Intelligent spatial-temporal characteristic analysis system based on electric power customer order

Abstract

The invention relates to the technical field of electric power data analysis, in particular to an intelligent spatial-temporal feature analysis system based on an electric power customer order, and discloses an electric power customer order spatial-temporal feature prediction module, an electric power customer order simulation delivery module and an electric power customer order delivery decision module. Through an electric power customer order spatio-temporal feature prediction module, an electric power customer order simulation distribution module and an electric power customer order distribution decision module, deep analysis is carried out on customer electricity consumption spatio-temporal features of a power grid based on an electric power customer order; the power consumption characteristics of each physical area in a future period are predicted by combining a long-short-term neural network model technology, the change of the power consumption condition of each physical area is dynamically simulated, and a power distribution scheme meeting the power consumption demand of each physical area is preset. And deeply analyzing the matching condition of the key power line for power peak load shifting under the power distribution scheme, and combining with the actual analysis to analyze the adjustability of the power system under the power distribution scheme.

Description

Technical Field

[0001] This invention relates to the field of power data analysis technology, and more specifically, to an intelligent analysis system for spatiotemporal characteristics based on power customer orders. Background Technology

[0002] Electricity customer orders serve as a crucial source of information reflecting user electricity demand, and their spatiotemporal characteristics contain rich information on electricity consumption behavior patterns and trends. However, most existing electricity data analysis systems lack in-depth mining and utilization of the spatiotemporal characteristics of electricity customer orders.

[0003] Furthermore, advanced artificial intelligence technologies such as Long Short-Term Memory (LSTM) models have demonstrated powerful capabilities in time series forecasting and pattern recognition. Existing research has attempted to apply models like LSTM to load forecasting for power grid systems. However, load forecasting of the power grid system alone lacks a comprehensive consideration and analysis of the spatiotemporal characteristics of electricity customer orders, and it fails to account for the availability of power instability under actual conditions. This makes it difficult to provide effective foresight and flexibility for the formulation of subsequent power allocation and dispatch strategies.

[0004] Therefore, developing a system capable of intelligent analysis based on the spatiotemporal characteristics of electricity customer orders, combined with long short-term neural network model technology to accurately predict the electricity consumption characteristics of various physical regions over a future period, dynamically simulating changes in electricity consumption in each physical region, pre-setting power allocation schemes to meet the electricity demand of each physical region, and deeply analyzing the coordination of key power lines with peak shaving and valley filling under the power allocation scheme and the adjustability of the power system are of great significance for improving the stability, reliability and economy of the power system. Summary of the Invention

[0005] In view of the shortcomings of existing technologies, the purpose of this invention is to provide an intelligent analysis system for spatiotemporal characteristics based on electricity customer orders.

[0006] To achieve the above objectives, the present invention provides the following technical solution: The intelligent analysis system based on the spatiotemporal characteristics of electricity customer orders includes a module for predicting the spatiotemporal characteristics of electricity customer orders, a module for simulating the delivery of electricity customer orders, and a module for making delivery decisions for electricity customer orders. The power customer order spatiotemporal feature prediction module: Based on the connection relationship of transmission lines, the power grid is divided into several physical regions, and the power prediction spatiotemporal feature set of each physical region is determined periodically; The power customer order simulation delivery module: Based on the spatiotemporal feature set of power forecast for each physical region, it constructs a power delivery simulation model and controls the power delivery simulation model to perform a power allocation; The power customer order delivery decision module: After the power allocation is completed, it obtains the allocation capacity reservation value corresponding to the power allocation, and determines whether to perform the power allocation to the power grid based on the comparison result between the allocation capacity reservation value and the allocation capacity reservation threshold.

[0007] Furthermore, the spatiotemporal feature set of power forecasting for a physical region is determined as follows: a physical region is selected, all customers within that physical region are identified, the power forecasting demand feature set for each customer is determined, and the power forecasting demand feature sets for each customer are combined as a dataset to obtain the spatiotemporal feature set of power forecasting for the physical region.

[0008] Furthermore, a customer's electricity forecast demand feature set is determined as follows: a customer is selected, the customer's forecast demand features for each electricity consumption feature are determined, and the forecast demand features for each electricity consumption feature are combined as a dataset to obtain the customer's electricity forecast demand feature set.

[0009] Furthermore, the predicted demand features of a customer for a specific electricity consumption characteristic are determined as follows: Select an electricity consumption characteristic, collect the customer's actual demand features for that electricity consumption characteristic over the previous n consecutive spatiotemporal characteristic analysis periods, combine the n actual demand features of that electricity consumption characteristic as a time series dataset to obtain an actual demand time series feature set, obtain the predicted demand feature time series model of the customer for that electricity consumption characteristic, input the actual demand time series feature set into the predicted demand feature time series model, and output the predicted demand features of the customer for that electricity consumption characteristic.

[0010] Furthermore, for a spatiotemporal feature analysis period, the actual demand features of an electricity consumption feature are collected in the following way: a spatiotemporal feature analysis period is selected, the electricity customer orders of the customer in the spatiotemporal feature analysis period are collected, an electricity consumption feature is selected, and the actual demand features of the electricity consumption feature are extracted from the electricity customer orders through feature extraction.

[0011] Furthermore, a power allocation capacity reservation threshold is set. When the reserved power allocation capacity value is greater than the reserved power allocation capacity threshold, the power allocation is performed on the power grid in the next spatiotemporal feature analysis cycle. Otherwise, the power distribution simulation model is controlled to perform the next power allocation.

[0012] Furthermore, the reserved allocation capacity value corresponding to power allocation is obtained based on the following method: Before power allocation begins, the power distribution simulation model determines all critical power lines within the model. The model then begins power allocation for one spatiotemporal feature analysis cycle. Within this cycle, each... The system continuously acquires the power load of each critical power line and the flexible load allocation values ​​of each simulated physical region within the power distribution simulation model. After power allocation is completed, it determines the line load depth analysis value of each critical power line and the sudden allocation satisfaction value of each simulated physical region. A sudden allocation satisfaction threshold is set. When the sudden allocation satisfaction value of a simulated physical region exceeds the sudden allocation satisfaction threshold, the simulated physical region is marked as a sudden allocation satisfaction region. The average line load depth analysis value is calculated by summing and averaging the power loads of all critical power lines. The total number of areas where emergency allocation is satisfied is marked as (yy), through + The reserved allocation capacity for this power distribution is calculated using (yy)*s4. , where s3 is the third coefficient and s4 is the fourth coefficient.

[0013] Furthermore, the burst allocation satisfaction value of the simulated physical region is determined as follows: a simulated physical region is defined, all load flexible allocation values ​​obtained in the simulated physical region within the spatiotemporal characteristic analysis period are collected, and the summation and average of all load flexible allocation values ​​are calculated to obtain the burst allocation satisfaction value of the simulated physical region.

[0014] Furthermore, the load flexibility allocation value of the simulated physical area is determined based on the following method: a simulated physical area is defined, and all key power lines connected to the simulated physical area are marked as associated lines. The power load of each associated line is obtained, and all associated lines are compared pairwise. The absolute difference between the power loads of the two compared associated lines is calculated to obtain the power load difference value. The summation and averaging of all power load difference values ​​are then calculated to obtain the load flexibility allocation value of the simulated physical area.

[0015] Furthermore, the line load depth analysis value of critical power lines is determined as follows: A critical power line is identified, and all power loads acquired for that critical power line within the spatiotemporal characteristic analysis period are collected. All power loads are sorted in the order of acquisition. The absolute difference between adjacent power loads after sorting is calculated to obtain the power fluctuation load. The average power fluctuation load is obtained by summing and averaging all power fluctuation loads. The system defines high and low power loads. Power loads falling between these two thresholds are marked as compliant; otherwise, they are marked as non-compliant. All compliant power loads are sorted by acquisition time. The acquisition time difference between adjacent compliant power loads is calculated to obtain the compliant load interval. The average compliant load interval is calculated by summing all compliant load intervals. ,pass The line load depth analysis value of this critical power line was calculated. Where s1 is the first coefficient and s2 is the second coefficient, with s1 having a value of 0.88 and s2 having a value of 0.54.

[0016] Compared with the prior art, the present invention has the following beneficial effects: Through the electricity customer order spatiotemporal feature prediction module, electricity customer order simulation delivery module, and electricity customer order delivery decision module, the system conducts in-depth analysis of the spatiotemporal characteristics of customer electricity consumption in the power grid based on electricity customer orders. It combines long-term and short-term neural network model technology to predict the electricity consumption characteristics of each physical area in the future and dynamically simulates the changes in electricity consumption in each physical area. It pre-sets a power allocation scheme to meet the electricity demand of each physical area and deeply analyzes the coordination of key power lines with peak shaving and valley filling under the power allocation scheme. It also combines actual analysis to determine the power system's adjustability under the power allocation scheme, ensuring the power adjustability of the power allocation scheme in the face of sudden power consumption. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the operation of an intelligent analysis system for the spatiotemporal characteristics of electricity customer orders. Figure 2 This is a block diagram of the system. Figure 3 A flowchart for determining the flexible load allocation value for the simulated physical region. Detailed Implementation

[0018] Reference Figures 1 to 3 The system is an intelligent analysis system based on the spatiotemporal characteristics of electricity customer orders, including a module for predicting the spatiotemporal characteristics of electricity customer orders, a module for simulating the delivery of electricity customer orders, and a module for making delivery decisions for electricity customer orders.

[0019] The electricity customer order spatiotemporal feature prediction module divides the power grid into several physical areas (physical areas can be residential areas, commercial areas, etc.) based on the connection relationship of transmission lines. It sets a spatiotemporal feature analysis cycle (the spatiotemporal feature analysis cycle is a preset cycle, and the length of the cycle can be adjusted according to the needs). After each spatiotemporal feature analysis cycle, it collects the electricity customer orders of each customer in the physical area, and then determines the power prediction spatiotemporal feature set of each physical area.

[0020] The spatiotemporal feature set of power forecasting for a physical region is determined as follows: a physical region is selected, all customers within that physical region are identified, the power forecasting demand feature set for each customer is determined, and the power forecasting demand feature sets for each customer are combined as a dataset to obtain the spatiotemporal feature set of power forecasting for the physical region.

[0021] A customer's electricity forecast demand feature set is determined as follows: Select a customer, determine the customer's forecast demand features for each electricity consumption characteristic (electricity consumption characteristics include electricity demand, peak and valley consumption characteristics, electricity fluctuation characteristics, etc.), combine the forecast demand features of each electricity consumption characteristic as a dataset, and obtain the customer's electricity forecast demand feature set.

[0022] The predicted demand features of a customer for a given electricity consumption characteristic are determined as follows: Select an electricity consumption characteristic, collect the customer's actual demand features for that electricity consumption characteristic over the previous n consecutive spatiotemporal feature analysis periods (each spatiotemporal feature analysis period corresponds to one actual demand feature for that electricity consumption characteristic), combine the n actual demand features for that electricity consumption characteristic as a time series dataset to obtain an actual demand time series feature set, obtain the predicted demand feature time series model for that customer for that electricity consumption characteristic, input the actual demand time series feature set into the predicted demand feature time series model, and output the predicted demand features for that customer for that electricity consumption characteristic.

[0023] A spatiotemporal feature analysis period is used to collect the actual demand features of a certain electricity consumption feature in the following way: Select a spatiotemporal feature analysis period, collect the electricity customer orders of the customer in the spatiotemporal feature analysis period, select an electricity consumption feature (electricity consumption feature includes electricity demand, peak and valley period features, electricity fluctuation features, etc.), and extract the actual demand features of the electricity consumption feature from the electricity customer orders through feature extraction.

[0024] The feature extraction method is as follows, taking electricity demand as an example: Aggregate the electricity consumption of electricity customer orders, calculate the electricity demand for each spatiotemporal feature analysis period, and take peak-valley electricity consumption characteristics as an example: Calculate the average electricity consumption by hour (or other time granularity) based on electricity customer orders. Identify the peak and valley values ​​of electricity consumption. Extract the start time, end time, and electricity consumption of peak and valley periods. Taking electricity consumption fluctuation characteristics as an example: Calculate the standard deviation and coefficient of variation of electricity consumption based on electricity customer orders. Analyze the frequency of electricity consumption changes (e.g., number of changes per hour). Extract volatility indicators (e.g., fluctuation amplitude, fluctuation frequency).

[0025] Each customer's electricity consumption characteristic corresponds to a time series model for predicted demand characteristics. Each time series model for predicted demand characteristics is built based on an independent LSTM model. For example, customer A's electricity demand and customer B's electricity demand each correspond to a time series model for predicted demand characteristics, and customer A's electricity demand and customer A's peak and valley electricity consumption characteristics each correspond to a time series model for predicted demand characteristics. This embodiment will take customer A's electricity demand and customer A's peak and valley electricity consumption characteristics as examples to illustrate the construction process of two time series models for predicted demand characteristics. The time series model for predicted demand characteristics of customer A's electricity demand... The model construction process is as follows: Construct an LSTM model, collect the actual demand time series feature sets of multiple electricity demand quantities of customer A, train the LSTM model with the actual demand time series feature sets of electricity demand quantities, assign a predicted demand feature to each actual demand time series feature set of electricity demand quantities, the predicted demand feature is the predicted electricity demand of customer A in the next spatiotemporal feature analysis period, divide the training data into training set, validation set and test set according to the set ratio of 5:2:1, train on the training set, validation set and test set, and obtain the predicted demand feature time series model of customer A's electricity demand quantity.

[0026] The process of constructing the time series model of predicted demand features for customer A's peak and valley electricity consumption characteristics is as follows: An LSTM model is constructed, and actual demand time series feature sets of multiple peak and valley electricity consumption characteristics of customer A are collected. The LSTM model is trained using these actual demand time series feature sets. A predicted demand feature is assigned to each actual demand time series feature set of peak and valley electricity consumption characteristics. The predicted demand feature is the predicted peak and valley electricity consumption characteristics of customer A in the next spatiotemporal feature analysis period. The training data is divided into a training set, a validation set, and a test set according to a set ratio of 5:2:1. The training, validation, and test sets are then used to train the time series model of predicted demand features for customer A's peak and valley electricity consumption characteristics.

[0027] Electricity customer order simulation delivery module: Based on the spatiotemporal feature set of electricity forecast for each physical region, a power delivery simulation model is constructed, and the power delivery simulation model is controlled to perform a power allocation (this power allocation corresponds to a power allocation scheme, which can meet the electricity demand of each physical region).

[0028] Based on the spatiotemporal feature set of power forecast for each physical region, a power distribution simulation model is constructed. Specifically, power simulation software is selected, and a power grid model is created within the software. According to the topology of the power grid model (including the location, connection relationship, and electrical parameters (such as voltage level, line impedance, etc.) of key power grid facilities such as substations, transmission lines, and switching stations), the power access point of each customer and the transmission line in the power grid model is set. Simultaneously, the power grid model is divided into several simulation physical regions. Based on the spatiotemporal feature set of power forecast for each physical region, the power consumption of each simulation physical region is simulated within a spatiotemporal feature analysis period, thus constructing a power distribution simulation model. The power distribution simulation model can allocate power based on the power consumption of each simulation physical region to ensure that the power demand of each simulation physical region is met (the simulation is driven by the spatiotemporal feature set of power forecast for each physical region, taking into account the changes in power consumption characteristics in time and space dimensions. It is not a simple static simulation, but can dynamically show the changes in power consumption of each physical region in different periods).

[0029] Power customer order delivery decision module: After the power allocation is completed, the allocation capacity reservation value corresponding to the power allocation is obtained, and the allocation capacity reservation threshold is set (the allocation capacity reservation threshold is a preset threshold used to compare with the allocation capacity reservation value). When the allocation capacity reservation value is greater than the allocation capacity reservation threshold, the power allocation is performed on the power grid in the next spatiotemporal feature analysis cycle; otherwise, the power delivery simulation model is controlled to perform the next power allocation.

[0030] The reserved capacity value for power allocation is obtained as follows: Before power allocation begins, the power distribution simulation model determines all critical power lines within the model (critical power lines are those that play a crucial role in the entire power grid, transmitting large-capacity power; for example, in peak shaving and valley filling scenarios, the function of critical power lines is as follows: excess power generated by residential lines during off-peak hours at night is gradually collected and transmitted through critical power lines before finally being supplied to commercial lines). The power distribution simulation model then begins power allocation for one spatiotemporal feature analysis cycle. Within this cycle, each... The system continuously acquires the power load of each key power line and the flexible load allocation values ​​of each simulated physical region within the power distribution simulation model. After power allocation is completed, it determines the line load depth analysis value of each key power line and the sudden allocation satisfaction value of each simulated physical region. A sudden allocation satisfaction threshold is set (this threshold is a preset threshold used for comparison with the sudden allocation satisfaction value). When the sudden allocation satisfaction value of a simulated physical region is greater than the sudden allocation satisfaction threshold, the simulated physical region is marked as a sudden allocation satisfaction region (otherwise, no action is taken). The average line load depth analysis value is calculated by summing and averaging the power loads of all key power lines. The total number of areas where emergency allocation is satisfied is marked as (yy), through + The reserved allocation capacity for this power distribution is calculated using (yy)*s4. Where s3 is the third coefficient and s4 is the fourth coefficient, with s3 having a value of 1.01 and s4 having a value of 1.55.

[0031] The burst allocation satisfaction value of the simulated physical region is determined as follows: a simulated physical region is defined, all load flexible allocation values ​​obtained in the spatiotemporal characteristic analysis period of the simulated physical region are collected, and the summation and average of all load flexible allocation values ​​are calculated to obtain the burst allocation satisfaction value of the simulated physical region.

[0032] The load flexibility adjustment value for the simulated physical area is determined as follows: A simulated physical area is defined, and all critical power lines connected to this area are marked as associated lines (a physical area typically connects multiple critical power lines). The power load of each associated line is obtained. All associated lines are compared pairwise, and the absolute difference between the power loads of the compared associated lines is calculated to obtain the power load difference value. The summation and averaging of all power load difference values ​​yields the load flexibility adjustment value for the simulated physical area. (Since the loads of different power lines differ at the same time, this itself provides a basis for the flexible allocation of the power system. In the event of sudden changes in power demand, such as a temporary large-scale event causing a surge in power load in a local area or extreme weather affecting power supply in some areas, the supply and demand relationship of the entire system can be balanced by adjusting the load of each power line.)

[0033] The line load depth analysis value of critical power lines is determined as follows: A critical power line is identified, and all power loads acquired for that critical power line within the spatiotemporal characteristic analysis period are collected. All power loads are sorted in the order of acquisition. The absolute difference between adjacent power loads after sorting is calculated to obtain the power fluctuation load. The average power fluctuation load is obtained by summing and averaging all power fluctuation loads. The system defines high and low power loads (high power load is higher than low power load; both are preset loads used for comparison). When a power load falls between the high and low power loads, it is marked as a compliant load; otherwise (power load greater than or equal to the high power load or less than or equal to the low power load), it is marked as a non-compliant load. All compliant loads are sorted in the order they were acquired. The acquisition time difference between adjacent compliant loads is calculated to obtain the compliant load interval. The average compliant load interval is calculated by summing all compliant load intervals. ,pass The line load depth analysis value of this critical power line was calculated. Where s1 is the first coefficient and s2 is the second coefficient, with s1 having a value of 0.88 and s2 having a value of 0.54.

[0034] Through the electricity customer order spatiotemporal feature prediction module, electricity customer order simulation delivery module, and electricity customer order delivery decision module, the system conducts in-depth analysis of the spatiotemporal characteristics of customer electricity consumption in the power grid based on electricity customer orders. It combines long-term and short-term neural network model technology to predict the electricity consumption characteristics of each physical area in the future and dynamically simulates the changes in electricity consumption in each physical area. It pre-sets a power allocation scheme to meet the electricity demand of each physical area and deeply analyzes the coordination of key power lines with peak shaving and valley filling under the power allocation scheme. It also combines actual analysis to determine the power system's adjustability under the power allocation scheme, ensuring the power adjustability of the power allocation scheme in the face of sudden power consumption.

[0035] The above formulas are all dimensionless calculations, and the preset parameters in the formulas should be set by those skilled in the art according to the actual situation.

[0036] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0037] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0038] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0039] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0040] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0041] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0042] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A spatiotemporal characteristic intelligent analysis system based on electricity customer orders, characterized in that, It includes a module for predicting the spatiotemporal characteristics of electricity customer orders, a module for simulating the delivery of electricity customer orders, and a module for making delivery decisions for electricity customer orders; The power customer order spatiotemporal feature prediction module: Based on the connection relationship of transmission lines, the power grid is divided into several physical regions, and the power prediction spatiotemporal feature set of each physical region is determined periodically; The power customer order simulation delivery module: Based on the spatiotemporal feature set of power forecast for each physical region, it constructs a power delivery simulation model and controls the power delivery simulation model to perform a power allocation; The power customer order delivery decision module: After the power allocation is completed, it obtains the allocation capacity reservation value corresponding to the power allocation, and determines whether to perform the power allocation to the power grid based on the comparison result between the allocation capacity reservation value and the allocation capacity reservation threshold.

2. The intelligent analysis system for spatiotemporal characteristics based on electricity customer orders according to claim 1, characterized in that, The spatiotemporal feature set of power forecasting for a physical region is determined as follows: a physical region is selected, all customers within that physical region are identified, the power forecasting demand feature set for each customer is determined, and the power forecasting demand feature sets for each customer are combined as a dataset to obtain the spatiotemporal feature set of power forecasting for the physical region.

3. The intelligent analysis system for spatiotemporal characteristics based on electricity customer orders according to claim 2, characterized in that, A customer's electricity forecast demand feature set is determined as follows: Select a customer, determine the customer's forecast demand features for each electricity consumption feature, combine the forecast demand features of each electricity consumption feature as a dataset, and obtain the customer's electricity forecast demand feature set.

4. The intelligent analysis system for spatiotemporal characteristics based on electricity customer orders according to claim 3, characterized in that, The predicted demand features of a customer for a specific electricity consumption characteristic are determined as follows: Select an electricity consumption characteristic, collect the customer's actual demand features for that electricity consumption characteristic over the previous n consecutive spatiotemporal characteristic analysis periods, combine the n actual demand features of that electricity consumption characteristic as a time series dataset to obtain the actual demand time series feature set, obtain the predicted demand feature time series model of the customer for that electricity consumption characteristic, input the actual demand time series feature set into the predicted demand feature time series model, and output the predicted demand features of the customer for that electricity consumption characteristic.

5. The intelligent analysis system for spatiotemporal characteristics based on electricity customer orders according to claim 4, characterized in that, A spatiotemporal feature analysis cycle collects the actual demand features of an electricity consumption feature based on the following method: a spatiotemporal feature analysis cycle is selected, the customer's electricity orders in that spatiotemporal feature analysis cycle are collected, an electricity consumption feature is selected, and the actual demand features of that electricity consumption feature are extracted from the electricity customer orders through feature extraction.

6. The intelligent analysis system for spatiotemporal characteristics based on electricity customer orders according to claim 1, characterized in that, A power allocation capacity reservation threshold is set. When the reserved power allocation capacity value is greater than the reserved power allocation capacity threshold, the power allocation is performed on the power grid in the next spatiotemporal feature analysis cycle. Otherwise, the power distribution simulation model is controlled to perform the next power allocation.

7. The intelligent analysis system for spatiotemporal characteristics based on electricity customer orders according to claim 1, characterized in that, The reserved allocation capacity value corresponding to power allocation is obtained based on the following method: Before power allocation begins, the power distribution simulation model determines all critical power lines within the model. The model then begins power allocation for one spatiotemporal feature analysis cycle. Within this cycle, each... The system continuously acquires the power load of each critical power line and the flexible load allocation values ​​of each simulated physical region within the power distribution simulation model. After power allocation is completed, it determines the line load depth analysis value of each critical power line and the sudden allocation satisfaction value of each simulated physical region. A sudden allocation satisfaction threshold is set. When the sudden allocation satisfaction value of a simulated physical region exceeds the sudden allocation satisfaction threshold, the simulated physical region is marked as a sudden allocation satisfaction region. The average line load depth analysis value is calculated by summing and averaging the power loads of all critical power lines. The total number of areas where emergency allocation is satisfied is marked as (yy), through + The reserved allocation capacity for this power distribution is calculated using (yy)*s4. , where s3 is the third coefficient and s4 is the fourth coefficient.

8. The intelligent analysis system for spatiotemporal characteristics based on electricity customer orders according to claim 7, characterized in that, The burst allocation satisfaction value of the simulated physical region is determined as follows: a simulated physical region is defined, all load flexible allocation values ​​obtained in the spatiotemporal characteristic analysis period of the simulated physical region are collected, and the summation and average of all load flexible allocation values ​​are calculated to obtain the burst allocation satisfaction value of the simulated physical region.

9. The intelligent analysis system for spatiotemporal characteristics based on electricity customer orders according to claim 7, characterized in that, The load flexibility allocation value of the simulated physical area is determined as follows: a simulated physical area is defined, and all key power lines connected to the simulated physical area are marked as associated lines. The power load of each associated line is obtained. All associated lines are compared pairwise. The absolute difference between the power loads of the two compared associated lines is calculated to obtain the power load difference value. All power load difference values ​​are summed and averaged to obtain the load flexibility allocation value of the simulated physical area.

10. The intelligent analysis system for spatiotemporal characteristics based on electricity customer orders according to claim 7, characterized in that, The line load depth analysis value of critical power lines is determined as follows: A critical power line is identified, and all power loads acquired for that critical power line within the spatiotemporal characteristic analysis period are collected. All power loads are sorted in the order of acquisition. The absolute difference between adjacent power loads after sorting is calculated to obtain the power fluctuation load. The average power fluctuation load is obtained by summing and averaging all power fluctuation loads. The system defines high and low power loads. Power loads falling between these two thresholds are marked as compliant; otherwise, they are marked as non-compliant. All compliant power loads are sorted by acquisition time. The acquisition time difference between adjacent compliant power loads is calculated to obtain the compliant load interval. The average compliant load interval is calculated by summing all compliant load intervals. ,pass The line load depth analysis value of this critical power line was calculated. Where s1 is the first coefficient and s2 is the second coefficient, with s1 having a value of 0.88 and s2 having a value of 0.54.

Images