Multi-source load management system and method based on dynamic regulation and control

Abstract

The invention discloses a multi-source load management system based on dynamic regulation and control, and the system comprises a space-time clustering feature calculation module which employs a multivariable recombination formula to calculate a regulation and control task recombination value, combines a regulation and control region, a time period, and a load space-time position, generates space-time features of each load through a space-time clustering feature formula, and stores the space-time features of each load; quantifying feature significance based on a distribution probability formula, and generating a feature sequence according to a probability descending order; the load regulation and control task matching module is used for dividing loads into a plurality of sets through feature distance calculation, calculating comprehensive features of the sets, calculating matching values of tasks and the sets by adopting a multi-dimensional matching formula, and selecting an optimal matching set; and the load regulation and control task execution module executes a regulation and control task on the load in the matching set according to the priority of the feature sequence, ensures that the high-probability feature load responds preferentially, and supports a dynamic adjustment strategy. According to the invention, stable and efficient operation of a power system can be guaranteed, the intelligent management level is improved, and fine management of complex multi-source loads is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of load management, and particularly to a multi-source load management system and method based on dynamic regulation and control. Background Art

[0002] In the operation and management of modern power systems, with the diversification of energy structures and the increasing complexity of electricity demand, multi-source load management faces many severe challenges, specifically manifested as follows: A large number of different types of loads are connected to the power system, such as the complex and variable electrical equipment in industrial production processes, the electrical facilities with obvious peak-valley characteristics in commercial premises, and various types of scattered and diverse household appliances in residents' lives, etc.; These loads have significant differences in power consumption time, power demand, and adjustability. The traditional unified management mode is difficult to effectively coordinate their operations and cannot fully exploit their adjustment potential to meet the dynamic demands of the system; The power system needs to simultaneously consider multiple important dynamic regulation and control objectives. For example, in the context of tight energy supply and increasing environmental pressure, it is necessary to achieve peak shaving and valley filling to ensure the stable operation of the power grid and improve energy utilization efficiency, while reducing the electricity cost of users and ensuring the power supply reliability of important loads; However, existing management methods often only focus on single or a few objectives and are difficult to achieve a good balance among multiple objectives, resulting in the overall performance of the system not reaching the optimal; With the large-scale access of new energy power generation, its randomness and volatility further exacerbate the imbalance between power supply and demand, making the difficulty of multi-source load management increase significantly. Summary of the Invention

[0003] The object of the present invention is to address the deficiencies of the prior art and provide a multi-source load management system based on dynamic regulation and control, including:

[0004] A spatio-temporal clustering feature calculation module, configured to use a preset multi-variable recombination formula to recombine a preset load regulation task to obtain a recombination value of the preset load regulation task, calculate the spatio-temporal clustering features of each current load based on the recombination value, the regulation area in the preset load regulation task, the regulation time period in the preset load regulation task, the time period where each current load is located, and the area where each current load is located by using a preset spatio-temporal clustering feature formula, calculate the distribution probability of the spatio-temporal clustering features of each current load in the set of spatio-temporal clustering features of all current loads by using a preset spatio-temporal clustering feature distribution probability formula, and arrange the spatio-temporal clustering features of each current load in descending order according to the corresponding distribution probability to obtain a spatio-temporal clustering feature sequence;

[0005] The load regulation task matching module is used to calculate the feature distances between the spatio-temporal clustering features of current loads, classify all current loads into several current load sets according to a plurality of preset feature distance intervals, calculate the comprehensive features of each current load set by using a preset comprehensive feature formula of the current load set, calculate the matching values between the preset load regulation tasks and each current load set by using a preset multi-dimensional matching formula according to the comprehensive features of each current load set, and take the current load set with the largest matching value as the matching result;

[0006] The load regulation task execution module is used to execute the preset load regulation tasks for each load in the current load set as the matching result in the order of the corresponding spatio-temporal clustering features in the spatio-temporal clustering feature sequence.

[0007] Further, in the spatio-temporal clustering feature calculation module, the specific method for reorganizing the preset load regulation task by using a preset multi-variable reorganization formula to obtain the reorganization value of the preset load regulation task is as follows:

[0008]

[0009] where G i is the load value regulation target in the preset load regulation task i, T i is the regulation time period in the preset load regulation task i, A i is the regulation area in the preset load regulation task i, P i is the load type regulation priority in the preset load regulation task i, δ1 to δ4 are the weights corresponding to the load value regulation target, regulation time period, regulation area and load type regulation priority, and R i is the reorganization value of the preset load regulation task i.

[0010] Further, in the spatio-temporal clustering feature calculation module, the specific method for calculating the spatio-temporal clustering features of current loads by using a preset spatio-temporal clustering feature formula based on the reorganization value, the regulation area in the preset load regulation task, the regulation time period in the preset load regulation task, the time periods where current loads are located, and the areas where current loads are located is as follows:

[0011]

[0012] where b1 is the influence factor of the regulation time period obtained by regression analysis, b2 is the influence factor of the regulation area obtained by regression analysis, e is the natural logarithm base, t is the time period where the current load m is located, a is the area where the current load m is located, S t,a is the spatio-temporal clustering feature of the time period t where the current load m is located and the area a where the current load m is located, and the spatio-temporal clustering feature set S = {S t,a}.

[0013] Further, in the spatio-temporal clustering feature calculation module, the specific method for calculating the distribution probability of the spatio-temporal clustering features of each current load in the set of spatio-temporal clustering features of all current loads using a preset spatio-temporal clustering feature distribution probability formula is as follows:

[0014]

[0015] Among them, S j is the j-th interval among n equally long intervals into which the spatio-temporal clustering feature set S = {S t,a} is divided, N j is the number of spatio-temporal clustering features falling within the interval [S j , S j+1 ), ∑ n N n is the total number of empty clustering features falling within the n intervals, is the distribution probability corresponding to the spatio-temporal clustering features falling within the j-th interval. The spatio-temporal clustering features in the spatio-temporal clustering feature set S = {S t,a} are sorted in descending order according to the distribution probability of the spatio-temporal clustering features to obtain a spatio-temporal clustering feature sequence.

[0016] Further, in the load regulation task matching module, the specific method for calculating the feature distance between the spatio-temporal clustering features of each current load is as follows:

[0017] The calculation methods of the feature distance include Euclidean distance, Manhattan distance, Chebyshev distance, and Minkowski distance.

[0018] Further, in the load regulation task matching module, the specific method for classifying all current loads into several current load sets according to a preset number of feature distance intervals is as follows:

[0019] The preset number of feature distance intervals is determined based on the value range of the feature distances between the spatio-temporal clustering features of historical loads.

[0020] Further, in the load regulation task matching module, the specific method for calculating the comprehensive feature of each current load set using a preset current load set comprehensive feature formula is as follows:

[0021]

[0022] Among them: I t,a (C k ) is an indicator function. If there are loads related to the preset load regulation task in the current time period t when the current load set C k is located and the current area a where the current load m is located, I t,a (C k ) = 1; otherwise, I t,a(C k ) = 0; ∑ t,a S t,a is the sum of the spatio-temporal clustering characteristics of the current load m at the current time period t and in the current area a where the load m is located; is the vector after spatio-temporal clustering feature mapping of the time period quantization value and the area quantization value of the current load m in the current load set C k , k is the number of the current load m in the current load set C k , is the comprehensive eigenvalue of the current load set C k .

[0023] Furthermore, in the load regulation task execution module, the specific method for calculating the matching value between the preset load regulation task and each current load set by using a preset multi-dimensional matching formula according to the comprehensive characteristics of each current load set is as follows:

[0024]

[0025] where G i is the load value regulation target of the preset load regulation task i, ΔT is the difference between the regulation time period in the preset load regulation task i and the time period quantization value of the current load m, ΔA is the difference between the regulation area in the preset load regulation task i and the area quantization value of the current load m, P i is the load type regulation priority in the preset load regulation task i, max[ ] represents taking the maximum value, M(C k , i) is the calculated matching value between the preset load regulation task i and the current load set C k .

[0026] A multi-source load management method based on dynamic regulation includes:

[0027] Using a preset multi-variable recombination formula to recombine the preset load regulation task to obtain the recombination value of the preset load regulation task, and calculating the spatio-temporal clustering characteristics of each current load based on the recombination value, the regulation area in the preset load regulation task, the regulation time period in the preset load regulation task, the current time period of each current load, and the current area where each current load is located by using a preset spatio-temporal clustering feature formula, and calculating the distribution probability of the spatio-temporal clustering characteristics of each current load in the set of spatio-temporal clustering characteristics of all current loads by using a preset spatio-temporal clustering feature distribution probability formula, and arranging the spatio-temporal clustering characteristics of each current load in descending order according to the corresponding distribution probability to obtain the spatio-temporal clustering feature sequence;

[0028] Calculate the feature distance between the spatio-temporal clustering characteristics of each current load, classify all current loads into several current load sets according to a preset number of feature distance intervals, calculate the comprehensive characteristics of each current load set using a preset comprehensive characteristic formula for the current load set, calculate the matching value between the preset load regulation task and each current load set using a preset multi-dimensional matching formula according to the comprehensive characteristics of each current load set, and take the current load set with the largest matching value as the matching result;

[0029] For each load in the current load set that is the matching result, execute the preset load regulation task in sequence according to the order of the corresponding spatio-temporal clustering characteristics in the spatio-temporal clustering characteristic sequence.

[0030] A computer program product includes a computer program / instructions, and when the computer program / instructions are executed by a processor, the above-mentioned multi-source load management method based on dynamic regulation is implemented.

[0031] The beneficial effects of the present invention are as follows:

[0032] 1. In view of the problems such as the complex load types and the difficulty of multi-objective balance in the power system, based on the comprehensive consideration of regulation objectives, time periods, regions and priorities, spatio-temporal clustering characteristics and their sequences are generated, the load characteristics and distribution laws are accurately grasped, and then multi-source current loads are clustered to form specific current load sets, and finally the efficient matching between the preset load regulation task and the current load set is realized, so that the regulation potential of multi-source loads can be fully explored, the stable and efficient operation of the power system can be guaranteed, the intelligent management level can be improved, and the refined management of complex multi-source loads is realized.

[0033] 2. Based on the spatio-temporal clustering characteristic sequence, the preset load regulation task is executed in the matched current load set in sequence, realizing an orderly and reasonable regulation implementation. In view of the dynamic and changeable operation of the power system, the spatio-temporal clustering characteristic sequence obtained based on the spatio-temporal clustering characteristic distribution probability realizes the implementation of regulation in the matched current load set, making the regulation adapt to the system changes based on the comprehensive consideration of the spatio-temporal clustering characteristic sequence and priority, optimizing the regulation process, ensuring the stable energy supply, and strengthening the intelligent management effect of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a block diagram of the system of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0035] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0036] Embodiment 1

[0037] Reference Figure 1 , a multi-source load management system based on dynamic regulation, comprising:

[0038] A spatio-temporal clustering feature calculation module, configured to use a preset multi-variable recombination formula to recombine a preset load regulation task to obtain a recombination value of the preset load regulation task, and based on the recombination value, the regulation area in the preset load regulation task, the regulation time period in the preset load regulation task, the current time period where each load is located, and the current area where each load is located, calculate the spatio-temporal clustering features of each current load using a preset spatio-temporal clustering feature formula, and calculate the distribution probability of the spatio-temporal clustering features of each current load in the set of spatio-temporal clustering features of all current loads using a preset spatio-temporal clustering feature distribution probability formula, and sort the spatio-temporal clustering features of each current load in descending order according to the corresponding distribution probability to obtain a spatio-temporal clustering feature sequence;

[0039] A load regulation task matching module, configured to calculate the feature distance between the spatio-temporal clustering features of each current load, classify all current loads into several current load sets according to a preset number of feature distance intervals, calculate the comprehensive feature of each current load set using a preset current load set comprehensive feature formula, calculate the matching value between the preset load regulation task and each current load set using a preset multi-dimensional matching formula according to the comprehensive feature of each current load set, and use the current load set with the largest matching value as the matching result;

[0040] A load regulation task execution module, configured to execute the preset load regulation task for each load in the current load set that is used as the matching result in the order of the corresponding spatio-temporal clustering features in the spatio-temporal clustering feature sequence.

[0041] Among them, the preset load regulation task includes regulation contents such as the load value regulation target, the regulation time period, the regulation area, and the load type regulation priority, and realizes regulation requirements such as peak-valley period optimization through the preset load regulation task.

[0042] (1) In the spatio-temporal clustering feature calculation module, the preset multi-variable recombination formula is:

[0043]

[0044] Wherein, G i is the load value regulation target in the preset load regulation task i, T i is the regulation time period in the preset load regulation task i, A i is the regulation area in the preset load regulation task i, P i is the load type regulation priority in the preset load regulation task i, δ1 to δ4 are the weights corresponding to the load value regulation target, the regulation time period, the regulation area, and the load type regulation priority, Ri is the recombined value for the preset load regulation task i.

[0045] By recombining various preset load regulation tasks in the preset load regulation task, the complex and diverse preset load regulation tasks are transformed into quantifiable and comparable recombined values. In the face of the dilemma of huge differences in the electricity consumption characteristics of different types of loads and the failure of traditional management models, the regulation targets, time periods, regions, and priorities are accurately quantified. The recombined value is calculated according to the formula, laying a solid foundation for the subsequent generation of spatio-temporal clustering characteristics, enabling the preset load regulation task to be measured by a unified standard, providing a well-organized classification basis for multi-source load management, and greatly improving the scientificity and accuracy of management decisions.

[0046] (2) In the spatio-temporal clustering feature calculation module, the preset spatio-temporal clustering feature formula is:

[0047]

[0048] Among them, b1 is the influence factor of the regulation time period obtained through regression analysis, b2 is the influence factor of the regulation region obtained through regression analysis, e is the natural logarithm base, t is the time period in which the current load m is located, a is the region in which the current load m is located, and S t,a is the spatio-temporal clustering feature of the time period t in which the current load m is located and the region a in which the current load m is located. The spatio-temporal clustering feature set S = {S t,a}.

[0049] The region a in which the current load m is located is collected using the preset region coding rule, and the time period t in which the current load m is located is collected using the preset load quantity collection rule. The load quantity collection rule and the region coding rule are formulated based on the common knowledge in the art. The specific load quantity collection rule can be but is not limited to setting a load quantity threshold, and when it belongs to the load quantity threshold, the time series corresponding to the load quantity is collected. The specific region coding rule can be but is not limited to the conversion based on the preset region coding table.

[0050] By calculating the spatio-temporal clustering features of each current load, the accurate construction of the clustering features reflecting spatio-temporal characteristics is realized. Due to the high complexity of the actual multi-source load in spatio-temporal distribution, it is necessary to combine the influence factors of the regulation time period and region to calculate the spatio-temporal clustering features, so as to effectively capture the internal correlation and change trend of the load under different spatio-temporal conditions, be able to provide key guidance for multi-source load clustering, ensure that similar loads can be effectively clustered, improve management pertinence, and enable the power system to respond to dynamically changing demands.

[0051] (3) In the spatio-temporal clustering feature calculation module, the spatio-temporal clustering feature distribution probability calculation formula is:

[0052]

[0053] Among them, S j is the j-th interval among n equally long intervals into which the spatio-temporal clustering feature set S = {S t,a} is divided, N j is the number of spatio-temporal clustering features falling within the interval [S j , S j+1 ), ∑ n N n is the total number of empty clustering features falling within the n intervals, is the distribution probability corresponding to the spatio-temporal clustering features falling within the j-th interval. According to the distribution probabilities of the spatio-temporal clustering features, the spatio-temporal clustering features in the spatio-temporal clustering feature set S = {S t,a} are sorted in descending order to obtain a spatio-temporal clustering feature sequence.

[0054] By calculating the distribution probabilities of spatio-temporal clustering features, the distribution probabilities of spatio-temporal clustering features are obtained to get a sequence. Considering the diverse variations of the load in different spatio-temporal dimensions, the distribution probabilities are calculated by dividing intervals, counting the proportion of quantities, and then sorting, intuitively showing the distribution trend of spatio-temporal clustering features, providing a probability basis for the subsequent execution of the preset load regulation task, enabling the management strategy to optimize the regulation of resource allocation based on probability laws, and enhancing the operational stability of the system.

[0055] In the load regulation task matching module described in (4), the specific method for calculating the feature distance between the current spatio-temporal clustering features of each load is as follows:

[0056] The calculation methods of feature distance include Euclidean distance, Manhattan distance, Chebyshev distance, and Minkowski distance.

[0057] The specific method for classifying all current loads into several current load sets according to a preset number of feature distance intervals is as follows:

[0058] The preset number of feature distance intervals is determined based on the value range of the feature distances between the spatio-temporal clustering features of historical loads.

[0059] Based on the multi-source load feature distance calculation and the preset feature distance intervals, accurate and efficient load clustering is achieved. Aiming at the problem of diverse load electricity consumption characteristics, the distance is calculated according to the regulation time period, regional quantization value, and load own feature quantization value, and clustering is selected according to the preset feature distance intervals, promoting loads with similar electricity consumption characteristics to be grouped together, facilitating unified management and coordinated regulation, tapping the potential of group regulation, and enhancing the ability of the power system to cope with complex loads.

[0060] (5) In the load regulation task matching module, the comprehensive feature calculation formula of the current load set is:

[0061]

[0062] Where: I t,a (C k ) is an indicator function. If there is a load related to the preset load regulation task in the time period t when the current load m is located and the area a where the current load m is located, I k (C t,a ) = 1; otherwise, I k (C t,a ) = 0; ∑ k S t,a is the sum of the spatio-temporal clustering characteristics in the time period t when the current load m is located and the area a where the current load m is located; t,a is the vector after the time period quantization value of the current load m and the area quantization value of the current load m in the current load set C are mapped by the spatio-temporal clustering characteristics. k is the number of the current load m in the current load set C is the current load m in the current load set C k ; k is the comprehensive eigenvalue of the current load set C is the current load set C k .

[0063] Using the preset calculation formula for the comprehensive characteristics of the current load, the comprehensive characteristics of the current load are refined for multi-dimensional matching. Considering that the current load needs to be accurately docked with the preset load regulation task, the comprehensive characteristics are calculated by combining spatio-temporal clustering characteristics, indicator functions, etc., covering all key information of the set in all directions, creating an adaptation standard for multi-dimensional matching, making the matching between the preset load regulation task and the current load more accurate, ensuring the effective implementation of the regulation instructions, and improving the system regulation efficiency.

[0064] In the load regulation task execution module described in (6), the multi-dimensional matching formula is:

[0065]

[0066] Where, G i is the load value regulation target of the preset load regulation task i, ΔT is the difference between the regulation time period in the preset load regulation task i and the time period quantization value when the current load m is located, ΔA is the difference between the regulation area in the preset load regulation task i and the area quantization value when the current load m is located, P i is the load type regulation priority in the preset load regulation task i, max[ ] represents finding the maximum value, M(C k , i) is the matching value calculated for the preset load regulation task i and the current load set C k .

[0067] Through multi-dimensional matching, an intelligent agent set that can quickly and accurately screen out the most suitable ones for the preset load regulation tasks is realized. Under the multi-source load management problem that needs to balance multiple objectives and multi-dimensional matching, by quantitatively comparing the differences in each dimension between the preset load regulation tasks and the current load set, the matching value is calculated according to the formula and the maximum value is taken to determine the matching result, ensuring that the regulation execution is accurate and in place, avoiding mismatching and wasting resources, and improving the economy and reliability of the power system operation.

[0068] (7) The specific method for executing the preset load regulation task (such as peak shaving and valley filling) is as follows: The power grid dispatching center receives superior regulation instructions (such as peak shaving capacity, valley filling period, peak shaving and valley filling area, load regulation target curve, etc.). The power grid dispatching center clarifies the regulation target (such as reducing the peak load by XX megawatts, or transferring the load in XX period to the low valley period), determines the time window (for example, the peak shaving period is 14:00-16:00, and the valley filling period is 00:00-6:00), and clarifies the scope of the regulated area. Then, through the SCADA system, smart meters, etc., the load values, power generation outputs, equipment operation states (such as transformer load rates, line power flows) of each load in the current load set are obtained. Combining factors such as historical data and weather (such as a sharp increase in air-conditioning load due to high temperature), the load curve for the next few hours to several days is predicted. Then, through demand-side management and generation-side control, the peak load is reduced (such as remotely cutting off interruptible loads, using price incentives (such as time-of-use electricity prices, peak electricity prices) to guide users to actively reduce peak electricity consumption, reducing the output of non-essential units, and giving priority to ensuring the consumption of renewable energy), and the valley load is increased through measures such as energy storage charging, incentive electricity consumption, and adjusting the generation plan on the generation side.

[0069] The following are examples:

[0070] Peak shaving: In a certain place, the load reaches 5000 MW in the afternoon of summer, and the target is to reduce 300 MW. Steps: Cut off 50 MW of industrial interruptible load + demand response to reduce 100 MW (raise the commercial air conditioner by 2°C) + energy storage discharge 50 MW + inter-provincial mutual assistance 100 MW.

[0071] Valley filling: The load at night is only 2000 MW, and it needs to be increased to 2500 MW. Steps: Pumped storage power station charges 200 MW + aluminum smelter increases production by 300 MW (preferential low valley electricity price).

[0072] Example 2

[0073] A multi-source load management method based on dynamic regulation, including:

[0074] Recombine the preset load regulation task using a preset multi-variable recombination formula to obtain the recombination value of the preset load regulation task. Based on the recombination value, the regulated area in the preset load regulation task, the regulated time period in the preset load regulation task, the current time period where each load is located, and the current area where each load is located, calculate the spatio-temporal clustering characteristics of each current load using a preset spatio-temporal clustering feature formula, and calculate the distribution probability of the spatio-temporal clustering characteristics of each current load in the set of spatio-temporal clustering characteristics of all current loads using a preset spatio-temporal clustering feature distribution probability formula. Arrange the spatio-temporal clustering characteristics of each current load in descending order according to the corresponding distribution probability to obtain a spatio-temporal clustering feature sequence;

[0075] Calculate the feature distance between the spatio-temporal clustering characteristics of each current load. Classify all current loads into several current load sets according to a preset number of feature distance intervals. Calculate the comprehensive feature of each current load set using a preset current load set comprehensive feature formula. Calculate the matching value between the preset load regulation task and each current load set using a preset multi-dimensional matching formula according to the comprehensive feature of each current load set. Take the current load set with the largest matching value as the matching result;

[0076] For each load in the current load set that is the matching result, sequentially execute the preset load regulation task in the order of the corresponding spatio-temporal clustering characteristics in the spatio-temporal clustering feature sequence.

[0077] Embodiment 3

[0078] A computer program product includes computer programs / instructions which, when executed by a processor, implement the multi-source load management method based on dynamic regulation in Embodiment 2.

[0079] The content not detailed in this specification belongs to the prior art well-known to those skilled in the art. Those skilled in the art should understand that the embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can be in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0080] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and combinations of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable data processing devices to produce a machine, such that the instructions executed by the processors of the computer or other programmable data processing devices produce means for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or means for implementing the functions specified in multiple blocks.

[0081] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including instruction means that implement the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or means for implementing the functions specified in multiple blocks.

[0082] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or means for implementing the functions specified in multiple blocks.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit the scope of its protection. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that after reading the present invention, various changes, modifications, or equivalent replacements can still be made to the specific implementation manners of the invention, but these changes, modifications, or equivalent replacements are all within the scope of the claims of the invention pending approval.

Claims

1. A multi-source load management system based on dynamic regulation, characterized in that: include: A spatiotemporal clustering feature calculation module, which is used to use a preset multivariate reorganization formula to reorganize the preset load control task to obtain a reorganization value of the preset load control task, and to use a preset spatiotemporal clustering feature formula to calculate the spatiotemporal clustering features of each current load based on the reorganization value, the control area in the preset load control task, the control period in the preset load control task, the period in which each load is currently located, and the area in which each load is currently located, and to use a preset spatiotemporal clustering feature distribution probability formula to calculate the distribution probability of the spatiotemporal clustering features of each current load in the set of spatiotemporal clustering features of all current loads, and to arrange the spatiotemporal clustering features of each current load in descending order according to the corresponding distribution probability to obtain a spatiotemporal clustering feature sequence; The load control task matching module is used to calculate the characteristic distance between the spatiotemporal clustering characteristics of each current load, classify all current loads into several current load sets according to several preset characteristic distance intervals, calculate the comprehensive characteristics of each current load set using a preset current load set comprehensive characteristic formula, calculate the matching value between the preset load control task and each current load set using a preset multi-dimensional matching formula based on the comprehensive characteristics of each current load set, and take the current load set with the largest matching value as the matching result; The load regulation task execution module is used to execute the preset load regulation task in sequence for each load in the current load set as the matching result according to the order of the corresponding spatiotemporal clustering features in the spatiotemporal clustering feature sequence.

2. The multi-source load management system based on dynamic regulation according to claim 1 is characterized in that: In the spatiotemporal clustering feature calculation module, the specific method of using the preset multivariate reorganization formula to reorganize the preset load control task to obtain the reorganized value of the preset load control task is: The preset load control tasks include load value control target, control period, control area and load type control priority. i is the load value control target in the preset load control task i, T i is the control period in the preset load control task i, A i is the control area in the preset load control task i, P i is the load type control priority in the preset load control task i, δ1~δ4 are the weights of the corresponding load value control target, control period, control area and load type control priority, R i is the reorganization value of the preset load regulation task i.

3. The multi-source load management system based on dynamic regulation according to claim 2 is characterized in that: In the spatiotemporal clustering feature calculation module, the specific method of calculating the spatiotemporal clustering features of each current load using the preset spatiotemporal clustering feature formula based on the reorganization value, the control area in the preset load control task, the control period in the preset load control task, the period in which each load is currently located, and the area in which each load is currently located is: Among them, b1 is the influencing factor of the regulation period obtained by regression analysis, b2 is the influencing factor of the regulation area obtained by regression analysis, e is the natural base, t is the period of the current load m, a is the area where the current load m is located, S t,a is the calculated spatiotemporal clustering features of the current load m in the time period t and the current load m in the region a. The spatiotemporal clustering feature set S = {S t,a }.

4. The multi-source load management system based on dynamic regulation according to claim 3 is characterized in that: In the spatiotemporal clustering feature calculation module, the specific method of using the preset spatiotemporal clustering feature distribution probability formula to calculate the distribution probability of the spatiotemporal clustering feature of each current load in the set of spatiotemporal clustering features of all current loads is: Among them, S j To transform the spatiotemporal clustering feature set S = {S t,a } is divided into the jth interval of n intervals of equal length, N j For the interval [S j ,S j+1 ) within the spatiotemporal clustering features, ∑ n N n is the total number of empty cluster features falling in n intervals, is the distribution probability corresponding to the spatiotemporal clustering feature falling in the jth interval. According to the distribution probability of the spatiotemporal clustering feature, the spatiotemporal clustering feature set S = {S t,a } are sorted in descending order to obtain a spatiotemporal clustering feature sequence.

5. The multi-source load management system based on dynamic regulation according to claim 4 is characterized in that: In the load control task matching module, the specific method for calculating the characteristic distance between the spatiotemporal clustering characteristics of each current load is: The calculation methods of feature distance include Euclidean distance, Manhattan distance, Chebyshev distance, and Minkowski distance.

6. The multi-source load management system based on dynamic regulation according to claim 5, characterized in that: In the load control task matching module, the specific method of classifying all current loads into several current load sets according to several preset characteristic distance intervals is: The preset characteristic distance intervals are determined based on the value range of characteristic distances between the spatiotemporal clustering characteristics of the historical load.

7. The multi-source load management system based on dynamic regulation according to claim 6 is characterized in that: In the load control task matching module, the specific method of using the preset current load set comprehensive characteristic formula to calculate the comprehensive characteristics of each current load set is: Where: I t,a (C k ) is the indicator function, if the current load set C k When there is a load related to the preset load control task in the time period t of the current load m and the area a where the current load m is located, t,a (C k )=1, otherwise I t,a (C k )=0;∑ t,a S t,a is the sum of the spatiotemporal clustering characteristics of the current load m in the time period t and the current load m in the region a; is the current load set C k The vector of the time period quantization value of the current load m and the regional quantization value of the current load m after spatiotemporal clustering feature mapping, k is the current load set C k The number of current loads m in is the current load set C k The comprehensive characteristic value of .

8. The multi-source load management system based on dynamic regulation according to claim 7 is characterized in that: In the load control task execution module, the specific method of calculating the matching value between the preset load control task and each current load set using a preset multi-dimensional matching formula according to the comprehensive characteristics of each current load set is: Among them, G i is the load value control target of the preset load control task i, ΔT is the difference between the control period in the preset load control task i and the quantitative value of the current load m in the period, ΔA is the difference between the quantitative value of the control area in the preset load control task i and the area where the current load m is located, P i is the load type control priority in the preset load control task i, max[] represents the maximum value, M(C k , i) is the calculated preset load control task i and the current load set C k The matching value of .

9. A multi-source load management method based on dynamic regulation, characterized in that: include: Reorganize the preset load control task by using a preset multivariable reorganization formula to obtain a reorganization value of the preset load control task, calculate the spatiotemporal clustering characteristics of each current load by using a preset spatiotemporal clustering characteristic formula based on the reorganization value, the control area in the preset load control task, the control period in the preset load control task, the period in which each current load is located, and the area in which each current load is located, and calculate the distribution probability of the spatiotemporal clustering characteristics of each current load in the set of spatiotemporal clustering characteristics of all current loads by using a preset spatiotemporal clustering characteristic distribution probability formula, and arrange the spatiotemporal clustering characteristics of each current load in descending order according to the corresponding distribution probability to obtain a spatiotemporal clustering characteristic sequence; Calculate the characteristic distance between the spatiotemporal clustering characteristics of each current load, classify all current loads into several current load sets according to several preset characteristic distance intervals, use the preset current load set comprehensive characteristic formula to calculate the comprehensive characteristics of each current load set, and use the preset multi-dimensional matching formula to calculate the matching value between the preset load control task and each current load set according to the comprehensive characteristics of each current load set, and take the current load set with the largest matching value as the matching result; Each load in the current load set as the matching result is sequentially subjected to the preset load control task according to the order of the corresponding spatiotemporal clustering features in the spatiotemporal clustering feature sequence.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the multi-source load management method based on dynamic regulation described in claim 9 is implemented.

Images