.19，0.16，0.23，0.14 <![CDATA[X7]]> 0.7，0.6，0.7，0.7，0.8 0.28，0.19，0.16，0.23，0.14 <![CDATA[X8]]> 0.8，0.5，0.7，0.6，0.5 0.28，0.19，0.16，0.23，0.14

[0139] Table 2 Transportation Time

[0140]

[0141]

[0142] Based on the data regarding transportation time and safety in Table 2, we can obtain numerous scheduling schemes for the eight material demand locations. Following the principles of traditional scheduling models, a scheduling scheme can be obtained for each demand location. Simultaneously, according to the proposed scheduling model, scheduling schemes can also be obtained for these locations. The solution process for these scheduling schemes is shown in Table 2. In this table, we comprehensively compare the optimized values ​​of transportation time and scheduling schemes, as well as the quantity of materials that can potentially safely reach the demand locations.

[0143] from Figure 8 It can be seen that the traditional scheduling model differs from the proposed model in the solution process. The traditional model tends to consider transportation time or safety more. However, the proposed model focuses more on the optimized value, which needs to consider not only time constraints but also the number of metering devices that can safely reach the demand locations. Meanwhile, we can obtain the number of meters transported from four warehouses to eight demand locations through the solution process in Table 3, as shown in Table 3. The final population of CMOPSO-IMSC based on different RFs on MW9 is as follows... Figure 3 As shown.

[0144] Table 3 Number of scheduling metering devices

[0145]

[0146]

[0147] ②This example uses the 2024 spring metering equipment replacement project of a provincial power company as a case study. The company needed to complete the replenishment and scheduling of metering equipment for eight prefecture-level warehouses across the province within 30 days. The company has four large metering equipment production bases located in cities A, B, C, and D. The prefecture-level warehouses are designated W1, W2, W3, W4, W5, W6, W7, and W8. The scheduling objective is to meet the actual needs of each warehouse while considering constraints such as transportation timeliness, safety, and load balancing at supply points, achieving intelligent and optimal scheduling.

[0148] 1. Demand points (warehouses) and their demand quantities: W1: 1200 units; W2: 900 units; W3: 1800 units; W4: 700 units; W5: 1500 units; W6: 1100 units; W7: 1600 units; W8: 1300 units.

[0149] 2. Supply points and their available quantities: A: 2500 units; B: 2000 units; C: 3000 units; D: 2500 units.

[0150] 3. Transportation time and transportation safety: Taking supply point A as an example, the transportation time and safety from A to various demand points are as follows:

[0151] A to W1: 4 hours, safety 0.98; A to W2: 6 hours, safety 0.96;

[0152] A to W3: 8 hours, safety 0.95; A to W4: 10 hours, safety 0.93;

[0153] A to W5: 12 hours, safety 0.92; A to W6: 14 hours, safety 0.90;

[0154] A to W7: 16 hours, safety 0.89; A to W8: 18 hours, safety 0.88;

[0155] The transportation time and safety data from B, C, and D to each warehouse are similar to those for A, with specific values ​​determined based on actual geographical location and road conditions.

[0156] 4. Importance indicators of demand points

[0157] The importance of scheduling at each demand point is determined by five factors: scheduling urgency, supply difficulty, time constraints, route conditions, and weather impact. For example, W1, due to its role in handling large-scale urban garment replacement, has the highest scheduling urgency, significant supply difficulty, strict time constraints, relatively good road conditions, and minimal weather impact, thus making it the most important. W4, located in a remote area, has the lowest importance due to its small demand. The importance of other warehouses falls between these two extremes.

[0158] 5. Demand rate

[0159] The demand rate for each warehouse is the proportion of its demand to the total demand. For example, W3 has the largest demand, accounting for about 18% of the total demand, while W4 has the smallest demand, accounting for about 7%.

[0160] In this embodiment, the scheduling model automatically calculates the priority based on the overall importance and demand rate of each demand point. Warehouses with higher priority will receive scheduling resources first, while warehouses with lower priority will be appropriately delayed when resources are scarce.

[0161] Based on the above data, the priority order is: W1, W3, W5, W7, W8, W6, W2, W4. That is, W1 has the highest priority and W4 has the lowest priority.

[0162] Implementation process of intelligent scheduling algorithm

[0163] 1. Initialize parameters: The system first enters the basic data of all demand and supply points, including demand, availability, transportation time, and security.

[0164] 2. Generate initial scheduling plans. The system automatically generates several initial scheduling plans, each of which specifies the number of devices to be allocated from each supply point to each demand point. Each plan must meet the basic constraints that supply points do not overproduce and demand points do not over-demand.

[0165] 3. Fitness assessment: The system evaluates each set of solutions, mainly considering the following aspects:

[0166] Does the actual arrival volume at the demand points meet the demand, especially for high-priority warehouses? Is the total transportation time the shortest possible? Is transportation safety guaranteed, and losses during transit should be minimized? Is the load on the supply points balanced to avoid excessive pressure on any particular supply point?

[0167] 4. Intelligent optimization process

[0168] The system employs an improved multi-subgroup particle swarm optimization algorithm. The algorithm divides all solutions into several subgroups, each optimizing independently, and periodically shuffles and regroups its members to prevent getting trapped in local optima. The main group then selects the optimal solution from all subgroups for further optimization. Throughout the optimization process, the system dynamically adjusts the search range, initially exploring a large range and gradually converging to the optimal feasible solution later.

[0169] 5. Constraint Handling

[0170] For schemes where supply points are allocated more than their own available capacity or demand points are allocated insufficiently, the system will automatically adjust, either by returning the excess to the critical value or by allocating the unmet demand to other supply points with surplus capacity.

[0171] Scheduling Results and Analysis: After intelligent optimization, the system finally outputs a set of optimal scheduling schemes. Some examples of the scheduling results are shown in Table 4:

[0172] Table 4 shows some scheduling results.

[0173]

[0174] Of the 1200 units of equipment in W1, 900 were supplied by supplier A and 300 by supplier C, ensuring sufficient supply to the highest priority demand point. All 1800 units of equipment in W3 were supplied by supplier C, due to the favorable transportation conditions and high security between C and W3. Equipment in W5 and W8 were supplied by a combination of suppliers A, B, C, and D, respectively, ensuring timely delivery to the demand points. The low-priority equipment in W4 was mainly supplied by supplier B, meeting its lower demand. The actual shipments from each supplier did not exceed its maximum available quantity.

[0175] Regarding transportation safety, the system prioritizes transportation routes with high safety coefficients. For example, the safety coefficient from supply point A to W1 is 0.98, and the safety coefficient from supply point C to W3 is 0.93, minimizing equipment damage during transportation. Actual deliveries almost entirely meet the requirements of the demand points, with high-priority warehouses achieving delivery rates exceeding 95%, and even lower-priority warehouses achieving delivery rates exceeding 90%.

[0176] In terms of transportation timeliness, the system automatically avoids long-distance, inefficient transportation. For example, supply point A supplies warehouses W1, W5, and W8 that are relatively close, while supply point C undertakes tasks with high demand and better transportation conditions, such as W3, W6, and W7, thereby effectively shortening the overall transportation time.

[0177] In terms of load balancing, the system rationally allocated the workload to each supply point, avoiding overloading of individual supply points. The four supply points A, B, C, and D evenly distributed the scheduling tasks across the province, maximizing overall capacity.

[0178] This embodiment, based on a real-world power metering equipment scheduling scenario, demonstrates in detail how the multi-constraint intelligent scheduling model and method of this invention can be applied in practical work. By rationally setting various parameters such as demand, supply, transportation, and safety, and combining innovative intelligent optimization algorithms, it achieves efficient, intelligent, and controllable solutions to complex scheduling problems. This method is not only applicable to the power industry but can also be extended to other large-scale material production and supply chain scheduling fields, possessing broad application prospects and significant economic and social benefits.

[0179] Example 3 is an embodiment of the present invention, which provides a multi-constraint intelligent scheduling system for power metering equipment, comprising a model building module, a multi-subgroup collaborative optimization module, and a scheduling module;

[0180] The model building module establishes an intelligent scheduling model for the demand points, supply points, transportation time, and safety of metering equipment. Based on the importance index of the demand points and the demand rate of metering equipment, the demand priority is calculated through weighted dual-objective proximity.

[0181] The multi-subgroup collaborative optimization module initializes the particle swarm and divides it into a main group and multiple dynamic subgroups. The multiple dynamic subgroups evolve independently and reorganize periodically. The main group updates its speed through a dynamic jittering strategy and learns the optimal particles of the subgroups. It adopts a progressive domain constraint strategy to adaptively adjust the range of the constraint feasible domain.

[0182] The scheduling module filters supply points that meet the latest arrival time limit, generates transportation queues in descending order of time urgency, and outputs a scheduling scheme through multi-subgroup collaborative optimization.

[0183] This embodiment also provides an electronic device applicable to a multi-constraint intelligent scheduling method for power metering equipment, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the multi-constraint intelligent scheduling method for power metering equipment as proposed in the above embodiment.

[0184] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, it implements a multi-constraint intelligent scheduling method for power metering equipment as proposed in the above embodiment.

[0185] The storage medium proposed in this embodiment and the method for implementing a multi-constraint intelligent scheduling of power metering equipment proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0186] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0187] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A multi-constraint intelligent scheduling method for power metering equipment, characterized in that: include, Establish an intelligent scheduling model for the demand points, supply points, transportation time, and safety of metering equipment. Based on the importance index of demand points and the demand rate of metering equipment, calculate the demand priority through weighted dual-objective proximity. The particle swarm is initialized and divided into a main swarm and multiple dynamic subgroups. The multiple dynamic subgroups evolve independently and reorganize periodically. The main swarm updates its velocity and learns the optimal particles of the subgroups through a dynamic jittering strategy. An incremental domain constraint strategy is adopted to adaptively adjust the range of the constraint feasible domain. Supply points that meet the latest arrival time limit are selected, and transportation queues are generated in descending order of time urgency. The output scheduling scheme is optimized through multi-subgroup collaboration.

2. The multi-constraint intelligent scheduling method for power metering equipment as described in claim 1, characterized in that: The intelligent scheduling model includes classifying the warehouses for metering equipment into X1, X2, X3, ..., X... M The number of measuring devices needed for the warehouse is x. i (i=1,2,3,…,M),Y1,Y2,Y3,…,Y N The location from which metering equipment is supplied is given, with a supply quantity of y. j (j = 1, 2, 3, ..., N), y ij Represents Y j To X i The quantity of metering equipment supplied; Make X i The optimal scheduling model for obtaining an accurate number of metering devices within a specified time is expressed as follows: Among them, s ji (0≤s ji ≤1) indicates from Y j To X i Safety, φ is the scheduling scheme, ω is the distance from the transport metering equipment to X i The set of all scheduling schemes, A i (φ) is the distance to X i The possible number of measuring devices, T(φ) is from Y j (j=1,2,3,…,Z) Transport metering equipment to X i The time set, X i Belongs to X1, X2, X3, ..., X M Y j Belongs to Y1, Y2, Y3, ..., Y N .

3. The multi-constraint intelligent scheduling method for power metering equipment as described in claim 2, characterized in that: The calculation of demand priority includes the order of prioritizing demand based on the importance indicators of the location of metering equipment demand and the demand rate of metering equipment; The importance indicators include the urgency of transporting the metering equipment, the difficulty of supplying it, time constraints for transportation, route conditions, and weather conditions. The importance of the locations where the metering equipment is needed is expressed as follows: Where, α i This represents the importance index of the location where the i-th metering device is needed, k is the variable index, and δ k ε is the weighting coefficient of the k-th indicator. k This represents the original evaluation value of the k-th indicator; The demand rate for metering equipment is expressed as follows: Where, β i Location X i The demand rate for metering equipment, x i It is X i The demand for meters, where M is the total number of warehouses storing metering equipment.

4. The multi-constraint intelligent scheduling method for power metering equipment as described in claim 3, characterized in that: The calculation of demand priority also includes balancing the importance index value and the metering equipment demand rate by weighting the proximity of the two objectives, based on the fact that the importance index value is directly proportional to the priority and the demand rate is inversely proportional to the priority. Will affect X i Priority importance index value and the closeness μ of the metering equipment demand rate to the optimal value i1 Perform balancing: Similarly, μ is obtained. i2 : X i Relative closeness μ to the optimal value i for: Where ξ1 is the weight of the importance of the demand location, and ξ2 is the weight of the demand rate of metering equipment, with ξ1 and ξ2 being 0.7 and 0.3 respectively; α i1 α i2 These are the maximum and minimum values ​​of the importance index for the demand location, respectively; β i1 β i2 These are the minimum and maximum values ​​of the demand rate for metering equipment, respectively.

5. The multi-constraint intelligent scheduling method for power metering equipment as described in claim 4, characterized in that: The intelligent scheduling model includes calculating X, taking into account the time constraints of metering equipment supply. i The optimized values ​​of all scheduling schemes are used to construct the scheduling scheme with the shortest execution time, as follows: Where ζ1 and ζ2 are two weighting indicators, with weights of 0.6 and 0.4 respectively; t 1i ,t 2i ,t 3i ,…,t pi ,…,t Ni From Y j (j=1,2,3,…,N) to X i The transportation time, T i It is the transport meter to X i The latest time.

6. The multi-constraint intelligent scheduling method for power metering equipment as described in claim 5, characterized in that: The initialization of the particle swarm and the division into a main swarm and multiple dynamic sub-swarms includes improving the constrained multi-objective particle swarm optimization method for multi-sub-swarm cooperation through multi-objective optimization methods and constraint handling techniques. The particles are initialized using the CMOPSO-IMSC algorithm. N particles are randomly selected to form the main group, and the remaining N particles are divided into multiple subgroups (where S is the number of particles in each subgroup). For subgroups, a cyclical dynamic recombination strategy is used to iteratively update using the standard PSO update formula, randomly redistributing subgroups at a preset frequency, and providing the main group with the best subgroup particles as learning objects. For the main group, a dynamic dithering update strategy is used to help it escape local optima. The main group learns by relying on the best particles in the subgroups. The velocity update formula for particle i in the main group is as follows: Where ω is the inertia weight, c1 and c2 are learning factors, r1 and r2 are random numbers in the interval (0,1), d is the dimension, and subp d It is the optimal particle in all subgroups. It is the particle's own historical best.

7. The multi-constraint intelligent scheduling method for power metering equipment as described in claim 6, characterized in that: The dynamic jitter update strategy includes introducing a random jitter term into the velocity update formula. Through the random jitter term, the particle velocity update is subject to random disturbance, thus limiting the jitter range. randfactor=unifrnd(-0.1,0.1,N,D) Update speed based on the limited jitter range: The proposed method employs a progressive confinement strategy to adaptively adjust the feasible region range. This strategy sets an initial search range in the early stages of optimization, allowing particles to search within this range and increasing the likelihood of finding the global optimum. As the number of iterations increases, the value of ε further decreases, resulting in the feasible region range of: The value of ε is approximately equal to 0, meaning that the feasible region of ε gradually becomes equivalent to the real feasible region.

8. A multi-constraint intelligent dispatching system for power metering equipment, employing the multi-constraint intelligent dispatching method for power metering equipment as described in any one of claims 1 to 7, characterized in that, include: Model building module, multi-subgroup collaborative optimization module, scheduling module; The model building module establishes an intelligent scheduling model for the demand points, supply points, transportation time, and safety of metering equipment. Based on the importance index of the demand points and the demand rate of metering equipment, the demand priority is calculated through weighted dual-objective proximity. The multi-subgroup collaborative optimization module initializes the particle swarm and divides it into a main group and multiple dynamic subgroups. The multiple dynamic subgroups evolve independently and reorganize periodically. The main group updates its speed through a dynamic jittering strategy and learns the optimal particles of the subgroups. It adopts a progressive domain constraint strategy to adaptively adjust the range of the constraint feasible domain. The scheduling module filters supply points that meet the latest arrival time limit, generates transportation queues in descending order of time urgency, and outputs a scheduling scheme through multi-subgroup collaborative optimization.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the multi-constraint intelligent scheduling method for power metering equipment as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the multi-constraint intelligent scheduling method for power metering equipment as described in any one of claims 1 to 7.

Images