Comprehensive energy supervision method and supervision system thereof

Abstract

The invention discloses a comprehensive energy supervision method and system, and relates to the technical field of comprehensive energy supervision, and the method comprises the steps: collecting and integrating distributed energy data, and achieving the data cleaning and processing based on a multi-modal data processing platform; constructing a fusion model based on historical and real-time data, and generating reliable data input through time alignment, space matching and characteristic extraction; based on historical and real-time environment monitoring data, equipment operation parameters are adjusted in real time; historical and real-time data are combined, future energy consumption requirements are predicted, a dynamic scheduling plan is generated according to the future energy consumption requirements, and equipment operation is optimized; based on a block chain technology; a complete energy management closed-loop system is formed through data acquisition, optimization, regulation and control, prediction and storage, the problems of incomplete acquisition, fixed regulation and control, inaccurate prediction and the like in the prior art are solved, and the operation efficiency, the reliability and the intelligent level of the system are greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of integrated energy supervision, in particular to an integrated energy supervision method and its supervision system. Background Art

[0002] With the continuous growth of global energy demand and the improvement of energy utilization efficiency, integrated energy supervision technology has gradually become one of the core research directions in the energy field. In this context, the application of distributed sensors and smart meters provides basic support for the real-time collection and monitoring of energy data. At the same time, the development of blockchain technology provides innovative means for data storage, verification, and traceability, providing strong guarantees for the transparency and data security of the energy management system. Moreover, multi-modal data fusion and artificial intelligence algorithms are gradually applied to fields such as energy demand forecasting and equipment scheduling optimization, significantly improving the operation efficiency and prediction accuracy of the system; However, the existing integrated energy supervision systems still face certain technical problems in data collection, fusion, and decision support, and urgently need to be broken through; Currently, the existing energy supervision technologies on the market mainly have the following deficiencies: First of all, the distribution of multi-modal data in space and time is uneven, resulting in insufficient data collection coverage and making it difficult to comprehensively reflect the dynamic changes in energy use and environmental monitoring; Secondly, due to the complex data sources and diverse frequencies, the algorithm limitations in the data cleaning and processing links of the existing systems make the anomaly detection and denoising performance poor, thus affecting the accuracy of the prediction and optimization models; In addition, the traditional data fusion method lacks the ability of in-depth mining in the time and space dimensions and cannot effectively capture the complex correlations between historical and real-time data; Finally, the dynamic adjustment of equipment operation parameters still relies on fixed strategies, lacking flexibility and making it difficult to adapt to the changing energy demands and external environments; Therefore, the existing technology urgently needs an integrated energy supervision method and its supervision system that can achieve comprehensive supervision and efficient management of energy data by combining multi-modal data processing, trend prediction, dynamic scheduling, and blockchain technology. Summary of the Invention

[0003] In view of the above existing problems, the present invention is proposed.

[0004] Therefore, the present invention provides a comprehensive energy supervision method and its supervision system, which solve the following problems: 1. Traditional data fusion methods lack the ability to deeply mine in the time and space dimensions and cannot effectively capture the complex correlation between historical and real-time data; 2. The algorithm limitations in the data cleaning and processing links of existing systems result in poor anomaly detection and denoising performance; 3. The dynamic adjustment of equipment operation parameters still relies on fixed strategies, lacking flexibility and being difficult to adapt to the changing energy demands and external environments.

[0005] To solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a comprehensive energy supervision method, which includes: Collect and integrate distributed energy data, and implement data cleaning and processing based on a multi-modal data processing platform; Construct a fusion model based on historical and real-time data, and generate reliable data inputs through time alignment, space matching, and feature extraction; Based on historical and real-time environmental monitoring data, adjust the equipment operation parameters in real time; Combine historical and real-time data to predict future energy consumption demands, and generate a dynamic scheduling plan accordingly to optimize equipment operation; Based on blockchain technology, combine the prediction and scheduling results, continuously optimize the control model, and achieve closed-loop control and dynamic adjustment.

[0006] As a preferred solution of the comprehensive energy supervision method of the present invention, the collection and integration of distributed energy data and the implementation of data cleaning and processing based on a multi-modal data processing platform include: Based on the sensors and smart meters installed on distributed energy nodes, by optimizing the spatial distribution and sampling coverage of the sensors, as shown in the following formula: ; Wherein, is the sampling coverage efficiency, which is used to represent the coverage ability of the sensors for the target area in a given spatial area; is the target coverage area of the sensor, which is used to define the integration range; is the position of the th sensor, which represents the coordinates of the sensor in the target area; is the center point of the target area, which defines the reference center of the coverage; is the spatial coordinate, which is used to calculate the coordinate distribution within the integration area; By setting edge network nodes, establish a preliminary data processing model to optimize its load, as shown in the following formula: ; Wherein, is the edge network node Load; Is the acquisition time period; Is the number of sensors connected to the gateway node ; Is the location of the gateway node; Is the th sensor at time Data volume; Using a low-power communication network, based on the optimization objectives of data transmission distance and delay, as follows: ; Among them, Is the communication network performance; Is the data transmission delay; Is the acquisition time period; Is the node And Location; Based on the characteristics of sensor historical data, construct an acquisition frequency dynamic adjustment model and related parameter expressions: ; Among them, Is the th type of sensor at time Acquisition frequency; Is the th type of sensor at time Historical data characteristic value, used to evaluate the dynamic impact of historical data on the acquisition frequency; ; Among them, Is the dynamic acquisition performance index of the system within the time; Is the time interval of the acquisition period; Is the total number of sensor categories; Is the th type of sensor acquisition frequency, indicating the real-time acquisition times of the sensor within the time ; Is the th type of sensor historical data value, used to assist the reinforcement learning algorithm to judge the dynamic adjustment of the acquisition frequency; Is the time decay function, used to weaken the lag effect of historical data; Indicates the time change rate of the acquisition frequency of the i-th type of sensor; Is the spatial coverage range of the i-th type of sensor, is the spatial sampling intensity, and describes the spatial optimization degree within the sensor coverage range in combination with the cosine distribution; Is the total number of noise sources; Is the th type of noise at time The dynamic value; Is a function for non - linear smoothing of noise, used to weaken the negative impact of noise on acquisition performance; Is the dynamic characteristic of noise changing over time; Perform outlier detection and removal on the multi - modal data collected by the sensor, and achieve data cleaning by combining time - series analysis. Its detection rule is as follows: ; Where, Is the outlier detection result; Is the Actual data of the th type of sensor; Expected data of the th type of sensor; Is the outlier detection threshold; ; Where, Is the fused data; Is the Data value of the th data source at time Is the Outlier detection value of the th data source;

[0007] As a preferred solution of the integrated energy supervision method described in the present invention, wherein, the construction of the fusion model based on historical and real - time data, and the generation of reliable data input through time alignment, space matching, and feature extraction include: Extract trend features from the fused data to capture the dynamics of data changing over time. The formula is as follows: ; Where, Is the trend feature, used to capture the change trend of data over time; Is the fused data value, representing the data fusion result at time ; Is the time variable; Based on the extracted trend information, analyze the dynamic changes of energy forms, and combine with spatial distribution characteristics to construct an optimization model: ; Where, Is the energy form conversion efficiency; Is the dynamic characteristic of energy form conversion, used to describe the changes during the conversion process; Is the energy supply characteristic; Used to describe the non-linear spatial characteristics of energy conversion; Represents the time dynamic change; The value range is ; The value range of depends on the dynamic characteristics of and the spatial periodicity of; After the trend and conversion optimization are completed, the collaborative fusion of data is achieved through the following formula. Combining the time and space distribution characteristics, the fusion effect is further optimized: ; Among them, is the efficiency of the collaborative fusion of multi-modal data combining time and space characteristics; is the dynamic data intensity of the th is the dynamic anomaly characteristic of the th is the feature distribution in the two-dimensional space; is the complexity description of the spatial characteristic distribution; is the abnormal data weight adjustment function, used to suppress the influence of abnormal data on the fusion performance; is the abnormal data weight adjustment function, used to suppress the influence of abnormal data on the fusion performance; By fusing data and combining trend, periodic and random characteristics, the future energy consumption demand is predicted. The formula is as follows: ; Among them, is the predicted value of future energy consumption at time ,; is the trend characteristic term; is the periodic characteristic term, , where represents the periodic frequency; is the random characteristic term, representing the influence of random fluctuations. The specific formula is ; is the influence weight of external environmental factors, representing the interference intensity of the external environment on the energy consumption demand. This term is used to reduce the influence of environmental changes on the prediction result; To improve data consistency, the historical fusion data and real-time data are standardized. The formula is as follows: ; Among them, is the normalized data, used to label data with different dimensions; is the original data; is the maximum and minimum values of the data; And further extract the trend characteristics. The formula is: ; Among them, is the trend characteristic, which is used to extract the dynamic change trend of data; is the normalized data; is the natural logarithm function, which is used to smooth the data and reduce the non-linear influence; is the second-order time derivative, which is used to analyze the second-order dynamic characteristics of data changes; Based on the fused data, combined with time series analysis, determine the trend characteristics, periodicity and stochastic characteristics of future demand, and finally generate a prediction formula: ; Among them, is the predicted value of future demand; is the trend characteristic value; is the reciprocal weighted value of the fused data; is the fused data; is the environmental parameter, which is collected by sensors.

[0008] As a preferred solution of the integrated energy supervision method described in the present invention, among them, the real-time adjustment of device operation parameters based on historical and real-time environmental monitoring data includes: On the basis of data dynamic fusion, further optimize the path transmission efficiency, as the following formula: ; Among them, is the path planning performance; is the path delay characteristic; is the node position; is the gateway node position; After optimizing the path planning, further evaluate the rationality of the data through a verification model, and combine the dynamic correlation to evaluate the logical relationship between historical and real-time data: ; Among them, is the verification efficiency, which is used to evaluate the performance verification effect of the system or model within time ; is the th type of data channel noise characteristic, indicating the noise influence of different channels at time ; Further combine the dynamic correlation model to evaluate the relationship between historical and real-time data: ; Among them, It is a dynamic correlation index used to measure the dynamic relationship between the historical parameters, real-time parameters of the device and environmental factors; is the historical parameter of the th device, representing the historical record characteristics of the device operation, such as the historical change data of the energy consumption pattern; is the real-time parameter of the jth device, representing the real-time operation state of the current device, such as power consumption, power demand, etc.; is the environmental parameter of the jth device, representing the influence of the current environment on the device operation, such as temperature, humidity, etc.; Detect and eliminate parameter abnormal points; After completing the correlation analysis, optimize the data storage with the help of a hierarchical blockchain architecture: ; Among them, is the storage efficiency of the hierarchical blockchain; is the high-frequency data feature; is the high-frequency data transmission delay.

[0009] As a preferred solution of the comprehensive energy supervision method described in the present invention, wherein, combining historical and real-time data to predict future energy consumption demand and generating a dynamic scheduling plan accordingly to optimize device operation, including: Generate a scheduling plan according to the predicted energy consumption data and device operation parameters, which is specifically implemented through the following formula: ; Among them, is the calculation result of the scheduling plan; is the time device power; is the operation weight of the device; The formula for predicting future demand is: ; Among them, is the predicted value of future energy consumption demand; is the trend characteristic; is the representation of the periodic characteristic; is the result of the fusion of historical and real-time data; Predict future energy consumption demand, and its prediction formula is: Correct the difference between the predicted and actual energy consumption data, which is expressed by the following formula: ; Among them, is the error correction value, representing the deviation size between the predicted data and the actual data within the time ; is the time The actual data value at a moment, obtained from the actually collected data; is the time The predicted data value at the moment; is the environmental parameter weight, obtained by sensor collection; Normalize historical data and real-time data to eliminate the dimension difference. The normalization formula is: ; where, is the normalized data, used to standardize data with different dimensions; is the time of the original data; are the maximum and minimum values of the data.

[0010] As a preferred solution of the integrated energy supervision method described in the present invention, wherein, based on the blockchain technology, combining the prediction and scheduling results, continuously optimizing the control model, and realizing closed-loop control and dynamic adjustment includes: Encrypt and store the prediction and scheduling results based on the blockchain technology, as shown in the following formula: ; where, is the storage efficiency, used to measure the load balance of storage nodes; is the time range of storage optimization; is the high-frequency data characteristic, used to describe the storage efficiency of high-frequency data; is the low-frequency data characteristic, used to describe the storage performance of low-frequency data; is the low-frequency data trade-off function, used to balance the storage requirements of high- and low-frequency data; Realize the dynamic calibration of the control model through the blockchain feedback mechanism, and achieve it through the following formula: ; where, is the optimized control ability, used to describe the dynamic optimization achieved by the system by adjusting the scheduling parameters; is the time period, representing the total duration of the optimization process; is the change in the power time of the th device, used to measure the load capacity of the device; is the operation weight of the th device, used to describe the importance of the device, and is calculated from the real-time data collected by the sensor; ; where, For anomaly detection performance, which is used to evaluate the effectiveness and accuracy of anomaly detection in device operating states or environmental parameters; For abnormal fluctuations in device operating parameters, which is used to quantify the degree to which device parameters deviate from the normal state; For abnormal changes in environmental parameters, which is used to describe abnormal fluctuations in the impact of the environment on device operating parameters.

[0011] In a second aspect, the present invention provides an integrated energy supervision system, including, A data acquisition module, which is used to collect real-time status data of distributed energy through distributed sensors and smart meters; A data cleaning and processing module, which is used to clean the collected multimodal data and eliminate abnormal points; A data fusion module, which is used to fuse the spatio-temporal characteristics of historical data and real-time data; An energy consumption characteristic extraction module, which is used to extract trend characteristics, periodic characteristics, and random characteristics from the fused data; An energy consumption prediction module, which is used to predict future energy consumption requirements based on historical and real-time data; A dynamic scheduling module, which is used to generate a dynamic scheduling plan and optimize device operation according to the energy consumption prediction results; A blockchain storage module, which is used to encrypt and store prediction and scheduling data, and optimize based on high-low frequency hierarchical storage; An anomaly detection and feedback module, which is used to evaluate abnormal fluctuations in device operating states and environmental parameters, and dynamically calibrate the prediction model and scheduling model; A visualization and user interaction module, which is used to display the system operating state, prediction results, and scheduling plan in real time, and at the same time support users to adjust parameter inputs and policy modifications; Among them, the data cleaning and processing module includes: A noise filtering unit, which is used to eliminate data points that do not conform to the rules using an anomaly detection model; A normalization processing unit, which is used to standardize data from different sources and eliminate dimension differences; A data completion unit, which is used to complete missing data points through interpolation and historical pattern matching; A data classification unit, which is used to classify data into critical data and auxiliary data for processing; Among them, the data fusion module includes: A spatio-temporal alignment unit, which is used to align historical and real-time data and match spatial characteristics; A multimodal data fusion unit, which is used to merge data from different sources using a mathematical model to generate comprehensive characteristics; A characteristic weight allocation unit, which is used to allocate weights according to data characteristics; Among them, the blockchain storage module includes: A data encryption unit for encrypting stored data; A hierarchical storage unit for storing data in layers according to high-frequency and low-frequency characteristics; A blockchain network unit for implementing data storage and management based on blockchain technology, including data traceability and storage verification; A storage optimization unit for dynamically adjusting the storage strategy by combining prediction and scheduling data priorities.

[0012] In a third aspect, the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, any step of the comprehensive energy supervision method described in the first aspect of the present invention is implemented.

[0013] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, any step of the comprehensive energy supervision method described in the first aspect of the present invention is implemented.

[0014] The beneficial effects of the present invention are as follows: The energy data is collected through a distributed sensor network and smart meters, and the spatial coverage accuracy of data collection is improved by using the coverage range optimization formula, realizing the comprehensive and accurate capture of dynamic information such as energy consumption, heating efficiency, and environmental data, and providing high-quality raw data support for subsequent data fusion and prediction, avoiding the monitoring distortion problem caused by insufficient data collection coverage in the prior art, thereby significantly improving the reliability and accuracy of data collection; The sensor data transmission path is optimized by the path planning formula, effectively reducing the transmission delay and energy consumption, realizing the high efficiency of the communication network, and reducing the possible data loss rate and transmission energy consumption during the transmission process, providing timely data support for dynamic control devices, and ultimately improving the system operation efficiency and extending the service life of sensor nodes; Combined with the verification efficiency calculation model and dynamic correlation indicators, the device operation parameters are adjusted in real time. The dynamic optimization of the operation state of energy devices is realized, the energy utilization rate is improved, and by optimizing the energy supply-demand matching, resource waste or insufficient supply is avoided, and finally the energy allocation is made more accurate and efficient, improving the overall efficiency of the energy management system; Based on historical and real-time data, the future energy demand is analyzed and a scheduling plan is formulated by using a fusion algorithm and a trend prediction model; The advance estimation and optimized scheduling of energy demand are realized, which is used to reduce the volatility and uncertainty of system operation, ensure the stability and reliability of energy supply, and thus greatly improve the prediction ability and emergency response level of the energy management system; Combined with blockchain technology, the adaptive adjustment of the control model is realized through dynamic feedback to ensure the security and transparency of data storage. At the same time, real-time feedback regulation is achieved, the adaptability of the system to environmental parameter changes is enhanced, and the impact of abnormal states on energy regulation is reduced. In this way, a complete energy management closed-loop system is formed through data collection, optimization, regulation, prediction, and storage, which not only overcomes the problems of incomplete collection, fixed regulation, and inaccurate prediction in the prior art, but also greatly improves the operating efficiency, reliability, and intelligence level of the system. Brief Description of the Drawings

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0016] Figure 1 It is a flowchart of the comprehensive energy supervision method in Embodiment 1.

[0017] Figure 2 It is a schematic diagram of the comprehensive energy supervision system in Embodiment 1. Detailed Embodiments

[0018] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the specific embodiments of the present invention will be described in detail below with reference to the drawings in the specification.

[0019] Many specific details are set forth in the following description in order to provide a thorough understanding of the present invention. However, the present invention may be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0020] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation manner of the present invention. The appearances of "in one embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that exclude each other with other embodiments.

[0021] Embodiment 1, referring to Figure 1 and Figure 2 , is the first embodiment of the present invention. This embodiment provides a comprehensive energy supervision method and its supervision system, including the following steps: Collect and integrate distributed energy data, and implement data cleaning and processing based on a multimodal data processing platform; this step realizes real-time collection of data through a variety of distributed sensor devices to ensure the comprehensiveness of data sources. The integration of data can reduce the information fragmentation problem caused by data silos and provide a unified data basis for subsequent analysis; Build a fusion model based on historical and real-time data, and generate reliable data inputs through time alignment, spatial matching, and feature extraction; the fusion model effectively combines the trend information of historical data with the dynamic changes of real-time data, improving the depth and accuracy of data analysis. This model can capture the complex correlations of spatio-temporal data and optimize energy scheduling decisions; Based on historical and real-time environmental monitoring data, adjust the device operation parameters in real time; real-time adjustment of device operation parameters can dynamically optimize the operation state of the device, reduce energy waste and extend the device life. By monitoring and responding to environmental factors, the system's ability to adapt to complex environmental changes is improved; Combine historical and real-time data to predict future energy consumption demands, and generate a dynamic scheduling plan accordingly to optimize device operation; the dynamic scheduling plan can reduce energy waste caused by scheduling lags and achieve reasonable allocation of resources by accurately predicting future demands; Based on blockchain technology, combine the prediction and scheduling results, continuously optimize the control model, and achieve closed-loop control and dynamic adjustment; the introduction of blockchain technology ensures the immutability of data and the transparency of sharing. The optimized control model synchronously optimizes the energy flow and information flow by coordinating energy data and scheduling schemes.

[0022] Step 1: Collect and integrate distributed energy data, and the data cleaning and processing based on a multimodal data processing platform include: Based on the sensors and smart meters installed at distributed energy nodes, by optimizing the spatial distribution and sampling coverage of sensors, as shown in the following formula: ; where, is the sampling coverage efficiency, which is used to represent the coverage ability of sensors for the target area within a given spatial area; is the target coverage area of the sensor, which is used to define the integration range; is the th sensor position, representing the coordinates of the sensor in the target area; is the center point of the target area, defining the reference center of coverage; is the spatial coordinate, which is used to calculate the coordinate distribution within the integration area; By setting edge network nodes, establish a preliminary data processing model to optimize its load, as shown in the following formula: ; Among them, is the load of the edge gateway node ; is the acquisition time period; is the number of sensors connected to the gateway node ; is the location of the gateway node; is the th sensor's data volume at time ; Using a low-power communication network, based on the optimization objectives of data transmission distance and delay, as shown in the following formula: ; Among them, is the communication network performance; is the data transmission delay; is the acquisition time period; is the node and 's location; Based on the characteristics of sensor historical data, construct an acquisition frequency dynamic adjustment model and related parameter expressions: ; Among them, is the acquisition frequency of the th type of sensor at time ; is the historical data characteristic value of the th type of sensor at time , used to evaluate the dynamic impact of historical data on the acquisition frequency; ; Among them, is the dynamic acquisition performance index of the system within a certain time; is the time interval of the acquisition period; is the total number of sensor categories; is the th type of sensor's acquisition frequency, indicating the real-time acquisition times of the sensor within time ; is the historical data value of the th type of sensor, used to assist the reinforcement learning algorithm to judge the dynamic adjustment of the acquisition frequency; is the time decay function, used to weaken the lag effect of historical data; represents the time change rate of the acquisition frequency of the i-th type of sensor; is the spatial coverage range of the i-th type of sensor, is the spatial sampling intensity, and describes the spatial optimization degree within the sensor coverage range in combination with the cosine distribution; is the total number of noise sources; is the dynamic value of the type of noise at time ; is a function for non - linear smoothing of noise, used to weaken the negative impact of noise on acquisition performance; is the dynamic characteristic of noise changing with time; Perform outlier detection and elimination on the multi - modal data collected by sensors, and achieve data cleaning by combining time - series analysis. Its detection rule is as follows: ; where is the outlier detection result; is the actual data of the type of sensor; is the expected data of the type of sensor; is the outlier detection threshold; By fusing multi - modal data and optimizing spatial distribution, eliminate outlier data points. Its model is as follows: ; where is the fused data; is the data value of the data source at time ; is the outlier detection value of the data source; is the target space range; In this embodiment, construct an experimental scenario to simulate an energy distributed management scenario, including multiple sensor nodes and smart meters, covering different regional ranges; This experiment is divided into two stages: The existing technology method: Scheme A, and the method of the present invention: Scheme B, are compared. The experimental scenario includes different energy nodes, each node is equipped with multiple types of sensors, as well as data storage and transmission functions. The experimental scenario also includes different historical data input conditions and real - time monitoring requirements, involving dynamic data scheduling, fusion, prediction, and optimization; The experimental parameters include: Data acquisition range coverage rate, recording the coverage ability of sensors for the target area; Data delay time, counting the total delay of data transmission and processing; Data accuracy rate, calculating the deviation between the monitored data and the true value; Energy prediction error, counting the average error between the predicted value and the true value; Transmission energy consumption, calculating the energy consumption of sensor nodes; The detailed experimental process: Data collection and coverage optimization, optimizing the sensing coverage through the sensor layout algorithm. In Scheme A, the traditional fixed node distribution is adopted, while in Scheme B, the dynamic coverage model of the present invention is used for optimization; Data transmission and fusion. In Scheme A, simple weighted fusion is used, while in Scheme B, the fusion model constructed by the present invention is utilized to dynamically match through multi-dimensional time series and spatial characteristics; Prediction and scheduling. In Scheme B, the future energy demand is calculated based on the formula and a dynamic scheduling plan is generated, while in Scheme A, a fixed prediction mode is adopted; Collection of evaluation indicators, recording key indicators such as energy consumption, prediction error, and data accuracy during each stage of the experiment; Experimental data record form for simulating the energy distribution management scenario (#Table 1) ; It can be seen from the experimental data that the comprehensive energy supervision method of the present invention is significantly superior to the traditional scheme in terms of coverage rate, latency, accuracy, prediction error, and energy consumption, especially in dynamic and complex scenarios.

[0023] Step 2: Construct a fusion model based on historical and real-time data, and generate reliable data inputs through time alignment, spatial matching, and feature extraction, including: Extract the trend features from the fusion data to capture the dynamics of the data changing over time. The formula is as follows: ; Wherein, is the trend feature, used to capture the changing trend of the data over time; is the fusion data value, representing the data fusion result at time ; is the time variable; Based on the extracted trend information, analyze the dynamic changes of the energy form, and construct an optimization model in combination with the spatial distribution characteristics: ; Wherein, is the energy form conversion efficiency; is the dynamic characteristic of the energy form conversion, used to describe the changes during the conversion process; is the energy supply characteristic; is used to describe the non-linear spatial characteristics of the energy conversion; is the representation of the dynamic change of time; The value range of ; The value range of depends on the dynamic characteristic of and the spatial periodicity of After the trend and transformation optimization are completed, the collaborative fusion of data is achieved through the following formula. Combining the temporal and spatial distribution characteristics, the fusion effect is further optimized: ; where is the efficiency of the multi-modal data collaborative fusion that combines temporal and spatial characteristics; is the dynamic data intensity of the th type of data source; is the dynamic anomaly characteristic of the th type of data source; is the feature distribution in two-dimensional space; is the complexity description of the spatial characteristic distribution; is the abnormal data weight adjustment function, which is used to suppress the influence of abnormal data on the fusion performance; is the abnormal data weight adjustment function, which is used to suppress the influence of abnormal data on the fusion performance; By fusing data and combining trend, periodic, and random characteristics, the future energy consumption demand is predicted. The formula is as follows: ; where, is the predicted value of future energy consumption at time ,; is the trend characteristic term; is the periodic characteristic term, , where represents the periodic frequency; is the random characteristic term, which represents the influence of random fluctuations. The specific formula is ; is the influence weight of external environmental factors, which represents the interference intensity of the external environment on energy consumption demand. This term is used to reduce the influence of environmental changes on the prediction result; To improve data consistency, the historical fusion data and real-time data are standardized. The formula is as follows: ; where, is the normalized data, which is used to label data with different dimensions; is the original data; is the maximum and minimum values of the data; And the trend characteristic is further extracted. The formula is: ; where, is the trend characteristic, which is used to extract the dynamic change trend of the data; is the normalized data; is the natural logarithm function, which is used to smooth the data and reduce the non-linear influence; is the second-order time derivative, which is used to analyze the second-order dynamic characteristic of the data change; Based on the fused data, combined with time series analysis, determine the trend characteristics, periodicity, and stochastic characteristics of future demand, and finally generate a prediction formula: ; Wherein, is the predicted value of future demand; is the trend characteristic value; is the reciprocal weighted value of the fused data; is the fused data; is the environmental parameter, which is collected by sensors; In this embodiment, in order to verify the effect of this method, an intelligent building energy management system is selected as the application scenario in this experiment, aiming to analyze and predict the energy consumption data through the comprehensive energy supervision method proposed in the invention content, and optimize the energy management strategy; The experiment is divided into three stages: data collection and preprocessing, model analysis and prediction, and implementation of the optimization plan; First, key energy-consuming devices in the building, such as air conditioners, lighting systems, elevators, etc., are selected as monitoring objects, and high-precision sensors are installed for data collection. The data collection module determines the spatial coverage range of the sensors according to formula (#1) in the invention content to ensure the spatio-temporal continuity of the collected data; Subsequently, the collected data is preprocessed using a multi-modal data processing platform, including data standardization, noise filtering, etc. According to formula (#3), the time-delay characteristics of sensor data transmission are optimized; In the model analysis stage, the future energy demand is predicted through formula (#14), and the operating parameters of the equipment are dynamically adjusted in combination with the prediction results; Finally, in the implementation stage of the optimization plan, the fusion optimization of multi-modal data is realized through formula (#10), and an energy scheduling plan is formulated; Intelligent Building Energy Optimization and Comparative Analysis Data Table (#Table 2) ; First, based on the prediction model of formula (#14), in the prediction of future energy demand, the average error is only 4%, which is significantly lower than the average error of more than 10% in the existing technology; Secondly, by optimizing the data transmission time delay through formula (#3), the overall time delay is reduced by an average of 15 ms, significantly improving the real-time performance and stability of data transmission; In addition, after the optimization of multi-modal data fusion is achieved through formula (#10), the energy consumption is significantly reduced. The tabular data shows that the energy consumption of the air conditioner, lighting system, and elevator is reduced by 10%, 10%, and 10% respectively, and the total energy consumption reduction reaches 10%. These data indicate that the method of the present invention has significant advantages in reducing energy consumption and improving energy utilization efficiency.

[0024] Step 3: Based on historical and real-time environmental monitoring data, adjust the device operation parameters in real time, including: On the basis of dynamic data fusion, further optimize the path transmission efficiency, as shown in the following formula: ; Where, is the path planning performance; is the path delay characteristic; is the position of node ; is the position of the gateway node ; After optimizing the path planning, further evaluate the rationality of the data through a verification model, and combine the dynamic association to evaluate the logical relationship between historical and real-time data: ; Where, is the verification efficiency, used to evaluate the performance verification effect of the system or model within time ; is the noise characteristic of the th type of data channel, indicating the noise impact of different channels at time ; Further combine the dynamic association model to evaluate the relationship between historical and real-time data: ; Where, is the dynamic association index, used to measure the dynamic relationship between the historical parameters, real-time parameters, and environmental factors of the device; is the historical parameter of the th device, indicating the historical record characteristics of the device operation, such as the historical change data of the energy consumption pattern; is the real-time parameter of the jth device, indicating the real-time operation state of the current device, such as the electricity consumption, power demand, etc.; is the environmental parameter of the jth device, indicating the impact of the current environment on the device operation, such as temperature, humidity, etc.; Detect and eliminate parameter abnormal points; After completing the association analysis, optimize the data storage by means of a hierarchical blockchain architecture: ; Among them, is the hierarchical blockchain storage efficiency; is the high-frequency data feature; is the high-frequency data transmission delay; In this embodiment, to verify the actual effect of this method, a certain intelligent building is selected as the research object. This building uses common electric energy, heat energy and cooling systems to provide the daily operation energy demand. The test objective is to optimize the energy consumption path and verify the effectiveness of the path optimization model and the dynamic regulation algorithm. The traditional method conducts energy distribution through fixed path planning and static energy consumption analysis, while this test introduces historical data analysis and dynamic adjustment mechanism, so as to solve the problems of low energy consumption distribution accuracy and slow response speed of the existing methods; The test includes the following steps: Use the sensor network installed in the building to collect data such as power consumption, heat supply and environmental temperature, and calculate the sensor coverage range using formula (#1), and optimize the sensor layout to ensure the comprehensiveness and accuracy of data collection; Combine the sensor historical data through formula (#15) to optimize the energy consumption path to reduce transmission delay and energy consumption; Adopt a dynamic verification model to calculate the verification efficiency, formula (#16), and combine it with formula (#17) to adjust the device operation parameters in real time to improve the energy utilization rate; Analyze the historical data and environmental data through formula (#14) and predict the future energy demand to form a dynamic regulation plan; Comprehensive energy supervision system optimization effect data table (#Table 3) ; Through the comparison of the above table data, it can be known that: The power consumption rate has decreased by 22.15%, indicating that the optimized path effectively reduces unnecessary energy consumption; The heat energy utilization rate has increased by 18.26%, reflecting the significant improvement of historical data and dynamic regulation on the energy distribution efficiency; The cooling system efficiency has increased by 22.87%, indicating that real-time data analysis and prediction play an important role in optimizing the device performance; The response time has decreased by 35.20%, indicating that the dynamic regulation ability of the system has been significantly improved, and it responds more quickly to the real-time energy demand of the building; The data transmission delay has decreased by 43.33%, indicating the remarkable effect of optimizing the sensor layout and improving the communication model; In summary, the present invention shows significant advantages in aspects such as path optimization, dynamic regulation and data prediction, and effectively improves the deficiencies of the existing technologies.

[0025] Step 4: Combine historical and real-time data to predict future energy consumption demand, and generate a dynamic scheduling plan based on this to optimize equipment operation, including: Generate a scheduling plan based on the predicted energy consumption data and equipment operation parameters, which is specifically implemented through the following formula: ; where, is the calculation result of the scheduling plan; is the time of the equipment power; is the operation weight of the equipment; The formula for predicting future demand is: ; where, is the predicted value of future energy consumption demand; is the trend characteristic; is to represent the periodic characteristic; is the result of the fusion of historical and real-time data; Predict future energy consumption demand, and its prediction formula is: Correct the difference between the predicted and actual energy consumption data, which is expressed by the following formula: ; where, is the error correction value, indicating the deviation magnitude between the predicted data and the actual data within the time ; is the time of the actual data value, obtained from the actually collected data; is the predicted data value at the time ; is the environmental parameter weight, obtained by sensor collection; Standardize the historical data and real-time data to eliminate the dimension difference, and the standardization processing formula is: ; where, is the normalized data, used to standardize data with different dimensions; is the time of the original data; is the maximum and minimum values of the data; In this embodiment, a data collection and processing experiment based on the actual building environment is designed to verify the effectiveness of the integrated energy supervision method of the present invention in optimizing energy scheduling and improving energy utilization efficiency; An energy management system of a large building was selected for the experiment. The test objects included key parameters such as power consumption, heat energy utilization rate, cooling system efficiency, average response time, and data transmission delay. The performance of the traditional energy supervision method and the optimized method of the present invention was compared respectively; In the experiment, an advanced multi-mode sensor network was deployed in the data acquisition module. The sensor layout scheme calculated the optimal transmission path according to the path optimization formula of formula (#19), and predicted the future energy demand through the fusion model of historical and real-time data, formula (#20). Then, the dynamic regulation module corrected the deviation between the actual and predicted demands according to formula (#21) to ensure a high degree of matching between energy supply and demand. At the same time, to verify the energy utilization efficiency, all data was standardized using the normalization formula (#22) to eliminate the dimension difference and ensure the comparability of data between different time periods and indicators; The test data recording period was 5 consecutive days, and the acquisition frequency was once every 10 minutes. The data was uploaded to the central control system through a low-power wireless communication network and was processed and optimized in real time and scheduled in combination with the models of formulas (#19) to (#22). The average value of the existing energy management system of the building was used as the control group for the test data of the traditional method; Comprehensive energy supervision system performance comparison table (#Table 4) ; It can be seen from the comparison of the data in the table that: Among them, the power consumption decreased by 21.60%, indicating that the optimized path design and prediction model reduced unnecessary energy waste; The heat energy utilization rate increased by 15.66%, reflecting the accurate energy supply ability of the system under dynamic demand regulation; The cooling system efficiency increased by 21.13%, indicating that the operation of the cooling equipment was effectively improved by optimizing the cooling path and adjusting the system workload; The response time was shortened by 19.44%, showing that its dynamic regulation ability was better than the traditional method and could adapt to the real-time demand changes of the building more quickly; The data transmission delay decreased by 43.48%, reflecting the remarkable achievement of the sensor layout and data optimized transmission model; This proves that the method of the present invention has significant advantages in aspects such as energy path optimization, demand prediction, and data transmission compared with traditional technologies. It not only improves the energy utilization efficiency but also reduces energy waste and operating costs.

[0026] Step 5: Based on blockchain technology, combined with the prediction and scheduling results, continuously optimize the control model to achieve closed-loop control and dynamic adjustment.

[0027] The prediction and scheduling results are encrypted and stored based on blockchain technology, as shown in the following formula: ; in, Storage efficiency is used to measure the load balance of storage nodes; Time range optimized for storage; It is a high-frequency data characteristic and is used to describe the storage efficiency of high-frequency data; It is a low-frequency data characteristic, used to describe the storage performance of low-frequency data; It is a low-frequency data trade-off function, used to balance the storage requirements of high-frequency and low-frequency data; The dynamic calibration of the control model is achieved through the blockchain feedback mechanism, which is implemented through the following formula: ; in, The control capability for optimization is used to describe the dynamic optimization achieved by the system by adjusting the scheduling parameters; is the time period, which indicates the total duration of the optimization process; For the Power time of each device The change in is used to measure the load capacity of the equipment; For the The operation weight of a device is used to describe the importance of the device and is calculated based on the real-time data collected by the sensor; By detecting the difference between the predicted and actual energy consumption and correcting the prediction model, it is achieved through the following formula: ; in, Anomaly detection performance, which is used to evaluate the effectiveness and accuracy of anomaly detection in equipment operating status or environmental parameters; It is the abnormal fluctuation of equipment operating parameters, used to quantify the degree to which equipment parameters deviate from the normal state; To indicate abnormal changes in environmental parameters, it is used to describe the abnormal fluctuations in the impact of the environment on equipment operating parameters; In this embodiment, the blockchain feedback mechanism refers to the data transparency, distributed storage and immutability achieved through blockchain technology, which provides a reliable data storage and real-time feedback path for the energy management system, and solves the data lag and inconsistency problems existing in the traditional centralized energy management model; High-frequency data characteristics refer to data collected by sensors that change rapidly in a short period of time, such as changes in ambient temperature and humidity; low-frequency data characteristics refer to data that change slowly over a longer time scale, such as equipment failure rate, long-term energy consumption trend, etc. By analyzing the characteristics of these two types of data separately, it is helpful to optimize data storage strategies and improve system response speed; Therefore, by constructing a storage architecture based on blockchain technology, hierarchical optimization storage of high-frequency and low-frequency data is realized, and the storage ratio is dynamically adjusted according to the characteristics of data types. This mechanism can reduce data redundancy, improve data transmission efficiency, and avoid the risks of data loss and modification through the distributed characteristics of blockchain.

[0028] This embodiment also provides a comprehensive energy supervision system, including: A data acquisition module, configured to collect real-time status data of distributed energy through distributed sensors and smart meters; A data cleaning and processing module, configured to clean the collected multimodal data and remove outliers; A data fusion module, configured to perform spatio-temporal feature fusion on historical data and real-time data; An energy consumption feature extraction module, configured to extract trend features, periodic features, and random features from the fused data; An energy consumption prediction module, configured to predict future energy consumption requirements based on historical and real-time data; A dynamic scheduling module, configured to generate a dynamic scheduling plan and optimize device operation according to the energy consumption prediction results; A blockchain storage module, configured to encrypt and store prediction and scheduling data, and optimize based on high-low frequency hierarchical storage; An anomaly detection and feedback module, configured to evaluate abnormal fluctuations in device operation status and environmental parameters, and dynamically calibrate the prediction model and scheduling model; A visualization and user interaction module, configured to display the system operation status, prediction results, and scheduling plan in real time, and at the same time support users to adjust parameter inputs and policy modifications; Among them, the data cleaning and processing module includes: A noise filtering unit, configured to use an anomaly detection model to remove data points that do not conform to the rules; A normalization processing unit, configured to standardize data from different sources and eliminate the dimension difference; A data completion unit, configured to complete missing data points through interpolation and historical pattern matching; A data classification unit, configured to classify data into critical data and auxiliary data for processing; Among them, the data fusion module includes: A spatio-temporal alignment unit, configured to align historical and real-time data and match spatial features; A multimodal data fusion unit, configured to merge data from different sources using a mathematical model to generate comprehensive features; A feature weight assignment unit, configured to assign weights according to data features; Among them, the blockchain storage module includes: A data encryption unit for encrypting stored data; A hierarchical storage unit for storing data in layers according to high-frequency and low-frequency characteristics; A blockchain network unit for implementing data storage and management based on blockchain technology, including data traceability and storage verification; A storage optimization unit for dynamically adjusting the storage strategy by combining prediction and scheduling data priorities.

[0029] This embodiment also provides a computer device applicable to a comprehensive energy supervision method, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement a comprehensive energy supervision method as proposed in the above embodiment.

[0030] The computer device can be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, 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 achieved through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, a trackball, or a touchpad set on the computer device housing, or an external keyboard, a touchpad, or a mouse, etc.

[0031] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the comprehensive energy supervision method proposed in the above embodiment; 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, abbreviated as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, abbreviated as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, abbreviated as EPROM), programmable read-only memory (Programmable Red-Only Memory, abbreviated as PROM), read-only memory (Read-Only Memory, abbreviated as ROM), magnetic memory, flash memory, magnetic disk or optical disc.

[0032] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A comprehensive energy supervision method, characterized in that: include, Collect and integrate distributed energy data, and implement data cleaning and processing based on a multimodal data processing platform; Build a fusion model based on historical and real-time data to generate reliable data input through time alignment, spatial matching and feature extraction; Adjust equipment operating parameters in real time based on historical and real-time environmental monitoring data; Combine historical and real-time data to predict future energy demand and generate dynamic scheduling plans to optimize equipment operation; Based on blockchain technology, combined with prediction and scheduling results, the control model is continuously optimized to achieve closed-loop control and dynamic adjustment.

2. The integrated energy supervision method according to claim 1, characterized in that: The collection and integration of distributed energy data and the data cleaning and processing based on the multimodal data processing platform include: Based on the sensors and smart meters installed at distributed energy nodes, the spatial distribution of sensors and sampling coverage are optimized as follows: ； in, is the sampling coverage efficiency; is the target coverage area of ​​the sensor; For the The location of the sensors; is the center point of the target area, defining the reference center of coverage; is the spatial coordinate; By setting up edge gateway nodes, a preliminary data processing model is established to optimize its load, as shown below: ； in, For edge gateway nodes Load; is the collection time period; For gateway nodes The number of connected sensors; is the location of the gateway node; For the Sensors at time The amount of data; Using low-power communication networks, the optimization goals of data transmission distance and latency are as follows: ； in, For communication network performance; Delay in data transmission; is the collection time period; For Node and location; Based on the historical data characteristics of the sensor, a dynamic adjustment model for the acquisition frequency and related parameter expressions are constructed: ； in, For the Class sensor in time The collection frequency; For the Class sensor in time The historical data characteristic values ​​are used to evaluate the dynamic impact of historical data on the collection frequency; ； in, Dynamically collect performance indicators for the system over time; is the time interval of the collection cycle; is the total number of sensor categories; For the The acquisition frequency of the sensor, indicating the sensor Real-time collection times within For the Historical data values ​​of class sensors; is the spatial coverage of the i-th sensor; is the total number of noise sources; For the Noise-like Dynamic value of The multimodal data collected by the sensor is detected and eliminated for outliers, and data cleaning is achieved by combining time series analysis. The detection rule is as follows: ； in, is the abnormal detection result; For the Actual data of class sensors; For the Expected data from the sensor class; is the anomaly detection threshold; By fusing multimodal data and optimizing spatial distribution, abnormal data points are removed. The model is as follows: ； in, is the fused data; For the Data source at time The data value of For the Anomaly detection value of the data source; is the target space range.

3. The integrated energy supervision method according to claim 2, characterized in that: The construction is based on the fusion model of historical and real-time data, and generates reliable data input through time alignment, spatial matching and feature extraction, including: The trend characteristics are extracted from the fused data to capture the dynamics of data changes over time. The formula is as follows: ； in, It is a trend feature, which is used to capture the changing trend of data over time; is the fused data value, indicating time Data fusion results at each moment; is the time variable; Based on the extracted trend information, the dynamic changes of energy forms are analyzed, and the optimization model is constructed in combination with the spatial distribution characteristics: ； in, The efficiency of energy conversion; It is the dynamic characteristic of energy form conversion, which is used to describe the changes during the conversion process; For energy supply characteristics; Used to describe the nonlinear spatial characteristics of energy conversion; After the trend and conversion optimization is completed, the coordinated fusion of data is achieved through the following formula, and the fusion effect is further optimized by combining the time and space distribution characteristics: ； Among them, the efficiency of collaborative fusion of multimodal data combining temporal and spatial characteristics; For the Dynamic data strength of class data sources; For the Dynamic exception characteristics of class data sources; is the feature distribution in two-dimensional space; By integrating data, combining trends, cycles and random characteristics, we can predict future energy demand. The formula is as follows: ； in, For in time The future energy consumption forecast value, is the trend characteristic item; is the periodic characteristic term, ,in represents the periodic frequency; is a random characteristic term, which indicates the influence of random fluctuations. The specific formula is: ; Indicates the interference intensity of the external environment on energy demand; Improve data consistency and standardize historical fusion data and real-time data. The formula is as follows: ； in, It is the normalized data, used to mark data of different dimensions; is the original data; is the maximum and minimum value of the data; And further extract the trend characteristics, the formula is: ； in, It is a trend feature, which is used to extract the dynamic change trend of data; is the normalized data; Based on the fused data and combined with time series analysis, the trend characteristics, periodicity and random characteristics of future demand are determined, and finally the prediction formula is generated: ； in, Forecast values ​​for future demand; is the trend characteristic value; To fuse data; is the environmental parameter collected by the sensor.

4. The integrated energy supervision method according to claim 3, characterized in that: The real-time adjustment of equipment operating parameters based on historical and real-time environmental monitoring data includes: On the basis of dynamic data fusion, the path transmission efficiency is further optimized as follows: ； in, for path planning performance; is the path delay characteristic; For Node location; For gateway nodes location; After optimizing the path planning, the rationality of the data is further evaluated through the verification model, and the logical relationship between historical and real-time data is evaluated in combination with dynamic association: ； in, To verify the efficiency, it is used to evaluate the system and model in time Performance verification effect within the system; For the The noise characteristics of the data channel indicate that different channels The impact of noise; Further combined with the dynamic correlation model, the relationship between historical and real-time data is evaluated: ； in, It is a dynamic correlation indicator used to measure the dynamic relationship between the historical parameters of the equipment and the real-time parameters and environmental factors; For the The historical parameters of a device indicate the historical record characteristics of the device operation; is the real-time parameter of the jth device; is the environmental parameter of the jth device; Detect and remove parameter outliers; After completing the correlation analysis, the data storage is optimized with the help of the layered blockchain architecture: ； in, Storage efficiency for layered blockchains; It is a high-frequency data feature; Delay for high-frequency data transmission.

5. The integrated energy supervision method according to claim 4, characterized in that: The above-mentioned combination of historical and real-time data to predict future energy demand and generate dynamic scheduling plans accordingly to optimize equipment operation includes: The dispatch plan is generated based on the predicted energy consumption data and equipment operating parameters, which is specifically implemented through the following formula: ； in, is the calculation result of the scheduling plan; For time Equipment power; is the operating weight of the device; The formula for predicting future demand is: ； in, is the forecast value of future energy demand; For trend characteristics; To represent periodic characteristics; It is the result of the fusion of historical and real-time data; To predict future energy demand, the prediction formula is: The difference between the predicted and actual energy consumption data is corrected and expressed by the following formula: ； in, is the error correction value, indicating that The deviation between the internal prediction data and the actual data; For time The actual data value at the moment is obtained from the actual collected data; For time The predicted data value at the moment; The historical data and real-time data are standardized to eliminate the dimension difference. The standardization formula is: ； in, It is the normalized data, which is used to standardize data of different dimensions; For time The original data; are the maximum and minimum values ​​of the data.

6. The integrated energy supervision method according to claim 5, characterized in that: The blockchain technology is used to combine prediction and scheduling results to continuously optimize the control model and achieve closed-loop control and dynamic adjustment, including: The prediction and scheduling results are encrypted and stored based on blockchain technology, as shown below: ； in, Storage efficiency is used to measure the load balance of storage nodes; Time range optimized for storage; It is a high-frequency data characteristic; It is a low-frequency data characteristic; The dynamic calibration of the control model is achieved through the blockchain feedback mechanism, which is implemented through the following formula: ； in, The control capability for optimization is used to describe the dynamic optimization achieved by the system by adjusting the scheduling parameters; is the time period; For the Power time of each device changes; For the The operation weight of each device is calculated through real-time data collected by sensors; By detecting the difference between the predicted and actual energy consumption and correcting the prediction model, it is achieved through the following formula: ； in, Anomaly detection performance, which is used to evaluate the effectiveness and accuracy of anomaly detection in equipment operating status and environmental parameters; Abnormal fluctuations in equipment operating parameters; To indicate abnormal changes in environmental parameters.

7. An integrated energy supervision system, based on the integrated energy supervision method according to any one of claims 1 to 6, characterized in that: include: Data acquisition module, used to collect real-time status data of distributed energy through distributed sensors and smart meters; Data cleaning and processing module, used to clean the collected multimodal data and remove abnormal points; Data fusion module, used to fuse the temporal and spatial characteristics of historical data and real-time data; Energy characteristics extraction module, used to extract trend characteristics, periodic characteristics and random characteristics from fused data; Energy consumption forecasting module, used to predict future energy demand based on historical and real-time data; Dynamic scheduling module, used to generate dynamic scheduling plans and optimize equipment operation based on energy consumption forecast results; Blockchain storage module, used to encrypt and store prediction and scheduling data, based on high- and low-frequency tiered storage optimization; Anomaly detection and feedback module, used to evaluate abnormal fluctuations in equipment operating status and environmental parameters, and dynamically calibrate prediction models and scheduling models; The visualization and user interaction module is used to display the system operation status, prediction results and scheduling plans in real time, and supports users to adjust parameter input and modify strategies.

8. The integrated energy monitoring system according to claim 6, It is characterized in that in, The data cleaning and processing modules include: A noise filtering unit is used to remove data points that do not conform to the rules using an anomaly detection model; Normalization processing unit, used to standardize data from different sources and eliminate dimensional differences; Data completion unit, used to complete missing data points through interpolation and historical pattern matching; A data classification unit, used to classify data into key data and auxiliary data for processing; The data fusion module includes: The spatiotemporal alignment unit is used to align historical and real-time data and match spatial characteristics; A multimodal data fusion unit, which is used to combine data from different sources using mathematical models to generate comprehensive features; A characteristic weight allocation unit, used for allocating weights according to data characteristics; Among them, the blockchain storage module includes: A data encryption unit, used for encrypting stored data; A hierarchical storage unit for storing data in layers according to high-frequency and low-frequency characteristics; Blockchain network unit, used to implement data storage and management based on blockchain technology, including data traceability and storage verification; The storage optimization unit is used to dynamically adjust the storage policy based on the prediction and scheduling data priorities.

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, the steps of a comprehensive energy supervision method described in any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a comprehensive energy supervision method described in any one of claims 1 to 6 are implemented.

Images