An intelligent optimization and big data analysis system of an energy power system

Abstract

The application provides an intelligent optimization and big data analysis system of an energy power system, which comprises a collection and monitoring module, an analysis and prediction module, an optimization and management module and a user management module; the collection and monitoring module is used for obtaining operation data at each node in the energy power system; the analysis and prediction module is used for predicting and analyzing the load demand and supply of each node in the power system according to the obtained operation data; the optimization and management module is used for analyzing the power supply abnormality of each node in the power system according to the output result of the analysis and prediction module; and the user management module is used for completing the interaction between the management personnel and the coefficient data; through real-time data collection, prediction analysis and optimization management, the application effectively improves the supply and demand balance control ability of the power system and enhances the intelligence and stability of the power system operation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of digital data processing, in particular to an intelligent optimization and big data analysis system for energy power systems. BACKGROUND

[0002] In modern energy power systems, with the increasing demand for electricity and the growing penetration of renewable energy, the supply and demand balance and stable operation of the power system face unprecedented challenges. The power grid not only needs to meet the real-time load demand, but also needs to cope with the uncertainty brought by the volatility of renewable energy. At the same time, the complex power network structure and multi-node, multi-type power supply equipment further increase the difficulty of supply and demand coordination. Therefore, accurate load forecasting and supply analysis, real-time monitoring, and early warning mechanisms have become key requirements for the intelligent management of power grids. Although existing technologies have improved in data analysis and fault detection, they still lack dynamic prediction of future power load and supply and potential risk warning, and there is an urgent need for a more intelligent power system optimization and big data analysis method to ensure the efficient and reliable operation of the power system.

[0003] After reviewing relevant published technical solutions, the technical solution disclosed in CN118469126A proposes an energy power data analysis method, system, and medium, aiming to solve the problem that the generation and equipment failure of the power grid still need to be determined by data and manual coordination. The method includes: obtaining original energy power data under multiple power simulation states, and performing initial classification and processing on the original energy power data; determining the overall data characteristics of each type of data according to an aggregation algorithm; determining the index weight of each type of data according to a combination weighting method; calculating and analyzing the generation and equipment failure in a specific scenario according to the index weight and overall data characteristics of the energy power data; after processing the original energy power data, using the aggregation algorithm and combination weighting method to generate feature data and analyze the generation and equipment failure in a specific scenario, significantly improving the lean management level and operation and maintenance efficiency of the power grid. However, this solution mainly relies on the aggregation algorithm and combination weighting method for static analysis of generation and failure, lacking dynamic prediction capability for future load demand and supply. In addition, this solution mainly analyzes equipment failure and generation, and fails to provide an early warning mechanism to address potential supply and demand imbalance. SUMMARY

[0004] The present application relates to the technical field of digital data processing, in particular to an intelligent optimization and big data analysis system for energy power systems.

[0005] The present application adopts the following technical solutions:

[0006] The application discloses an intelligent optimization and big data analysis system of an energy power system, and belongs to the field of power system optimization and analysis.

[0007] The collection and monitoring module is used for acquiring operation data at each node in the energy power system; the analysis and prediction module is used for predicting and analyzing the load demand and supply amount of each node in the power system according to the acquired operation data; the optimization and management module is used for analyzing the power supply abnormality of each node in the power system according to the output result of the analysis and prediction module; and the user management module is used for completing the interaction between the management personnel and the coefficient data.

[0008] The collection and monitoring module comprises a data acquisition unit, a data transmission unit and a storage management unit; the data acquisition unit is used for acquiring operation data from each node in the energy power system, wherein the operation data comprises power load demand data, power supply amount data, environmental data and other data of key influencing factors of power demand and supply of each node; the data transmission unit is used for completing the data transmission of the operation data; and the storage management unit is used for receiving the operation data transmitted from the data transmission unit and performing storage management on the operation data according to a time line.

[0009] Further, the analysis and prediction module comprises a data preprocessing unit, a demand prediction unit and a power generation prediction unit; the data preprocessing unit is used for performing preprocessing operation on the operation data transmitted by the collection and monitoring module, wherein the preprocessing operation comprises data cleaning and feature extraction processing; the demand prediction unit is used for predicting the power load demand of the energy power system in future time; and the power generation prediction unit is used for predicting the power supply amount of the energy power system in future time.

[0010] Further, the specific work flow of the demand prediction unit is as follows.

[0011] S11: according to the periodic variation law of the power load demand, a demand prediction period is set, the demand prediction period is a prediction time period of future power demand with the current time as a starting point, and operation data of a plurality of continuous demand reference periods in a time period corresponding to the demand prediction period in historical operation data is acquired.

[0012] S12: an initial load prediction model is established, the initial load prediction model is used for completing the preliminary prediction of the power load demand in the demand prediction period, and the initial load prediction model is acquired through historical data training; for each node in the power system, the specific representation of the initial load prediction model is as follows.

[0013] .

[0014] wherein, For the initial load forecasting model at the forecast time The predicted time is the time when the electricity load demand is expected to be met. This refers to a specific point in time within the demand forecasting period. For the initial load forecasting model constants, representing the model's bias parameters, which are obtained through training with historical data; The number of historical demand reference periods selected for the initial load forecasting model; These are autoregressive parameters, representing the prediction time. The previous The impact of electricity load demand in a reference period on the current predicted electricity load demand is obtained through training with historical data. For predicting time The previous The corresponding forecast time within the demand reference period Reference weight at time, For predicting time The previous The corresponding forecast time within the demand reference period Power load demand at that time; This is a random error term, obtained through model training and fitting.

[0015] S13: Establish an impact model of external data on electricity load demand based on historical external data; the external data includes environmental data and other data that have key influencing factors on electricity demand.

[0016] S14: Obtain external data information for the demand forecasting period from third-party data sources.

[0017] S15: Combine external data to complete the final forecast of electricity load demand.

[0018] Furthermore, the specific workflow of the power generation prediction unit is as follows.

[0019] S21: Based on the cyclical change pattern of power supply, a supply forecast period is set, which is the forecast period of future power supply starting from the current time; and the operating data of multiple consecutive supply reference periods corresponding to the supply forecast period are obtained from historical operating data.

[0020] S22: Establish an initial supply forecasting model, which is used to make a preliminary forecast of power supply within the supply forecasting period; the initial supply forecasting model is obtained through training on historical data; for each node in the power system, the specific representation of the initial supply forecasting model is as follows.

[0021] .

[0022] wherein, is the initial supply prediction model of the power supply amount at the prediction time , the prediction time is a certain time point within the supply prediction period; is the initial supply prediction model constant term, indicating the bias parameter of the model, obtained by training the historical data; is the number of historical supply reference periods selected by the initial supply prediction model; is the autoregressive parameter, indicating the influence of the power supply amount of the supply reference period before the prediction time on the current predicted power supply amount, obtained by training the historical data; is the reference weight at the prediction time in the supply reference period before the prediction time ; is the power supply amount at the prediction time in the supply reference period before the prediction time ; is the random error term, obtained by fitting the model training.

[0023] S23: An influence model of external data on power load supply is established according to historical external data; the external data includes environmental data and other data related to key influencing factors of power supply.

[0024] S24: External data information within the supply prediction period is obtained from a third-party data source.

[0025] S25: The final prediction of power load supply is completed in combination with the external data.

[0026] Further, the optimization management module includes a supply-demand matching unit and a power supply adjustment unit; the supply-demand matching unit is used to correspondingly match the power load demand predicted at a future time of each node in the power system with the power supply amount, and when the supply-demand difference is too large, a warning reminder is sent to the management personnel; the power supply adjustment unit is used to issue a prediction reminder to the management personnel for power supply abnormalities of a specific node in combination with historical power supply and power generation conditions.

[0027] The beneficial effects obtained by the present application are:

[0028] The application can accurately capture the periodic load characteristics and supply fluctuation trend by establishing a prediction model of future power demand and supply, and combining multiple historical reference periods for preliminary supply and demand analysis; the adaptability of the model to external changes is further enhanced by combining the preliminary analysis results with environmental and key influencing factors obtained from third-party data sources; by combining the predicted power load demand and power supply amount, the supply and demand balance status of future time is dynamically evaluated, potential supply and demand imbalance nodes are identified in time and early warning is issued, thereby effectively improving the response speed and operation stability of the power system. BRIEF DESCRIPTION OF DRAWINGS

[0029] The application can be further understood from the following description made with reference to the accompanying drawings. The components in the drawings are not necessarily drawn to scale, but emphasis is instead placed on illustrating the principles of the embodiments. In different views, the same reference numerals designate corresponding parts.

[0030] Figure 1 The figure is a schematic diagram of the overall module of the application.

[0031] Figure 2 The figure is a schematic diagram of the working process of the demand prediction unit of the application.

[0032] Figure 3 The figure is a schematic diagram of the working process of the power generation prediction unit of the application.

[0033] Figure 4 The figure is a schematic diagram of the working process of the power supply adjustment unit of the application. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical scheme and advantages of the application more clear, the application will be further described in detail below in combination with the embodiments thereof; it should be understood that the specific embodiments described herein are only used to explain the application, and are not used to limit the application; for those skilled in the art, after reading the following detailed description, other systems, methods and / or features of the embodiments will become apparent; all such additional systems, methods, features and advantages are intended to be included within the present specification; included within the scope of the application, and protected by the appended claims; additional features of the disclosed embodiments are described in the following detailed description, and will be apparent from the following detailed description.

[0035] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or component referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present patent. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0036] Example 1:

[0037] like Figure 1 As shown in the figure, this embodiment provides an intelligent optimization and big data analysis system for energy and power systems. The system includes a data acquisition and monitoring module, an analysis and prediction module, an optimization management module, and a user management module.

[0038] The data acquisition and monitoring module is used to acquire operational data at each node in the energy and power system; the analysis and prediction module is used to predict and analyze the load demand and supply of each node in the power system based on the acquired operational data; the optimization and management module is used to analyze power supply anomalies at each node in the power system based on the output results of the analysis and prediction module; and the user management module is used to facilitate interaction between management personnel and coefficient data.

[0039] The data acquisition and monitoring module includes a data acquisition unit, a data transmission unit, and a storage management unit. The data acquisition unit is used to acquire operational data from various nodes in the energy and power system. The operational data includes power load demand data, power supply data, environmental data, and other data that have a key impact on power demand and supply at each node. The data transmission unit is used to transmit the operational data. The storage management unit is used to receive the operational data transmitted from the data transmission unit and store and manage the operational data according to a timeline.

[0040] The analysis and prediction module includes a data preprocessing unit, a demand prediction unit, and a power generation prediction unit. The data preprocessing unit is used to preprocess the operational data input from the acquisition and monitoring module, and the preprocessing operations include data cleaning and feature extraction. The demand prediction unit is used to predict the power load demand of the energy and power system in the future. The power generation prediction unit is used to predict the power supply of the energy and power system in the future.

[0041] Furthermore, such as Figure 2 As shown, the specific workflow of the demand forecasting unit is as follows:

[0042] S11: According to the periodic variation law of the power load demand, a demand prediction period is set, which is a prediction time period of future power demand with the current time as the starting point; and the operation data of a plurality of continuous demand reference periods corresponding to the time period of the demand prediction period in the historical operation data is obtained.

[0043] S12: An initial load prediction model is established, which is used to complete the preliminary prediction of the power load demand in the demand prediction period; the initial load prediction model is obtained by historical data training; for each node in the power system, the specific representation of the initial load prediction model is as follows:

[0044] ;

[0045] Wherein, is the power load demand of the initial load prediction model at the prediction time , the prediction time is a certain time point in the demand prediction period; is the constant term of the initial load prediction model, which represents the bias parameter of the model and is obtained by historical data training; is the number of historical demand reference periods selected by the initial load prediction model; is the autoregressive parameter, which represents the influence of the power load demand of the th demand reference period before the prediction time on the current predicted power load demand, which is obtained by historical data training; is the reference weight at the prediction time in the th demand reference period before the prediction time ; is the power load demand at the prediction time in the th demand reference period before the prediction time ; is a random error term, which is obtained by model training fitting.

[0046] For satisfies:

[0047] ;

[0048] Wherein, is the initial decay coefficient of the demand weight, which is used to control the initial value variation of the demand reference weight, and is set by pre-experiment; is the exponential decay rate of the demand weight, which is used to control the decay speed of the demand reference weight, and is set by pre-experiment.

[0049] S13: Establishing an external data impact model for power load demand based on historical external data; the external data includes environmental data and other data that are key influencing factors for power demand.

[0050] S14: Obtaining external data information within the demand prediction period from a third-party data source.

[0051] S15: Combining external data to complete the final prediction of power load demand:

[0052] ;

[0053] wherein, is the final predicted power load demand at prediction time . is the external data impact factor at prediction time , obtained by inputting the external data information obtained from the third-party data source into the external data impact model for power load demand.

[0054] Further, in this embodiment, part of the code for realizing the function of the demand prediction unit is as follows:

[0055] import numpy as np

[0056] import pandas as pd

[0057] class DemandPredictionUnit:

[0058] def __init__(self, alpha, beta, historical_data, external_data_model):

[0059] """

[0060] Initialize the demand prediction unit.

[0061] :param alpha: initial demand weight decay coefficient

[0062] :param beta: demand weight exponential decay coefficient

[0063] :param historical_data: historical power load data (DataFrame, each row is a time point)

[0064] :param external_data_model: external data model object, used to generate external impact factors

[0065] """

[0066] self.alpha = alpha

[0067] self.beta = beta

[0068] self.historical_data = historical_data

[0069] self.external_data_model = external_data_model

[0070] def calculate_weight(self, i):

[0071] """

[0072] Computes the weight ω_(t-i) for the i-th demand reference period.

[0073] :param i: The i-th demand reference period

[0074] :return: The weight value

[0075] """

[0076] return 1 / (1 + self.alpha * np.exp(self.beta * i))

[0077] def initial_demand_prediction(self, forecast_period, P):

[0078] """

[0079] Predicts the future electricity demand based on the initial load forecasting model.

[0080] :param forecast_period: List of time points within the forecast period

[0081] :param P: Number of historical demand reference periods to use

[0082] :return: Preliminary predicted electricity load demand values

[0083] """

[0084] predictions = []

[0085] for t in forecast_period:

[0086] y_hat_t = 0

[0087] for i in range(1, P + 1):

[0088] # Get historical demand data

[0089] y_t_minus_i = self.historical_data.loc[t - i]

[0090] # Autoregressive coefficient (can be set according to actual conditions; assumed to be 1 in this example)

[0091] phi_i = 1

[0092] # Calculate reference weights

[0093] omega_t_minus_i = self.calculate_weight(i)

[0094] # Accumulate to the prediction model

[0095] y_hat_t += phi_i * omega_t_minus_i * y_t_minus_i

[0096] # Model bias term (C), assumed to be a constant 0 here.

[0097] C = 0

[0098] # Random error term (ε_t), assumed to be a constant 0.

[0099] epsilon_t = 0

[0100] y_hat_t += C + epsilon_t

[0101] predictions.append(y_hat_t)

[0102] return predictions

[0103] def apply_external_factors(self, initial_predictions, forecast_period):

[0104] """

[0105] By combining external data influencing factors, the final load forecast is generated.

[0106] :param initial_predictions: List of initial load prediction values

[0107] :param forecast_period: List of time points within the forecast period

[0108] :return: List of final load prediction values

[0109] """

[0110] final_predictions = []

[0111] for idx, t in enumerate(forecast_period):

[0112] # Get external influence factors

[0113] EIF_t = self.external_data_model.get_external_influence(t)

[0114] # Calculate final prediction values

[0115] y_hat_t_final = initial_predictions[idx] * (1 + EIF_t)

[0116] final_predictions.append(y_hat_t_final)

[0117] return final_predictions

[0118] # Usage example

[0119] # Assume historical load demand data (time series DataFrame)

[0120] historical_data = pd.Series({

[0121] -3: 100, -2: 110, -1: 105 # Historical demand data for the last 3 time points

[0122] })

[0123] # Example of external data influence model

[0124] class ExternalDataModel:

[0125] def get_external_influence(self, t):

[0126] # Assume external influence factor is generated based on time (example values)

[0127] return 0.05 # Example fixed influence factor

[0128] # Initialize demand prediction unit

[0129] alpha = 0.5

[0130] beta = 0.1

[0131] forecast_period = [0, 1, 2] # List of future time points to predict

[0132] P = 3 # Number of historical demand reference periods to use

[0133] external_data_model = ExternalDataModel()

[0134] # Instantiate the prediction unit

[0135] demand_predictor = DemandPredictionUnit(alpha, beta, historical_data, external_data_model)

[0136] # Preliminary prediction of load demand

[0137] initial_predictions = demand_predictor.initial_demand_prediction(forecast_period, P)

[0138] # Combine external factors to generate final predictions

[0139] final_predictions = demand_predictor.apply_external_factors(initial_predictions, forecast_period)

[0140] # Output final prediction results

[0141] print("Preliminary predicted load demand:", initial_predictions)

[0142] print("Final predictions:", final_predictions).

[0143] Further, as shown in Figure 3 , the specific workflow of the power generation prediction unit is as follows:

[0144] S21: According to the periodic variation law of power supply, set the supply prediction period, which is the prediction time period of future power supply starting from the current time; and obtain the operation data of multiple continuous supply reference periods corresponding to the time period of the supply prediction period in the historical operation data.

[0145] S22: Establish an initial supply prediction model, which is used to complete the preliminary prediction of power supply in the supply prediction period; the initial supply prediction model is obtained by historical data training; for each node in the power system, the specific representation of the initial supply prediction model is as follows:

[0146] ;

[0147] wherein, is the power supply amount of the initial supply prediction model at the prediction time , the prediction time is a certain time point in the supply prediction period; is the constant term of the initial supply prediction model, representing the bias parameter of the model, which is obtained by historical data training; is the number of historical supply reference periods selected by the initial supply prediction model; is the autoregressive parameter, representing the influence of the power supply amount of the th supply reference period before the prediction time on the current predicted power supply amount, which is obtained by historical data training; is the reference weight at the prediction time in the th supply reference period before the prediction time ; is the power supply amount at the prediction time in the th supply reference period before the prediction time ; is a random error term, which is obtained by model training fitting.

[0148] For , it satisfies:

[0149] ;

[0150] Wherein, is the initial decay coefficient of the supply weight, used to control the variation of the initial value of the supply reference weight, set by pre-experiment; is the exponential decay rate of the supply weight, used to control the decay speed of the supply reference weight, set by pre-experiment.

[0151] S23: Establish an external data impact model for power load supply according to historical external data; the external data includes environmental data and other data related to key factors affecting power supply.

[0152] S24: Obtain external data information in the supply prediction period from a third-party data source.

[0153] S25: Complete the final prediction of power load supply in combination with external data:

[0154] ;

[0155] Wherein, is the final predicted power load supply amount at the prediction time . is the external data impact factor at the prediction time , obtained by inputting the external data information obtained from the third-party data source into the external data impact model for power load supply.

[0156] The present scheme sets a demand prediction period and a supply prediction period, and the demand prediction unit and the power generation prediction unit can accurately capture the periodic variation characteristics of future power demand and supply; each initial prediction model introduces a nonlinear decay reference weight, thereby improving the sensitivity to recent data in the prediction of current load and supply, and combining the environment and key factors of the third-party data source, the model can flexibly adjust the prediction result to adapt to the influence of sudden changes and environmental factors on power demand and supply; thereby the system can obtain higher prediction accuracy, and further support efficient scheduling and resource optimization of the power system.

[0157] Embodiment two:

[0158] The present embodiment should be understood as at least containing all the features of any one of the preceding embodiments, and further improving on the basis thereof;

[0159] The present embodiment provides an intelligent optimization and big data analysis system for an energy power system, which comprises a collection and monitoring module, an analysis and prediction module, an optimization and management module, and a user management module.

[0160] The collection monitoring module is configured to acquire operation data of each node in the energy power system; the analysis and prediction module is configured to perform prediction analysis on the load demand and supply amount of each node in the power system according to the acquired operation data; the optimization management module is configured to analyze the power supply abnormality of each node in the power system according to the output result of the analysis and prediction module; and the user management module is configured to complete the interaction between the management personnel and the coefficient data.

[0161] Further, the optimization management module comprises a supply-demand matching unit and a power supply adjustment unit; the supply-demand matching unit is configured to correspondingly match the predicted power load demand of each node in the power system at a future time with the power supply amount, and send a pre-warning to the management personnel when the supply-demand difference is too large; and the power supply adjustment unit is configured to combine the historical power supply and power generation conditions to send a prediction warning to the management personnel about the power supply abnormality of a specific node.

[0162] Further, as shown in Figure 4 , the specific working process of the power supply adjustment unit is as follows:

[0163] S31: Extract the power load demand data and power supply amount of each node in the power system from the historical operation data.

[0164] S32: Acquire the power load demand predicted by the analysis and prediction module at a future time and the power supply amount at the corresponding time.

[0165] S33: Take the current time as a time cut point, and establish a prediction monitoring period and a historical monitoring period in the future time and the historical time, respectively.

[0166] S34: For each node in the power system, calculate the power supply abnormality parameter by combining the power load demand data and the power supply amount in the prediction monitoring period and the historical monitoring period:

[0167] ;

[0168] Among them, is the power supply abnormality parameter, represents the power load demand at a certain time point in the prediction monitoring period, represents the power supply amount at a certain time point in the prediction monitoring period, and satisfies , is the time period in which the prediction monitoring period is located; represents the maximum difference between the power load demand and the power supply amount at all time points in the prediction monitoring period; is the total time amount of the supply-demand difference in the historical monitoring period; is the total time amount of the historical monitoring period. is a supply-demand deviation adjustment coefficient, used to adjust the influence of the historical supply-demand imbalance frequency on the power supply abnormality parameter.

[0169] Further, for satisfies:

[0170] ;

[0171] wherein, is the power load demand at time point in the historical monitoring period, is the power supply amount at time point in the historical monitoring period, is a supply-demand difference threshold, used to evaluate the severity of the supply-demand imbalance; is an indicator function, when is true, time point is counted into .

[0172] S35: Set the power supply abnormality threshold, and compare the power supply abnormality parameter of each node with the power supply abnormality threshold, when there is a power supply abnormality parameter greater than the power supply abnormality threshold, it is predicted that the node may have a power supply abnormality, and the power supply abnormality prediction reminder of the node is sent to the management personnel.

[0173] The scheme combines the predicted power load demand and power supply amount to identify the possible supply-demand imbalance condition in the future time, thereby issuing an early warning to ensure the stability of the system; by analyzing the power load demand and power supply amount in the historical data and the predicted data, the potential power supply abnormality is finely evaluated, the accurate identification of the high-risk node is realized, and the reliability and predictability of the system to abnormal conditions are enhanced.

[0174] The above disclosed content is only the preferred feasible embodiment of the present application, and does not limit the protection scope of the present application, so any equivalent technical changes made according to the content of the present application specification and drawings are included in the protection scope of the present application, and in addition, the elements can be updated as the technology develops.

Claims

1. An intelligent optimization and big data analysis system for energy and power systems, characterized in that, The system includes a data acquisition and monitoring module, an analysis and prediction module, an optimization and management module, and a user management module; The data acquisition and monitoring module is used to acquire operational data at each node in the energy and power system; the analysis and prediction module is used to predict and analyze the load demand and supply of each node in the power system based on the acquired operational data; the optimization and management module is used to analyze power supply anomalies at each node in the power system based on the output results of the analysis and prediction module. The user management module is used to facilitate interaction between administrators and coefficient data; The data acquisition and monitoring module includes a data acquisition unit, a data transmission unit, and a storage management unit. The data acquisition unit is used to acquire operational data from various nodes in the energy and power system. The operational data includes power load demand data, power supply data, environmental data, and other data on factors that have a key impact on power demand and supply at each node. The data transmission unit is used to transmit the operational data. The storage management unit is used to receive the operational data transmitted from the data transmission unit and store and manage the operational data according to a timeline. The analysis and prediction module includes a data preprocessing unit, a demand prediction unit, and a power generation prediction unit. The data preprocessing unit is used to preprocess the operational data input from the acquisition and monitoring module, and the preprocessing operations include data cleaning and feature extraction. The demand prediction unit is used to predict the power load demand of the energy and power system in the future. The specific workflow of the demand forecasting unit is as follows: S11: Based on the periodic variation pattern of power load demand, a demand forecasting period is set, which is the forecasting period of future power demand starting from the current time; and the operating data of multiple consecutive demand reference periods corresponding to the demand forecasting period are obtained from historical operating data. S12: Establish an initial load forecasting model, which is used to make a preliminary forecast of the power load demand within the demand forecasting period; the initial load forecasting model is obtained through training on historical data; for each node in the power system, the specific representation of the initial load forecasting model is as follows: ； in, For the initial load forecasting model at the forecast time The predicted time is the time when the electricity load demand is expected to be met. This refers to a specific point in time within the demand forecasting period. The constant term of the initial load forecasting model represents the bias parameters of the model, which are obtained through training with historical data. The number of historical demand reference periods selected for the initial load forecasting model; The autoregressive parameter represents the prediction time. The previous The impact of electricity load demand in a reference period on the current predicted electricity load demand is obtained through training with historical data. For predicting time The previous The corresponding forecast time within the demand reference period Reference weight at time, For predicting time The previous The corresponding forecast time within the demand reference period Power load demand at that time; This is a random error term, obtained through model training and fitting. for satisfy: ； in, This is the initial decay coefficient for demand weights, used to control the initial value variation of demand reference weights, and is set through pre-experimentation; The decay rate of the demand weight index is used to control the decay speed of the demand reference weight, and is set through pre-experimentation. S13: Establish a model of the impact of external data on electricity load demand based on historical external data; the external data includes environmental data and other data that have key influencing factors on electricity demand. S14: Obtain external data information for the demand forecasting period from third-party data sources; S15: Combine external data to complete the final forecast of electricity load demand; The optimization management module includes a supply and demand matching unit and a power supply adjustment unit. The supply and demand matching unit is used to match the predicted power load demand and power supply of each node in the power system in the future time, and send an early warning to the management personnel when the supply and demand difference is too large. The power supply adjustment unit is used to combine historical power supply and generation data to issue a predictive reminder to the management personnel for power supply anomalies of specific nodes. The specific workflow of the power supply adjustment unit is as follows: S31: Extract power load demand data and power supply for each node in the power system from historical operating data; S32: Obtain the power load demand and corresponding power supply predicted by the analysis and forecasting module for future times; S33: Establish predictive monitoring periods and historical monitoring periods for both future and historical times, using the current time as the cutoff point; S34: For each node in the power system, calculate power supply anomaly parameters by combining the power load demand data and power supply data from the predictive monitoring period and the historical monitoring period. ； in, These are parameters indicating abnormal power supply. This indicates a specific point in time within the prediction and monitoring period. Electricity load demand at that time This indicates a specific point in time within the prediction and monitoring period. The power supply at that time meets , To predict the time period of the monitoring cycle; This refers to the total amount of time during which supply and demand differences exist within the historical monitoring period; This represents the total time span of the historical monitoring period; S35: Set the power supply anomaly threshold and compare the power supply anomaly parameter of each node with the power supply anomaly threshold. When there is a power supply anomaly parameter that is greater than the power supply anomaly threshold, it is predicted that there may be a power supply anomaly at that node, and a power supply anomaly prediction reminder for that node is sent to the management personnel.

2. The intelligent optimization and big data analysis system for an energy and power system according to claim 1, characterized in that, The power generation prediction unit is used to predict the power supply of the energy and power system in the future.

3. The intelligent optimization and big data analysis system for an energy and power system according to claim 2, characterized in that, The specific workflow of the power generation prediction unit is as follows: S21: Based on the cyclical variation of power supply, a supply forecast period is set, which is the forecast period of future power supply starting from the current time; and the operating data of multiple consecutive supply reference periods corresponding to the supply forecast period are obtained from historical operating data. S22: Establish an initial supply forecasting model, which is used to make a preliminary forecast of electricity supply within the supply forecasting period; the initial supply forecasting model is obtained through training on historical data; for each node in the power system, the specific representation of the initial supply forecasting model is as follows: ； in, For the initial supply forecasting model at the forecasting time The predicted power supply at that time This refers to a specific point in time within the supply forecasting cycle. The constant term of the initial supply forecasting model represents the bias parameters of the model, which are obtained through training with historical data. The number of historical supply reference periods selected for the initial supply forecasting model; The autoregressive parameter represents the prediction time. The previous The impact of electricity supply in a given supply reference period on the current forecast electricity supply is obtained through training with historical data; For predicting time The previous The corresponding forecast time within each supply reference cycle Reference weight at time, For predicting time The previous The corresponding forecast time within each supply reference cycle Electricity supply at that time; This is a random error term, obtained through model training and fitting. S23: Establish a model of the impact of external data on power load supply based on historical external data; the external data includes environmental data and other data that have key influencing factors on power supply. S24: Obtain external data information for the supply forecasting cycle from third-party data sources; S25: Combine external data to complete the final forecast of power load supply.

Images