Big data-based intelligent resource allocation and scheduling management method for transformer substation cabinets

Abstract

The invention provides a transformer substation cabinet intelligent resource allocation and scheduling management method based on big data, and relates to the technical field of transformer substation cabinet resource allocation and scheduling management. The method comprises the following steps: acquiring historical operation data of a substation cabinet, and constructing a cabinet resource network topology model; and establishing a cabinet resource three-dimensional evaluation model. And generating a dynamic optimal resource allocation pool. And generating an optimal resource scheduling route sequence. And generating a dynamically adjusted resource scheduling instruction. According to the method, the cabinet operation state data and the load dynamic change information are obtained in real time, the resource scheduling instruction for dynamic adjustment is generated according to the resource scheduling route deviation and the cabinet load deviation, good dynamic adaptability is achieved, various changes in actual operation can be quickly responded, and the resource scheduling efficiency is improved. The intelligent level and the overall operation efficiency of transformer substation cabinet resource management are further improved, and powerful support is provided for stable and reliable operation of a power system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of substation cabinet resource allocation and dispatching management, and particularly relates to a substation cabinet intelligent resource allocation and dispatching management method based on big data. BACKGROUND

[0002] As a key hub of the power system, the substation undertakes the core task of power conversion and distribution. The cabinet, as an important carrier of various electrical equipment in the substation, realizes power transmission and control. The efficiency of resource allocation and dispatching management of the cabinet directly affects the stability and reliability of the operation of the substation. With the continuous expansion of the power system and the diversification and dynamic development of the power demand, the traditional cabinet resource management mode gradually exposes many problems.

[0003] In terms of data utilization, the substation has accumulated a large amount of historical operation data of the cabinet, including operation time, load fluctuation, fault maintenance, energy consumption and other data, as well as rich equipment parameter information such as cabinet model, rated load capacity and interface specification. However, these data are often in a scattered storage state and are not fully integrated, making it difficult to achieve comprehensive and accurate characterization of cabinet resources. At the same time, the current power demand prediction lacks deep mining of historical data and multi-dimensional feature correlation analysis, resulting in a deviation between the prediction result and the actual demand, and unable to provide a reliable basis for cabinet resource allocation.

[0004] In terms of resource evaluation and dispatching, the traditional method for evaluating the key characteristics of cabinet resources such as equipment aging and degradation is relatively single, relying mainly on manual experience or simple mathematical statistics, and lacking intelligent evaluation based on big data, making it difficult to accurately reflect the actual available state and potential of cabinet resources. Moreover, in the resource dispatching process, the collaborative relationship between cabinets is not fully explored, and the generated dispatching route is mostly static planning, resulting in low dispatching efficiency, resource waste and untimely allocation. In addition, in the dispatching execution process, there is a lack of real-time monitoring and dynamic adjustment mechanism for the running state of the cabinet and the dynamic change of the load, which cannot quickly respond to the deviation in actual operation, further affecting the overall operation efficiency of the substation. SUMMARY

[0005] The present application provides a substation cabinet intelligent resource allocation and dispatching management method based on big data to solve the defects in the prior art.

[0006] The present application provides a substation cabinet intelligent resource allocation and dispatching management method based on big data, comprising: Collecting historical operation data of the substation cabinet, equipment parameter information and load demand records, and using a multi-dimensional feature correlation algorithm to construct a cabinet resource network topology model.

[0007] The current load data, equipment health state and future electricity demand prediction list of the target transformer substation are acquired, the capacity fluctuation interval of the cabinet resource is quantified, the equipment aging attenuation curve is modeled and the resource scheduling response threshold is analyzed, and a three-dimensional evaluation model of the cabinet resource is established.

[0008] Based on the cabinet resource network topology model and the current electricity demand prediction list, a dynamic preferred resource allocation pool is generated by using a resource association mining strategy.

[0009] The physical position coordinates of each cabinet in the resource allocation pool and the three-dimensional layout map of the transformer substation are acquired, a three-dimensional evaluation model of the cabinet resource is combined, a space-time coordination field construction technology and a global dynamic scheduling optimization algorithm are used, and an optimal resource scheduling route sequence is generated.

[0010] The cabinet running state data and load dynamic change information are acquired in real time, and a dynamic adjusted resource scheduling instruction is generated according to the resource scheduling route deviation and the cabinet load deviation.

[0011] According to the intelligent resource allocation and scheduling management method of the transformer substation cabinet based on big data provided by the application, the historical operation data includes cabinet operation duration data, load fluctuation data, fault maintenance record and energy consumption data. The equipment parameter information includes cabinet model, rated load capacity, interface specification model, rated working voltage, rated working current, heat dissipation power and design service life. The current electricity demand prediction list represents the planned electricity load list based on the historical electricity data of the transformer substation and the future electricity scene prediction.

[0012] According to the intelligent resource allocation and scheduling management method of the transformer substation cabinet based on big data provided by the application, the process of constructing the cabinet resource network topology model includes: The historical operation data of the transformer substation cabinet is preprocessed, and the resource load combination information in the complete operation cycle is extracted.

[0013] The cabinet resource association matrix is defined, the first-order coordination strength of any two cabinets is calculated, and the multi-order coordination relationship is extended. The first-order coordination strength represents the weighted average value of the load matching degree and the resource utilization rate index. The multi-order coordination relationship represents the resource coordination scheduling mode of the multi-cabinet combination.

[0014] The multi-order coordination relationship is stored in the form of a three-dimensional matrix, and the topology edge weight is generated by matrix decomposition.

[0015] The connection edges with a topology edge weight lower than a preset coordination strength threshold are removed, and a cabinet resource network topology graph is obtained, in which the nodes represent the cabinets and the edges represent the multi-cabinet combination coordination relationship.

[0016] The intelligent resource allocation and scheduling management method for a transformer substation cabinet based on big data provided by the application comprises the following steps of: The historical operation data of the transformer substation cabinet is preprocessed, and the process of extracting resource load combination information in a complete operation cycle comprises the following steps of: The historical operation data is divided according to types and identification information is added.

[0017] An abnormal data is identified and removed by using a double detection mechanism.

[0018] A differential completion strategy is used for different types of missing data.

[0019] Based on the start-stop record and load change characteristics of the cabinet, a complete operation cycle is divided, load-related characteristics in the cycle are extracted, and a resource load combination information data set is formed.

[0020] The intelligent resource allocation and scheduling management method for a transformer substation cabinet based on big data provided by the application comprises the following steps of: Based on the historical operation data of the cabinet, the load capacity arithmetic mean and variance of the cabinet in different operation periods are calculated.

[0021] The time periodicity characteristics of the cabinet load capacity are identified, and a fluctuation function combined with a seasonal attenuation factor is constructed.

[0022] The fluctuation function value is taken as the center line, and a dynamic fluctuation interval is generated according to the variance multiple.

[0023] The fluctuation interval range is dynamically adjusted by real-time monitoring of the current load capacity value of the cabinet.

[0024] The intelligent resource allocation and scheduling management method for a transformer substation cabinet based on big data provided by the application comprises the following steps of: The historical operation and fault maintenance records of the cabinet are analyzed, the average operation time from the cabinet being put into use to the first failure is calculated, and the average operation time is mapped as an initial health value.

[0025] An exponential decay function model is established, so that the initial health value presents a nonlinear decay characteristic with the passage of operation time.

[0026] The aging decay coefficient is dynamically adjusted according to the current operation environment parameters, including environmental temperature, humidity and dust concentration.

[0027] When the deviation between the actual health value and the predicted value of the cabinet exceeds the preset deviation threshold, the decay function parameters are corrected.

[0028] The intelligent resource allocation and scheduling management method for a transformer substation cabinet based on big data provided by the application comprises the following steps of: Extract response delay data in the box cabinet history resource scheduling record, and fit the benchmark threshold of the box cabinet scheduling response.

[0029] According to the benchmark threshold, a scheduling response efficiency conversion model is established, and a response efficiency function is calculated.

[0030] Real-time integration of the overall load pressure of the current substation, and dynamic updating of the resource scheduling response threshold.

[0031] The method for intelligent resource allocation and scheduling management of the substation box cabinet based on big data provided by the application adopts a resource association mining strategy to generate a dynamically optimized resource allocation pool, and the process includes: In the box cabinet resource network topology graph, the resource adaptation rate of all candidate box cabinets that can be reached through direct cooperation paths is calculated based on the current electricity demand prediction list.

[0032] Carry out multi-demand cooperation correction, and detect the second-order influence of the load combination in the current electricity demand prediction list on the candidate box cabinet.

[0033] Real-time monitoring of the attenuation rate of the pool box cabinet resource cooperation strength, and screening of the initial resource allocation pool according to the dynamically preset adaptation rate threshold.

[0034] The method for intelligent resource allocation and scheduling management of the substation box cabinet based on big data provided by the application generates an optimal resource scheduling route sequence, and the process includes: Map the physical location coordinates and resource value weights of each box cabinet in the resource allocation pool to a four-dimensional space-time scheduling field.

[0035] Superimpose the three-dimensional evaluation model of the box cabinet resource to form a comprehensive constraint field, establish a space-time scheduling optimization equation, obtain a continuous scheduling path curve, and the comprehensive constraint field includes the load constraint defined by the current capacity fluctuation interval, the maximum operation time determined by the equipment health value, and the time limit converted from the scheduling response threshold.

[0036] Discretize the continuous scheduling path curve into a scheduling node sequence, and sort and output it as a scheduling route instruction set according to physical accessibility to obtain an optimal resource scheduling route sequence.

[0037] The method for intelligent resource allocation and scheduling management of the substation box cabinet based on big data provided by the application generates a dynamically adjusted resource scheduling instruction, and the process includes: Real-time acquisition of box cabinet real-time position coordinates and load dynamic change data, establishment of a resource scheduling entropy monitoring model according to the offset of the real-time position coordinates and load data from the optimal resource scheduling route sequence and the resource allocation pool.

[0038] The dynamic re-planning condition is triggered when any offset exceeds a preset threshold, and the dynamic re-planning condition includes recalculating the space-time scheduling field gradient direction based on the latest data, and generating a revised instruction set including scheduling direction adjustment, speed optimization and cabinet replacement suggestions.

[0039] The power transformation station cabinet intelligent resource allocation and scheduling management method based on big data provided by the application fully collects and standardizes the preprocessing of historical operation data, equipment parameter information and load demand records of the power transformation station cabinet, constructs a cabinet resource network topology model by using a multi-dimensional feature correlation algorithm, fully integrates scattered data resources, accurately describes the multi-dimensional features and mutual correlation of cabinet resources, and effectively solves the problem of insufficient utilization of traditional data.

[0040] In terms of resource evaluation accuracy, by establishing a three-dimensional evaluation model of cabinet resources, the capacity fluctuation interval of cabinet resources is quantified, the equipment aging attenuation curve is modeled, and the resource scheduling response threshold is analyzed, which can accurately evaluate the actual state and potential of cabinet resources from multiple dimensions. Compared with the traditional single evaluation method, the accuracy of resource evaluation is greatly improved, which provides a more reliable basis for the reasonable allocation of resources and helps to avoid resource waste or insufficient allocation.

[0041] In terms of resource allocation and scheduling efficiency, based on the cabinet resource network topology model and the current electricity demand prediction list, a dynamic optimal resource allocation pool is generated by using resource correlation mining strategy, combined with the cabinet physical location coordinates and the three-dimensional layout map of the power transformation station, the optimal resource scheduling route sequence is generated by using the space-time coordination field construction technology and the global dynamic scheduling optimization algorithm, which fully excavates the cooperative relationship between cabinets, realizes the dynamic and optimal allocation of resources, and significantly improves the efficiency and rationality of resource scheduling, ensuring the efficiency of power transformation station operation. BRIEF DESCRIPTION OF DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the application or prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0043] Figure 1 It is a process schematic diagram of the power transformation station cabinet intelligent resource allocation and scheduling management method based on big data provided by the embodiment of the application; Figure 2 It is a process schematic diagram of constructing a cabinet resource network topology model in the embodiment of the application; Figure 3 It is a process schematic diagram of quantifying the capacity fluctuation interval of cabinet resources in the embodiment of the application; Figure 4 is a flowchart of modeling the equipment aging attenuation curve of the cabinet resource in the embodiment of the present application; Figure 5 is a flowchart of analyzing the resource scheduling response threshold of the cabinet resource in the embodiment of the present application. DETAILED DESCRIPTION

[0044] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be described clearly and completely below with reference to the drawings in the present application. Obviously, the described embodiments are some embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the present application.

[0045] The present application is described below with reference to the drawings. Figures 1-5 The present application provides a substation cabinet intelligent resource allocation and scheduling management method based on big data.

[0046] Figure 1 is a structural schematic diagram of the substation cabinet intelligent resource allocation and scheduling management method based on big data provided by the embodiment of the present application.

[0047] As shown in Figure 1 , the substation cabinet intelligent resource allocation and scheduling management method based on big data provided by the embodiment of the present application comprises: Collecting historical operation data, equipment parameter information and load demand records of the substation cabinet, using a multi-dimensional feature correlation algorithm to construct a cabinet resource network topology model.

[0048] The historical operation data includes cabinet operation duration data, load fluctuation data, fault maintenance records and energy consumption data. The equipment parameter information includes cabinet model, rated load capacity, interface specification model, rated working voltage, rated working current, heat dissipation power and design service life. The current power demand prediction list represents the planned power load list based on the historical power consumption data of the substation and the future power consumption scene prediction.

[0049] Figure 2 is a flowchart of constructing the cabinet resource network topology model in the embodiment of the present application.

[0050] As shown in Figure 2 , the process of constructing the cabinet resource network topology model comprises: Pretreating the historical operation data of the substation cabinet, and extracting resource load combination information in a complete operation cycle.

[0051] The box resource association matrix is defined, the first-order coordination strength of any two boxes is calculated, and the multi-order coordination relationship is extended. The first-order coordination strength represents the weighted average of the load matching degree and the resource utilization rate index, and is expressed by the formula: In the formula, is the first-order coordination strength of box i and box j, , is the weight coefficient, is the load matching value of box i and box j, is the maximum load matching value of all box combinations, is the resource utilization rate of the box and the box, is the maximum resource utilization rate of all box combinations. The multi-order coordination relationship represents the resource coordination scheduling mode of multiple box combinations.

[0052] The multi-order coordination relationship is stored in the form of a three-dimensional matrix, and the topological edge weight is generated by matrix decomposition, which is expressed by the formula: In the formula, is the topological edge weight between the box and the box, is the k-order coordination strength of box i and box j, M and N are the total number of boxes, is the maximum order of the coordination relationship.

[0053] Remove the connection edges with topological edge weights lower than the preset coordination strength threshold to obtain the box resource network topology graph. In the graph, the nodes represent the boxes, and the edges represent the multi-box combination coordination relationship.

[0054] The historical operation data of the substation box is preprocessed, and the process of extracting the resource load combination information in the complete operation cycle includes: The historical operation data is divided by type and added with identification information.

[0055] Anomaly data is identified and removed by using a double detection mechanism.

[0056] Differentiated completion strategies are used for different types of missing data.

[0057] Based on the box operation start-stop record and load change characteristics, the complete operation cycle is divided, the load-related characteristics in the cycle are extracted, and the resource load combination information dataset is formed.

[0058] The historical operation data is divided into operation state class data, fault class data and energy consumption class data according to types, the operation state class data includes operation duration, real-time load and voltage and current values, the fault class data includes fault occurrence time, fault type, fault position and maintenance duration, and the energy consumption class data includes unit time energy consumption, daily cumulative energy consumption and seasonal energy consumption peak value, and a timestamp, a cabinet unique identifier and a running scene label are added to each type of data.

[0059] The preliminary abnormal data exceeding the preset range is screened out through the 3 sigma criterion, and the data is standardized and input into the isolation forest model, and the finally determined abnormal data is removed.

[0060] The continuous data of the operation state class is interpolated by using the LSTM interpolation method based on the time sequence, the mode is used to complete the discrete data of the fault class, and the adjacent period mean weighted completion is used for the energy consumption class data.

[0061] Based on the start-stop record of the cabinet and the load change characteristics, a cycle is taken as "low load start-load climb-stable operation-load decline-low load stop", cycle nodes are identified through a sliding window method, load peak value, peak duration, average load and load fluctuation amplitude are extracted in each cycle, and a resource load combination information data set is formed.

[0062] Current load data, equipment health status and future electricity demand prediction list of the target transformer substation are acquired, capacity fluctuation interval quantization of the cabinet resource is carried out, equipment aging attenuation curve modeling and resource scheduling response threshold analysis are carried out, and a three-dimensional evaluation model of the cabinet resource is established.

[0063] Figure 3 It is a process schematic diagram for quantizing the capacity fluctuation interval of the cabinet resource in the embodiment of the application.

[0064] As shown in Figure 3 The process of quantizing the capacity fluctuation interval of the cabinet resource includes: Based on the historical operation data of the cabinet, the arithmetic mean and variance of the load capacity of the cabinet in different operation periods are counted, and the formula is represented as: In the formula, is the arithmetic mean of the load capacity, T is the total number of statistical periods, is the load capacity of the t period, is the variance.

[0065] The time periodicity characteristics of the cabinet load capacity are identified, a fluctuation function combined with a seasonal attenuation factor is constructed, and the formula is represented as: In the formula, is the load capacity fluctuation value at time t, is the arithmetic mean of the load capacity, is the standard deviation of the load capacity, is the seasonal attenuation factor, is the time of the last load peak, and f is the load fluctuation frequency, is the initial statistical time.

[0066] Taking the fluctuation function value as the center line, the dynamic fluctuation interval is generated according to the variance multiple, and the formula is: In the formula, is the standard deviation of the load capacity, is the confidence coefficient (usually 1.5-2.5).

[0067] By monitoring the current load capacity value of the cabinet in real time, the fluctuation interval range is dynamically adjusted.

[0068] Figure 4 is the process diagram of modeling the equipment aging attenuation curve of the cabinet resource in the embodiment of the application.

[0069] As Figure 4 shown, the process of modeling the equipment aging attenuation curve of the cabinet resource includes: Analyzing the cabinet historical operation and maintenance records, calculating the average running time from the cabinet being put into use to the first failure, and mapping the average running time as the initial health value, the formula is: In the formula, is the initial health value, is the average running time, is the designed service life of the cabinet.

[0070] An exponential decay function model is established, so that the initial health value presents a nonlinear decay characteristic with the running time, and the formula is: In the formula, is the cabinet health value at time t, is the aging attenuation coefficient.

[0071] The aging attenuation coefficient is dynamically adjusted according to the current running environment parameters, and the current running environment parameters include environmental temperature, humidity and dust concentration, and the adjustment formula is: In the formula, is the adjusted aging attenuation coefficient, a, b, and c are environmental influence coefficients, is the ambient temperature deviation value (relative to the standard operating temperature), is the ambient humidity deviation value (relative to the standard operating humidity), is the ambient dust concentration deviation value (relative to the standard operating dust concentration).

[0072] When the actual health value of the cabinet deviates from the predicted value by more than the preset deviation threshold, the decay function parameters are corrected, and the formula is: In the formula, is the corrected aging decay coefficient, is the learning rate, is the actual health value, is the predicted health value.

[0073] Figure 5 is the process flow diagram for analyzing the resource scheduling response threshold of the cabinet resources in the embodiment of the present application.

[0074] As Figure 5 shown, the process of analyzing the resource scheduling response threshold of the cabinet resources includes: Extracting the response delay data in the historical resource scheduling records of the cabinet, fitting the baseline threshold of the cabinet scheduling response, and the formula is: In the formula, is the baseline threshold, is the total number of scheduling records, is the response delay time of the ith scheduling.

[0075] According to the baseline threshold, a scheduling response efficiency conversion model is established, and a response efficiency function is calculated, and the formula is: In the formula, is the response efficiency, R is the actual response delay time, is the smoothing coefficient (to avoid the denominator being 0).

[0076] Real-time integration of the overall load pressure of the current substation, and dynamic updating of the resource scheduling response threshold, and the updating formula is: In the formula, is the updated response threshold, d is the load pressure influence coefficient, P is the current load pressure value (between 0 and 1).

[0077] Based on the cabinet resource network topology model and the current electricity demand prediction list, a dynamic optimal resource allocation pool is generated by using a resource association mining strategy, and the process includes: In the cabinet resource network topology diagram, taking the current electricity demand prediction list as the starting point, the resource adaptation rate of all candidate cabinets reachable through direct coordination path is calculated, which is expressed as: In the formula, is the resource adaptation rate of candidate cabinet C, is the number of associated cabinets, is the load matching value of candidate cabinet C and associated cabinet i, is the maximum load matching value of candidate cabinet C and all associated cabinets, is the load fluctuation gradient logarithm of candidate cabinet C and associated cabinet i.

[0078] Multi-demand coordination correction is carried out to detect the second-order influence of load combination in the current electricity demand prediction list on candidate cabinets.

[0079] The decay rate of the coordination strength of the cabinet resources in the pool is monitored in real time, and the initial resource allocation pool is selected according to the dynamic preset adaptation rate threshold, which is expressed as: In the formula, is the dynamic optimal resource allocation pool, is the adaptation rate threshold, is the adaptation rate decay rate, is the lower limit of the decay rate tolerance, and I is the demand load set, is the second-order coordination partial derivative of load and cabinet, is the coordination gain threshold.

[0080] The physical location coordinates of each cabinet in the resource allocation pool and the three-dimensional layout map of the substation are obtained, combined with the cabinet resource three-dimensional evaluation model, and the space-time coordination field construction technology and global dynamic scheduling optimization algorithm are used to generate the optimal resource scheduling route sequence, the process includes: The physical location coordinates and resource value weight of each cabinet in the resource allocation pool are mapped to a four-dimensional space-time scheduling field.

[0081] The cabinet resource three-dimensional evaluation model is superimposed to form a comprehensive constraint field, and a space-time scheduling optimization equation is established to obtain a continuous scheduling path curve. The comprehensive constraint field includes load constraints defined by the current capacity fluctuation interval, maximum operation time determined by the equipment health value, and time limit converted from the scheduling response threshold. The optimization equation is: In the formula, is the total loss of the scheduling path , , are the scheduling start and end times, , and These are the weighting coefficients. Let be the path distance loss at time t. Let t be the health value of the container. Let t be the scheduling response delay time. This is the updated response threshold.

[0082] The continuous scheduling path curve is discretized into a sequence of scheduling nodes, and then sorted by physical reachability to output a set of scheduling route instructions, thus obtaining the optimal resource scheduling route sequence, expressed by the formula: In the formula, This is the optimal resource scheduling route sequence. For the number of scheduling nodes, The location coordinates and arrival time of the container at the k2th node. .

[0083] The system acquires real-time data on the operational status of containers and dynamic load changes, and generates dynamically adjusted resource scheduling instructions based on resource scheduling route deviations and container load deviations. The process includes: Real-time location coordinates and load dynamic changes of containers are acquired. Based on the real-time location coordinates and load data, and the offsets from the optimal resource scheduling route sequence and resource allocation pool, a resource scheduling entropy monitoring model is established. The formula is as follows: In the formula, Let be the scheduling entropy at time t. For unscheduled container collection, For the resource matching rate of candidate container C, The path deviation coupling coefficient. This is the vector of the actual moving speed of the container. This is the planned scheduling path velocity vector.

[0084] When any offset exceeds a preset threshold, a dynamic replanning condition is triggered. The dynamic replanning condition includes recalculating the air conditioning field gradient direction based on the latest data and generating a set of correction instructions, including scheduling direction adjustment, speed optimization, and cabinet replacement suggestions. The replanning condition formula is expressed as follows: In the formula, For monitoring time windows, Let be the scheduling entropy at time t. For historical scheduling entropy, Entropy change threshold Real-time location of the container. For the planned scheduling path location, A path offset distance threshold.

[0085] In summary, the embodiment provides a substation cabinet intelligent resource allocation and scheduling management method based on big data. Through comprehensive collection and standardized preprocessing of historical operation data, equipment parameter information and load demand records of the substation cabinet, a cabinet resource network topology model is constructed using a multi-dimensional feature correlation algorithm, the dispersed data resources are fully integrated, the multi-dimensional features and mutual correlation of the cabinet resources are accurately described, and the problem of insufficient utilization of traditional data is effectively solved.

[0086] In terms of resource evaluation accuracy, the cabinet resource is quantified in the capacity fluctuation interval, the equipment aging attenuation curve is modeled, and the resource scheduling response threshold is analyzed through the establishment of a three-dimensional evaluation model of the cabinet resource, which can accurately evaluate the actual state and potential of the cabinet resource from multiple dimensions. Compared with the traditional single evaluation method, the accuracy of resource evaluation is greatly improved, which provides a more reliable basis for the reasonable allocation of resources and helps to avoid resource waste or insufficient allocation.

[0087] In terms of resource allocation and scheduling efficiency, based on the cabinet resource network topology model and the current power demand prediction list, a dynamic optimal resource allocation pool is generated using a resource correlation mining strategy, combined with the cabinet physical location coordinates and the three-dimensional layout map of the substation, the optimal resource scheduling route sequence is generated using the space-time coordination field construction technology and the global dynamic scheduling optimization algorithm, the cooperative relationship between cabinets is fully tapped, the dynamic and optimized allocation of resources is realized, and the efficiency and rationality of resource scheduling are significantly improved, ensuring the efficiency of substation operation.

[0088] In addition, real-time acquisition of cabinet operation state data and load dynamic change information, generation of dynamic adjustment resource scheduling instructions according to resource scheduling route deviation and cabinet load deviation, good dynamic adaptability, quick response to various changes in actual operation, further improvement of the intelligent level and overall operation efficiency of substation cabinet resource management, and strong support for stable and reliable operation of the power system.

[0089] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be realized by means of software and the necessary general hardware platform, and of course, it can also be realized by hardware. Based on such understanding, the above technical solutions essentially or say the part that makes a contribution to the prior art can be embodied in the form of a software product, which can be stored in a computer readable storage medium such as ROM / RAM, magnetic disk, optical disk, etc., and includes a plurality of instructions to make a computer device (which can be a personal computer, server, or network device, etc.) execute the method described in each embodiment or some part of the embodiment.

[0090] It should be pointed out finally that the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit the same; and although the present application has been described in detail with reference to the foregoing embodiments, it should be appreciated by those skilled in the art that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features thereof can be replaced equivalently; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A substation cabinet intelligent resource allocation and dispatch management method based on big data, characterized in that, The application relates to a power distribution station box cabinet resource scheduling method and device. The method comprises the following steps: Collecting historical operation data, equipment parameter information and load demand records of a power distribution station box cabinet, using a multi-dimensional feature correlation algorithm to construct a box cabinet resource network topology model; Obtaining current load data, equipment health status and future electricity demand prediction list of a target power distribution station, quantifying the capacity fluctuation interval of the box cabinet resource, modeling the equipment aging attenuation curve and analyzing the resource scheduling response threshold, and establishing a three-dimensional evaluation model of the box cabinet resource; Based on the box cabinet resource network topology model and the current electricity demand prediction list, a resource correlation mining strategy is used to generate a dynamic optimal resource allocation pool; Obtaining the physical location coordinates of each box cabinet in the resource allocation pool and the three-dimensional layout map of the power distribution station, combining the three-dimensional evaluation model of the box cabinet resource, using a time-space coordination field construction technology and a global dynamic scheduling optimization algorithm to generate an optimal resource scheduling route sequence; 2.The big data based intelligent resource allocation and dispatching management method of transformer substation cabinet according to claim 1, characterized in that, Real-time acquisition of box cabinet operation state data and load dynamic change information, and generation of a dynamically adjusted resource scheduling instruction according to the resource scheduling route deviation and the box cabinet load deviation. 3.The big data based intelligent resource allocation and dispatching management method of transformer substation cabinet according to claim 1, characterized in that, The historical operation data includes box cabinet operation duration data, load fluctuation data, fault maintenance records and energy consumption data; the equipment parameter information includes box cabinet model, rated load capacity, interface specification model, rated working voltage, rated working current, heat dissipation power and design service life; and the current electricity demand prediction list represents a planned electricity load list based on historical electricity data and future electricity scene prediction of the power distribution station. The process of constructing the box cabinet resource network topology model comprises the following steps: Pretreatment of the historical operation data of the power distribution station box cabinet, and extraction of resource load combination information in a complete operation cycle; Defining a box cabinet resource correlation matrix, calculating the first-order coordination strength of any two box cabinets, and extending the multi-order coordination relationship, wherein the first-order coordination strength represents the weighted average value of the load matching degree and the resource utilization rate index; and the multi-order coordination relationship represents the resource coordination scheduling mode of a plurality of box cabinet combinations; Storing the multi-order coordination relationship in the form of a three-dimensional matrix, and generating a topology edge weight through matrix decomposition; 4.The big data based intelligent resource allocation and dispatching management method of transformer substation cabinet according to claim 3, characterized in that, Removing the connection edges with a topology edge weight lower than a preset coordination strength threshold value to obtain a box cabinet resource network topology graph, wherein the nodes in the graph represent the box cabinets, and the edges represent the multi-box cabinet combination coordination relationship. The process of pretreating the historical operation data of the power distribution station box cabinet and extracting resource load combination information in a complete operation cycle comprises the following steps: Dividing the historical operation data according to types and adding identification information; Using a double detection mechanism to identify and eliminate abnormal data; Using a differential completion strategy for different types of missing data; 5.The big data based intelligent resource allocation and dispatching management method of transformer substation cabinet according to claim 1, characterized in that, Based on the box cabinet operation start-stop record and load change characteristics, dividing a complete operation cycle, extracting load-related characteristics in the cycle, and forming a resource load combination information data set. The process of quantifying the capacity fluctuation interval of the box cabinet resource comprises the following steps: Based on the historical operation data of the box cabinet, the arithmetic mean and variance of the load capacity of the box cabinet in different operation periods are counted; Identifying the time periodicity characteristics of the load capacity of the box cabinet, and constructing a fluctuation function combined with a seasonal attenuation factor; Taking the fluctuation function value as the center line, generating a dynamic fluctuation interval according to the variance multiple; Through real-time monitoring of the current load capacity value of the box cabinet, the fluctuation interval range is dynamically adjusted. 6.The big data based intelligent resource allocation and dispatching management method of transformer substation cabinet according to claim 1, characterized in that, The process of modeling the equipment aging attenuation curve of the cabinet resources includes: analyzing the cabinet historical operation and maintenance records, calculating the average running time of the cabinet from being put into use to the first failure, and mapping the average running time as an initial health value; establishing an exponential decay function model, so that the initial health value presents a nonlinear decay characteristic over time; dynamically adjusting the aging attenuation coefficient according to the current operating environment parameters, including temperature, humidity and dust concentration; when the actual health value of the cabinet deviates from the predicted value by more than a preset deviation threshold, the decay function parameters are corrected. 7.The big data based intelligent resource allocation and dispatching management method of transformer substation cabinet according to claim 1, characterized in that, The process of analyzing the resource scheduling response threshold of the cabinet resources includes: extracting response delay data from the historical resource scheduling records of the cabinet, fitting the baseline threshold of the cabinet scheduling response; establishing a scheduling response efficiency conversion model according to the baseline threshold, and calculating a response efficiency function; real-time integration of the overall load pressure of the substation, and dynamic updating of the resource scheduling response threshold. 8.The big data based intelligent resource allocation and dispatching management method of transformer substation cabinet according to claim 1, characterized in that, The process of generating a dynamic optimal resource allocation pool using resource association mining strategy includes: in the cabinet resource network topology graph, taking the current electricity demand prediction list as the starting point, calculating the resource adaptation rate of all candidate cabinets that can be reached through direct cooperation path; performing multi-demand cooperation correction to detect the second-order influence of the load combination in the current electricity demand prediction list on the candidate cabinet; real-time monitoring of the decay rate of the cooperation strength of the cabinet resources in the pool, and screening the initial resource allocation pool according to the dynamic preset adaptation rate threshold. 9.The big data based intelligent resource allocation and dispatching management method of transformer substation cabinet according to claim 1, characterized in that, The process of generating the optimal resource scheduling route sequence includes: mapping the physical location coordinates and resource value weights of each cabinet in the resource allocation pool to a four-dimensional space-time scheduling field; superimposing a three-dimensional evaluation model of the cabinet resources to form a comprehensive constraint field, establishing a space-time scheduling optimization equation, and obtaining a continuous scheduling path curve, the comprehensive constraint field includes load constraints defined by the current capacity fluctuation interval, maximum running time determined by the equipment health value, and time limit converted from the scheduling response threshold; discretizing the continuous scheduling path curve into a scheduling node sequence, and sorting it into a scheduling route instruction set according to physical accessibility to obtain the optimal resource scheduling route sequence. 10.The big data based intelligent resource allocation and dispatching management method of transformer substation cabinet according to claim 1, characterized in that, The process of generating dynamically adjusted resource scheduling instructions includes: real-time acquisition of cabinet real-time location coordinates and load dynamic change data, and establishment of a resource scheduling entropy monitoring model according to the offset between real-time location coordinates and load data, optimal resource scheduling route sequence and resource allocation pool; when any offset exceeds the preset threshold, trigger the dynamic re-planning condition, which includes recalculating the gradient direction of the space-time scheduling field based on the latest data, and generating a correction instruction set including scheduling direction adjustment, speed optimization and cabinet replacement suggestion.

Images