Urban demand side novel load resource scheduling system and method

Abstract

The invention discloses a novel load resource scheduling system and method for an urban demand side, and relates to the technical field of power system management and scheduling, and the method comprises the steps: extracting user power utilization features and energy production features from preprocessed urban data; establishing a user electricity consumption prediction model based on the historical user electricity consumption data, inputting the user electricity consumption characteristics into the user electricity consumption prediction model, and predicting the user electricity consumption; based on the historical energy production data, establishing an energy production prediction model, inputting the energy production characteristics into the energy production prediction model, and predicting the energy production amount; obtaining a personalized scheduling strategy based on the user electricity consumption and the energy production; and evaluating the scheduling effect of the personalized scheduling strategy, and performing feedback adjustment according to an evaluation result. According to the invention, through multi-channel data collection and preprocessing, the accuracy of load prediction and energy production prediction is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system management and dispatching, and in particular to a new load resource dispatching system and method on the urban demand side. Background Art

[0002] With the acceleration of urbanization and the continuous growth of energy demand, the stability and efficiency of urban power systems have become key issues that need to be addressed. In recent years, the development of smart grid technology has provided new possibilities for the optimal dispatch of urban power systems. Smart grids integrate advanced sensing, communication and control technologies to achieve real-time monitoring and management of all aspects of the power system. In terms of demand-side management, existing technologies mainly focus on load forecasting, demand response and energy dispatching. Load forecasting technology uses historical data and statistical methods, such as time series analysis and machine learning algorithms, to predict future electricity demand. Demand response technology guides users to adjust their electricity consumption behavior through price signals or incentive mechanisms to achieve the effect of peak shaving and valley filling. Energy dispatching technology ensures the balance of supply and demand by optimizing the allocation of power generation resources.

[0003] However, the existing technologies still have some shortcomings in practical applications. First, traditional load forecasting models are usually based on a single data source and lack comprehensive analysis of multi-channel data, resulting in limited forecasting accuracy. Secondly, existing dispatching strategies are often too dependent on fixed rules or simple optimization algorithms, and it is difficult to cope with the complex and changeable urban power system environment. In addition, although the existing demand response mechanism can alleviate the problem of supply and demand imbalance to a certain extent, its effect is limited by the user's participation and response speed, and fails to fully consider the user's personalized needs. Therefore, how to improve the accuracy of load forecasting, formulate more flexible and efficient dispatching strategies, and take into account user satisfaction has become an important direction of current research. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a new load resource scheduling system and method on the urban demand side to solve the problems of limited accuracy of traditional load forecasting and insufficient flexibility of scheduling strategies.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In the first aspect, an embodiment of the present invention provides a new load resource scheduling method on the urban demand side, which includes collecting urban data through multiple channels and preprocessing the urban data; extracting user electricity consumption characteristics and energy production characteristics from the preprocessed urban data; establishing a user electricity consumption prediction model based on historical user electricity consumption data, inputting user electricity consumption characteristics into the user electricity consumption prediction model, and predicting user electricity consumption; establishing an energy production prediction model based on historical energy production data, inputting energy production characteristics into the energy production prediction model, and predicting energy production; obtaining a personalized scheduling strategy based on user electricity consumption and energy production; evaluating the scheduling effect of the personalized scheduling strategy, and making feedback adjustments based on the evaluation results.

[0008] As a preferred solution of the new load resource scheduling method on the urban demand side of the present invention, wherein: the urban data is collected through multiple channels and pre-processed, and the specific steps are as follows:

[0009] Collect urban data through multiple channels such as IoT devices, smart meters and weather stations. The urban data collected through multiple channels include renewable energy production data, historical energy production data, user electricity consumption data and meteorological data;

[0010] The collected urban data are preprocessed through data cleaning, format unification, outlier detection, missing value processing and data standardization.

[0011] As a preferred solution of the new load resource scheduling method on the urban demand side of the present invention, wherein: the user power consumption characteristics and energy production characteristics are extracted from the pre-processed urban data, and the specific steps are as follows:

[0012] By analyzing the user's historical electricity consumption data, identifying the electricity consumption pattern, average electricity consumption, electricity consumption time and response to price changes, the user's electricity consumption characteristics are extracted;

[0013] Energy production characteristics are extracted by analyzing the total amount, fluctuation patterns and efficiency indicators in historical power generation data and combining them with meteorological information.

[0014] As a preferred solution of the new urban demand-side load resource scheduling method of the present invention, wherein: based on the historical user electricity consumption data, a user electricity consumption prediction model is established, the user electricity consumption characteristics are input into the user electricity consumption prediction model, and the user electricity consumption is predicted. The specific steps are as follows:

[0015] Using the linear regression model as the basic model, a user electricity consumption prediction model is established;

[0016] By splitting historical user electricity consumption data into consecutive training sets, validation sets, and test sets in chronological order, training a linear regression model using the training set, evaluating the prediction accuracy of the linear regression model using the test set, and adjusting the parameters of the linear regression model on the validation set according to the evaluation results, a user electricity consumption prediction model is obtained;

[0017] Input the user electricity consumption characteristics into the user electricity consumption prediction model to predict the user electricity consumption under the current characteristics.

[0018] As a preferred embodiment of the novel load resource scheduling method for the urban demand side described in the present invention, wherein: based on historical energy production data, an energy production prediction model is established, the energy production characteristics are input into the energy production prediction model to predict the energy production volume, and the specific steps are as follows.

[0019] Use the long short-term memory network model as the basic model to establish an energy production prediction model;

[0020] Divide the historical energy production data into training sets, validation sets, and test sets according to the random division method. The long short-term memory network model adjusts its internal parameters through the training set to minimize the prediction error. During the training process, the best model configuration is selected by monitoring the performance on the validation set, and finally a comprehensive evaluation is carried out on the test set. After sufficient training and verification, an energy production prediction model is obtained;

[0021] Input the energy production characteristics into the energy production prediction model to predict the energy production volume under the current characteristics.

[0022] As a preferred embodiment of the novel load resource scheduling method for the urban demand side described in the present invention, wherein: based on the user electricity consumption and energy production volume, a personalized scheduling strategy is obtained, and the specific steps are as follows.

[0023] Through the user electricity consumption y u and the energy production volume y e , calculate the total electricity demand and total energy supply at the current time point;

[0024] According to the total user electricity demand and total energy supply at the current time point, analyze the energy supply and demand situation and formulate a personalized scheduling strategy;

[0025] When D > S, it means that the energy supply is in short supply, and energy supply should be increased and energy demand should be reduced;

[0026] When D < S, it means that the energy supply is in oversupply, and excess energy should be stored and the power generation cost should be reduced;

[0027] When D = S, it means that the energy supply and electricity demand are balanced, and power generation and electricity consumption should be monitored and fine-tuned in real time;

[0028] The personalized dispatching strategy is converted into dispatching instructions, and the dispatching instructions are sent to the resource control device for resource dispatching through the smart grid management system.

[0029] As a preferred solution of the new urban demand-side load resource scheduling method of the present invention, the scheduling effect of the personalized scheduling strategy is evaluated, and feedback adjustment is performed according to the evaluation result. The specific steps are as follows:

[0030] Evaluate the scheduling effect of the executed personalized scheduling strategy;

[0031] The actual electricity consumption, energy production and user response data are collected in real time through the smart grid management system, and the supply and demand deviation value F is recalculated using the adjusted parameters. The scheduling strategy is re-formulated based on the calculated supply and demand deviation value F, and the updated scheduling strategy is converted into scheduling instructions, which are sent to the corresponding resource control equipment through the smart grid management system to adjust the resource supply and demand balance.

[0032] In the second aspect, the present invention provides a new type of load resource scheduling system on the urban demand side, including a data collection and preprocessing module, a feature extraction module, a user electricity consumption prediction module, an energy production prediction module, a scheduling strategy formulation module and an effect evaluation and feedback adjustment module; the data collection and preprocessing module is used to collect urban data through multiple channels and preprocess the urban data; the feature extraction module is used to extract user electricity consumption characteristics and energy production characteristics from the preprocessed urban data; the user electricity consumption prediction module is used to establish a user electricity consumption prediction model based on historical user electricity consumption data, input the user electricity consumption characteristics into the user electricity consumption prediction model, and predict the user electricity consumption; the energy production prediction module is used to establish an energy production prediction model based on historical energy production data, input the energy production characteristics into the energy production prediction model, and predict the energy production; the scheduling strategy formulation module obtains a personalized scheduling strategy based on user electricity consumption and energy production; the effect evaluation and feedback adjustment module is used to evaluate the scheduling effect of the personalized scheduling strategy, and make feedback adjustments based on the evaluation results.

[0033] In a third aspect, an embodiment of the present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the new load resource scheduling method on the urban demand side as described in the first aspect of the present invention is implemented.

[0034] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the new load resource scheduling method on the urban demand side as described in the first aspect of the present invention is implemented.

[0035] The beneficial effects of the present invention are as follows: through multi-channel data collection and preprocessing, the present invention improves the accuracy of load forecasting and energy production forecasting, and enhances the robustness of the system; the user electricity consumption forecasting model and energy production forecasting model established based on historical data realize accurate electricity consumption and energy production forecasting, and maintains the timeliness and reliability of the forecast through real-time update and feedback adjustment mechanism; formulates and implements personalized scheduling strategies, which not only efficiently responds to changes in supply and demand and reduces energy waste, but also fully considers the personalized needs of users and improves user satisfaction and participation. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0037] Figure 1 This is a flow chart of the new load resource scheduling method on the urban demand side in Example 1.

[0038] Figure 2 This is a flow chart of the new load resource dispatching system on the urban demand side in Example 1. DETAILED DESCRIPTION

[0039] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0040] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides a new load resource scheduling method on the urban demand side, comprising the following steps:

[0041] S1. Collect urban data through multiple channels and pre-process the urban data.

[0042] Furthermore, urban data are collected through multiple channels such as IoT devices, smart meters and weather stations. The urban data collected through multiple channels include renewable energy production data, historical energy production data, user electricity consumption data and meteorological data. The collected urban data are pre-processed through data cleaning, format unification, outlier detection, missing value processing and data standardization.

[0043] Data cleaning, format unification, outlier detection, missing value processing and data standardization are specifically as follows: through data cleaning, duplicate records are removed, inconsistencies in timestamps and units, and erroneous data are corrected to ensure data consistency and accuracy; format unification converts timestamps into UTC time, converts numerical data into floating-point numbers, and unifies temperature and electricity units into degrees Celsius and kilowatt-hours; outlier detection uses the Z-score method and box plot method to identify and process outliers, and uses the mean, median or interpolation method to replace them; missing value processing uses linear interpolation, forward filling / backward filling, mean / median filling and machine learning model prediction to fill missing values; data standardization uses Min-Max normalization, Z-score standardization and logarithmic transformation to make the data have the same scale, thereby ensuring that the collected urban data is of high quality and good consistency, providing a reliable foundation for subsequent feature extraction, model building and scheduling strategy formulation.

[0044] S2. Extract user electricity consumption characteristics and energy production characteristics from the preprocessed urban data.

[0045] Furthermore, by analyzing the historical electricity consumption data of users, the electricity consumption patterns, average electricity consumption, electricity consumption time and response to price changes are identified to extract the characteristics of user electricity consumption; by analyzing the total amount, fluctuation pattern and efficiency index in the historical power generation data and combining with meteorological information, the characteristics of energy production are extracted;

[0046] Analyze historical data to extract user electricity consumption characteristics and energy production characteristics. Specifically, by calculating the average electricity consumption and the proportion of electricity consumption in different time periods, and quantifying the user's response to price changes through regression analysis, we can extract average electricity consumption, peak hour electricity consumption, electricity consumption pattern category and price response coefficient; at the same time, by using time series analysis and regression analysis and other methods, we can calculate the total power generation, standard deviation of power generation, power generation efficiency and the influence coefficient of meteorological factors.

[0047] S3. Based on historical user electricity consumption data, a user electricity consumption prediction model is established, and user electricity consumption characteristics are input into the user electricity consumption prediction model to predict user electricity consumption.

[0048] Furthermore, a user electricity consumption prediction model is established using the linear regression model as the basic model;

[0049] The historical user electricity consumption data is divided into a continuous training set, a validation set and a test set in chronological order, the linear regression model is trained with the training set, the prediction accuracy of the linear regression model is evaluated with the test set, and the linear regression model parameters are adjusted on the validation set according to the evaluation results to obtain the user electricity consumption prediction model;

[0050] Input the user's electricity consumption characteristics into the user's electricity consumption prediction model to predict the user's electricity consumption under the current characteristics. The expression is:

[0051]

[0052] Among them, y u is the predicted user electricity consumption, β0 is the intercept term, n is the number of user electricity consumption features, β i represents the influence of the i-th user's electricity consumption characteristics on the user's electricity consumption, x ui is the user's electricity consumption feature vector x u The i-th eigenvalue in i The adjustment coefficient of the electricity consumption characteristics of the i-th user;

[0053] The calculation process of this expression is as follows: first, calculate the linear combination of the features and the corresponding weight coefficients; then, calculate the linear combination of the features and the corresponding weight coefficients and use it as the negative value of the exponential function; then, substitute this negative value into the exponential function and add 1 to get the denominator; finally, divide the linear combination obtained in the first step by the denominator and add the intercept term to get the final predicted value;

[0054] It should also be noted that β i It represents the influence of the electricity consumption characteristics of the i-th user on the user's electricity consumption, and the expression is:

[0055]

[0056] Linear regression is used as the basic model. Its advantages are: the simple and intuitive mathematical form, easy implementation and explanation, fast training speed and computational efficiency of the linear regression model make it a suitable basic model for user electricity consumption prediction. It can effectively capture the linear relationship between features and electricity consumption and provide reasonable prediction results.

[0057] S4. Based on historical energy production data, an energy production prediction model is established, and energy production characteristics are input into the energy production prediction model to predict energy production.

[0058] Furthermore, the long short-term memory network (LSTM) model is used as the basic model to establish an energy production prediction model;

[0059] The historical energy production data is divided into training set, validation set and test set according to the random partitioning method. The long short-term memory network (LSTM) model adjusts its internal parameters through the training set to minimize the prediction error. During the training process, the best model configuration is selected by monitoring the performance on the validation set. Finally, a comprehensive evaluation is performed on the test set. After sufficient training and verification, the energy production prediction model is obtained.

[0060] The energy production characteristics are input into the energy production prediction model to predict the energy production under the current characteristics. The expression is:

[0061] y e =tanh(W·h t +U·x et + b);

[0062] Among them, y e is the predicted energy production, W is the hidden state h t The weight matrix, U is the output feature x et The weight matrix, b is the bias term, x et is the energy production characteristic vector X e The eigenvalue at time step t, h t is the hidden state of the energy production forecasting model at time step t.

[0063] The calculation process of this expression is as follows: first, the matrix product of the hidden state and the weight matrix is ​​calculated; then, the matrix product of the input features and the weight matrix is ​​calculated; then, these two results are added together and the bias term is added to get an intermediate value; finally, this intermediate value is substituted into the hyperbolic tangent function to get the final output;

[0064] It should also be noted that the advantages of using the long short-term memory network (LSTM) as the basic model are: the long short-term memory network (LSTM) model can effectively process data with time dependence and seasonality, automatically model complex nonlinear relationships, adapt to a variety of input features, and provide high-precision predictions. It can capture long-term dependencies through its special gating mechanism and memory units, support end-to-end learning, and can be used in combination with other deep learning techniques, showing strong flexibility and broad application prospects.

[0065] S5. Obtain personalized scheduling strategies based on user electricity consumption and energy production.

[0066] Furthermore, through the user's electricity consumption u and energy production y e , calculate the total electricity demand and total energy supply at the current time point, the expression is:

[0067]

[0068] where y u is the user's electricity consumption, y e is the energy production, D is the total electricity demand, S is the total energy supply, m is the total number of users, and k is the number of energy production units;

[0069] Analyze the energy supply and demand situation based on the total electricity demand of users and the total energy supply at the current time point, and formulate personalized scheduling strategies;

[0070] When D > S, it means that the energy supply is in short supply, and energy supply should be increased and energy demand should be reduced;

[0071] When D < S, it means that the energy supply is in oversupply, and excess energy should be stored and the power generation cost should be reduced;

[0072] When D = S, it means that the energy supply and electricity demand are balanced, and power generation and electricity consumption should be monitored in real time and fine-tuned;

[0073] Convert the personalized scheduling strategy into scheduling instructions, and through the smart grid management system, send the scheduling instructions to the resource control equipment for resource scheduling;

[0074] Resource scheduling specifically includes adjusting electricity prices, shutting down some power generation equipment, selling excess energy, fine-tuning renewable energy production, and fine-tuning user electricity consumption;

[0075] The calculation process of this expression is as follows: First, for the calculation of the total electricity demand, sum the electricity consumption values of all users from 1 to the total number of users m, and the resulting value is the total electricity demand; then, for the calculation of the total energy supply, sum the energy production values of all energy production units from 1 to the number of energy production units k, and the resulting value is the total energy supply;

[0076] Sending the scheduling instructions to the resource control equipment for resource scheduling through the smart grid management system is specifically as follows: The smart grid management system sends these instructions to each resource control equipment (such as power generation equipment controllers, energy storage system controllers, demand response equipment, renewable energy controllers) through existing communication networks (such as power line carrier communication, wireless communication, etc.) and standardized protocols (such as IEC 61850, DNP3). After receiving the instructions, the resource control equipment performs corresponding operations. At the same time, the smart grid management system continuously monitors the operation status of the power grid, collects real-time data, and makes real-time adjustments and emergency responses as needed to ensure the stable operation of the power grid and the balance of energy supply and demand.

[0077] S6. Evaluate the scheduling effect of the personalized scheduling strategy and make feedback adjustments according to the evaluation results.

[0078] Furthermore, evaluate the scheduling effect of the executed personalized scheduling strategy. The expression is

[0079]

[0080] Among them, F represents the supply-demand deviation value, σ represents the standard deviation of the supply-demand balance, m is the total number of users, u is the user index variable, λ is the weight coefficient of user satisfaction relative to the supply-demand balance, and y u,t represents the user's expected electricity consumption, and γ represents the allowable deviation of the user's electricity consumption;

[0081] The actual electricity consumption, energy production and user response data are collected in real time through the smart grid management system, and the supply and demand deviation value F is recalculated using the adjusted parameters. The scheduling strategy is re-formulated based on the calculated supply and demand deviation value F, and the updated scheduling strategy is converted into scheduling instructions, which are sent to the corresponding resource control equipment through the smart grid management system to adjust the resource supply and demand balance.

[0082] The calculation process of this expression is as follows: first, calculate the square of the difference between the total electricity demand and the total energy supply, and divide it by a constant to get the first part; then, for each user from 1 to the total number of users m, calculate the square of the difference between their actual electricity consumption and the target electricity consumption, divide it by another constant, and then sum these results to get the second part; then, multiply the second part by a regularization parameter; finally, add the first part and the adjusted second part to obtain the minimized supply and demand difference and the deviation value between the user's electricity consumption and the target electricity consumption.

[0083] This embodiment also provides a new type of urban demand-side load resource scheduling system, including: a data collection and preprocessing module, a feature extraction module, a user power consumption prediction module, an energy production prediction module, a scheduling strategy formulation module, and an effect evaluation and feedback adjustment module;

[0084] Data collection and preprocessing module, used to collect urban data through multiple channels and preprocess the urban data;

[0085] A feature extraction module is used to extract user electricity consumption characteristics and energy production characteristics from the preprocessed urban data;

[0086] A user power consumption prediction module is used to establish a user power consumption prediction model based on historical user power consumption data, input user power consumption characteristics into the user power consumption prediction model, and predict user power consumption;

[0087] The energy production prediction module is used to establish an energy production prediction model based on historical energy production data, input energy production characteristics into the energy production prediction model, and predict energy production;

[0088] The scheduling strategy formulation module obtains personalized scheduling strategies based on user power consumption and energy production;

[0089] The effect evaluation and feedback adjustment module is used to evaluate the scheduling effect of the personalized scheduling strategy and make feedback adjustments based on the evaluation results.

[0090] This embodiment also provides a computer device, which is suitable for the case of a new type of load resource scheduling method on the urban demand side, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the new type of load resource scheduling method on the urban demand side proposed in the above embodiment.

[0091] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be realized through WI FI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0092] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for implementing a new type of load resource scheduling on the urban demand side proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, disk or optical disk.

[0093] In summary, the present invention improves the accuracy of load forecasting and energy production forecasting and enhances the robustness of the system through multi-channel data collection and preprocessing; the user electricity consumption forecasting model and energy production forecasting model established based on historical data realize accurate electricity consumption and energy production forecasting, and maintains the timeliness and reliability of the forecast through real-time update and feedback adjustment mechanism; formulates and implements personalized scheduling strategies, which not only efficiently responds to changes in supply and demand and reduces energy waste, but also fully considers the personalized needs of users and improves user satisfaction and participation.

Claims

1. A new load resource scheduling method on the urban demand side, characterized by: include, Collect urban data through multiple channels and pre-process the urban data; Extract user electricity consumption characteristics and energy production characteristics from the preprocessed urban data; Based on historical user electricity consumption data, a user electricity consumption prediction model is established, and user electricity consumption characteristics are input into the user electricity consumption prediction model to predict user electricity consumption; Based on historical energy production data, an energy production forecasting model is established, and energy production characteristics are input into the energy production forecasting model to predict energy production; Obtain personalized scheduling strategies based on user electricity consumption and energy production; Evaluate the scheduling effect of personalized scheduling strategies and make feedback adjustments based on the evaluation results.

2. The new urban demand-side load resource scheduling method according to claim 1, characterized in that: The city data is collected through multiple channels and preprocessed. The specific steps are as follows: Collect urban data through multiple channels such as IoT devices, smart meters and weather stations. The urban data collected through multiple channels include renewable energy production data, historical energy production data, user electricity consumption data and meteorological data; The collected urban data are preprocessed through data cleaning, format unification, outlier detection, missing value processing and data standardization.

3. The new urban demand-side load resource scheduling method according to claim 2, characterized in that: The specific steps of extracting user electricity consumption characteristics and energy production characteristics from the pre-processed urban data are as follows: By analyzing the user's historical electricity consumption data, identifying the electricity consumption pattern, average electricity consumption, electricity consumption time and response to price changes, the user's electricity consumption characteristics are extracted; Energy production characteristics are extracted by analyzing the total amount, fluctuation patterns and efficiency indicators in historical power generation data and combining them with meteorological information.

4. The new urban demand-side load resource scheduling method according to claim 3, characterized in that: The user electricity consumption prediction model is established based on the historical user electricity consumption data, and the user electricity consumption characteristics are input into the user electricity consumption prediction model to predict the user electricity consumption. The specific steps are as follows: Using the linear regression model as the basic model, a user electricity consumption prediction model is established; The historical user electricity consumption data is divided into a continuous training set, a validation set and a test set in chronological order, the linear regression model is trained with the training set, the prediction accuracy of the linear regression model is evaluated with the test set, and the linear regression model parameters are adjusted on the validation set according to the evaluation results to obtain the user electricity consumption prediction model; The user's electricity consumption characteristics are input into the user's electricity consumption prediction model to predict the user's electricity consumption under the current characteristics.

5. The new urban demand-side load resource scheduling method according to claim 4, characterized in that: Based on the historical energy production data, an energy production prediction model is established, energy production characteristics are input into the energy production prediction model, and energy production is predicted. The specific steps are as follows: Using the long short-term memory network model as the basic model, an energy production prediction model is established; The historical energy production data is divided into training set, validation set and test set according to the random partitioning method. The long short-term memory network model adjusts its internal parameters through the training set to minimize the prediction error. During the training process, the best model configuration is selected by monitoring the performance on the validation set. Finally, a comprehensive evaluation is performed on the test set. After sufficient training and verification, the energy production prediction model is obtained. The energy production characteristics are input into the energy production prediction model to predict the energy production under the current characteristics.

6. The new urban demand-side load resource scheduling method according to claim 5, characterized in that: The specific steps of obtaining a personalized scheduling strategy based on user power consumption and energy production are as follows: By user power consumption u and energy production y e , calculate the total electricity demand and total energy supply at the current time point; Analyze the energy supply and demand situation based on the total electricity demand of users and the total energy supply at the current time point, and formulate personalized scheduling strategies; When D > S, it indicates that the energy supply is in short supply, and energy supply should be increased and energy demand should be reduced; When D < S, it indicates that the energy supply is in oversupply, and surplus energy should be stored and the power generation cost should be reduced; When D = S, it indicates that the energy supply and electricity demand are balanced, and power generation and electricity consumption should be monitored and finely adjusted in real time; Convert the personalized scheduling strategy into a scheduling instruction, and through the smart grid management system, send the scheduling instruction to the resource control device for resource scheduling.

7. The new urban demand-side load resource scheduling method according to claim 6, characterized in that: Evaluate the scheduling effect of the personalized scheduling strategy, and make feedback adjustments according to the evaluation results. The specific steps are as follows: Evaluate the scheduling effect of the executed personalized scheduling strategy; Collect actual electricity consumption, energy production volume, and user response data in real time through the smart grid management system, recalculate the supply-demand deviation value F using the adjusted parameters, formulate a new scheduling strategy based on the recalculated supply-demand deviation value F, convert the adjusted scheduling strategy into a scheduling instruction, and send it through the smart grid management system to the corresponding resource control device to adjust the balance of resource supply and demand.

8. A new type of load resource dispatching system on the urban demand side, based on the new type of load resource dispatching method on the urban demand side according to any one of claims 1 to 7, characterized in that: It includes a data collection and preprocessing module, a feature extraction module, a user electricity consumption prediction module, an energy production prediction module, a scheduling strategy formulation module, and an effect evaluation and feedback adjustment module; The data collection and preprocessing module is used to collect urban data through multiple channels and preprocess the urban data; The feature extraction module is used to extract user electricity consumption characteristics and energy production characteristics from the preprocessed urban data; The user electricity consumption prediction module is used to establish a user electricity consumption prediction model based on historical user electricity consumption data, input the user electricity consumption characteristics into the user electricity consumption prediction model, and predict the user electricity consumption; The energy production prediction module is used to establish an energy production prediction model based on historical energy production data, input the energy production characteristics into the energy production prediction model, and predict the energy production volume; The scheduling strategy formulation module obtains a personalized scheduling strategy based on the user electricity consumption and the energy production volume; The effect evaluation and feedback adjustment module is used to evaluate the scheduling effect of the personalized scheduling strategy and make feedback adjustments according to the evaluation results.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, it implements the steps of the new load resource scheduling method for the urban demand side according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the new load resource scheduling method for the urban demand side according to any one of claims 1 to 7.

Images