Distributed energy management system based on multi-energy microgrid

Abstract

The invention discloses a distributed energy management system based on a multi-energy micro-grid, relates to the technical field of energy management, and solves the problems that the prior art is lack of considering the deviation between a prediction result and an actual result, so that the prediction data accuracy is low, and the energy consumption is low. And an effective solution when the actual production demand cannot meet the actual consumption demand is lacked, so that the accuracy and efficiency of the energy management system are low. The method comprises the steps of generating region priorities according to region data; generating a dynamic time window according to the energy data; generating predicted energy data according to the dynamic time window; an adjusting scheme is generated according to predicted energy data and regional priorities, energy management scheduling is realized according to the adjusting scheme, the size of a dynamic time window is adjusted by taking an error of a prediction result as a reference, so that the prediction result is more accurate, energy is dynamically distributed according to the regional priorities when supply cannot meet requirements, and the energy management efficiency is improved. And the accuracy and efficiency of the energy management system are improved.

Description

Technical Field

[0001] This application belongs to the field of energy management technology, specifically a distributed energy management system based on a multi-energy microgrid. Background Art

[0002] A multi-energy microgrid is a miniature autonomous power system that integrates "energy production, energy storage, and energy consumption". It has the characteristics of multiple energy forms, multiple load types, multiple power supply types, and multiple operating conditions. It can be flexibly configured according to resource distribution, user needs, and application scenarios to achieve reliable and economical operation; distributed energy refers to an energy comprehensive utilization system distributed at the user end. It uses all available resources, including renewable energy and fossil energy, to provide users with various forms of energy services such as cooling, heating, and electricity.

[0003] The existing technology (invention patent with publication number CN117236638B) discloses a canal microgrid distributed energy management system based on a multimodal network. The energy prediction module obtains all ways of obtaining electricity in the canal area, and after comprehensively analyzing the electricity supply status of all acquisition methods, predicts the future power supply in the canal area. The demand prediction module obtains historical data and load characteristics based on the database to predict the future electricity demand in the canal area. The judgment module compares the future power supply in the canal area with the future electricity demand. If it is judged that the supply of electricity in the canal area exceeds the demand, the decision module chooses to store or sell the remaining electricity based on the future electricity price. If it is judged that the supply of electricity in the canal area is less than the demand, the decision module needs to formulate a management strategy.

[0004] The above scheme predicts the future power supply and power demand in the canal area and then manages the power supply. However, the above scheme fails to consider the deviation between the predicted results and the actual results, resulting in low accuracy of the predicted data. It also fails to consider effective solutions when actual production demand cannot meet actual consumption demand, resulting in low accuracy and efficiency of the energy management system. Therefore, the energy management system still needs further improvement. Summary of the Invention

[0005] The present application aims to solve at least one of the technical problems existing in the prior art; to this end, the present application proposes a distributed energy management system based on a multi-energy microgrid, which is used to solve the technical problems that the prior art lacks consideration of the deviation between the predicted results and the actual results, resulting in low accuracy of the predicted data, and lacks consideration of effective solutions when the actual production demand cannot meet the actual consumption demand, resulting in low accuracy and efficiency of the energy management system.

[0006] To achieve the above-mentioned objectives, the first aspect of the present application provides a distributed energy management system based on a multi-energy microgrid, comprising: a data acquisition module, a data analysis module, an early warning module, and a database; the data acquisition module is electrically and / or communicatively connected to the data analysis module; the data analysis module is electrically and / or communicatively connected to the early warning module; the database is electrically and / or communicatively connected to the data acquisition module, the data analysis module, and the early warning module, respectively;

[0007] The data acquisition module is used to acquire regional data and energy data through data acquisition equipment;

[0008] The data analysis module generates regional priorities based on regional data; generates dynamic time windows based on energy data; generates predicted energy data based on the dynamic time windows; generates an adjustment plan based on the predicted energy data and regional priorities, and implements energy management scheduling based on the adjustment plan;

[0009] The early warning module: makes prompts according to the alarm signal and contacts the management personnel;

[0010] The database is used to store data collected by the data collection device and store historical data required for training the model.

[0011] Through the above steps, this application flexibly adjusts the size of the dynamic time window based on the prediction error feedback mechanism, aiming to significantly improve the accuracy of the prediction results; in addition, in the face of a situation where supply cannot meet demand, a dynamic energy allocation strategy is implemented according to regional priority, effectively enhancing the accuracy and operational efficiency of the energy management system.

[0012] Furthermore, the regional priority QY i Satisfies the following formula:

[0013]

[0014] Among them, α1 is the exponential coefficient, α1>1; α2 and α3 are proportional coefficients, α2 and α3∈(0,1); RMG i It represents the value after normalization of the i-th region; max() and min() represent the maximum and minimum value operations; i represents the number corresponding to the region ID, GD represents the critical facility level, RM represents the population density, and GDP represents the unit energy consumption.

[0015] Furthermore, generating a dynamic time window based on energy data includes:

[0016] Real-time acquisition of regional data, energy data, and corresponding historical forecast energy data; the energy data includes a number of actual energy values; the historical forecast energy data includes a number of historical energy forecast values;

[0017] Calculate the energy error NW based on the energy forecast value NYZ and the actual energy value NSZ i,j,t ; The energy error NW i,j,t Satisfies the following formula:

[0018] Among them, j represents the energy number and t represents the time number;

[0019] Calculate the average energy error PJNW based on the energy errors NW of the last N times i,j ; The average energy error PJNW i,j Satisfies the following formula:

[0020] Where n represents the number of the most recent N energy errors; N is an integer, N>1;

[0021] Determine whether the average energy error is within the error range WF; if yes, do nothing;

[0022] No, generate dynamic time windows based on the average energy error.

[0023] Furthermore, generating a dynamic time window according to the average energy error includes:

[0024] Get regional data, average energy error PJNW i,j and an error range WF; the error range includes a maximum error range WFD and a minimum error range WFX;

[0025] Based on the average energy error PJNW i,j The error deviation degree WP is calculated based on the error range WF; the error deviation degree WP satisfies the following formula:

[0026]

[0027] Based on the average energy error PJNW i,j The sign function F is calculated by summing the error range WF; the sign function F satisfies the following formula:

[0028]

[0029] Calculate the dynamic time window W based on the sign function F and the error deviation degree WP i,t+1 ; The dynamic time window W i,t+1 Satisfies the following formula:

[0030] W i，t+1 =clip(W i，t +ΔW×F×|WP| β , W min , Wmax ); Among them, clip() is to control the dynamic time window to the maximum value of the time window W max and the minimum value of the time window W min ΔW is the basic adjustment step, β is the exponential coefficient, and β>0.

[0031] Furthermore, generating predicted energy data according to the dynamic time window includes:

[0032] Acquire regional data, device data, user data, dynamic time windows, and their corresponding energy data and environmental data; the energy data includes several energy production values and several energy consumption values; the environmental data includes temperature, humidity, wind speed, and weather forecast data; the regional data includes regional ID and regional priority;

[0033] Integrate the dynamic time windows and their corresponding energy production values, equipment data, and environmental data into several energy production forecast sequences;

[0034] Inputting a plurality of energy production forecast sequences into an energy production estimation model to obtain a plurality of energy estimated production values; the energy production estimation model is constructed using a machine learning model;

[0035] Integrate the region ID and its corresponding dynamic time window, environmental data, user data, and several energy consumption values into several energy consumption prediction sequences;

[0036] Inputting a plurality of energy consumption prediction sequences into an energy consumption estimation model to obtain a plurality of energy consumption estimation values; the energy consumption estimation model is constructed by a machine learning model.

[0037] Furthermore, the energy production estimation model is constructed through a machine learning model, including:

[0038] Obtain several historical dynamic time windows and their corresponding equipment data and environmental data as well as several historical energy production values;

[0039] Integrate several historical dynamic time windows and their corresponding equipment data and environmental data as well as several historical energy production values into several historical energy production forecast sequences;

[0040] Several historical energy production forecast series are divided into training set, test set and validation set;

[0041] Select a machine learning model as the base model;

[0042] Train the basic model using the training set, and adjust the learning rate or other hyperparameters on the validation set to obtain a pre-trained model;

[0043] By verifying the pre-trained model on the test set, we finally obtain an energy production estimation model whose input is several energy production prediction sequences and output is several energy estimated production values.

[0044] Furthermore, the energy consumption estimation model is constructed through a machine learning model, including:

[0045] Obtain several historical area IDs and their corresponding dynamic time windows, environmental data, user data, and several historical energy consumption values;

[0046] Integrate several historical area IDs and their corresponding dynamic time windows, environmental data, user data, and several historical energy consumption values into several historical energy consumption prediction sequences;

[0047] Several historical energy consumption forecast series are divided into training set, test set and validation set;

[0048] Select a machine learning model as the base model;

[0049] Train the basic model using the training set, and adjust the learning rate or other hyperparameters on the validation set to obtain a pre-trained model;

[0050] By verifying the pre-trained model on the test set, we finally obtain an energy consumption estimation model whose input is a number of energy consumption prediction sequences and whose output is a number of energy consumption estimated values corresponding to a number of regional IDs.

[0051] Furthermore, generating a regulation plan based on the predicted energy data and regional priorities includes:

[0052] Obtaining predicted energy data and regional data; the predicted energy data includes a number of estimated energy production values, a number of estimated energy consumption values, and a number of energy storage values; the regional data includes a region ID and a region priority;

[0053] A plurality of overall estimated energy consumption values are obtained by summing a plurality of estimated energy consumption values of a plurality of regions;

[0054] A plurality of estimated energy supply values are obtained by summing a plurality of estimated energy production values and a plurality of energy storage values;

[0055] determining whether a number of estimated overall energy consumption values are greater than a number of estimated energy supply values;

[0056] Yes, generating an adjustment plan based on a number of estimated overall energy consumption values and a number of estimated energy supply values;

[0057] No, do nothing.

[0058] Furthermore, generating an adjustment plan based on a plurality of estimated overall energy consumption values and a plurality of estimated energy supply values includes:

[0059] Obtaining a number of overall energy estimated consumption values, a number of energy estimated supply values, and regional priorities;

[0060] A plurality of energy estimated difference values are obtained by performing difference calculation between a plurality of overall energy estimated consumption values and a plurality of energy estimated supply values;

[0061] Sort the regional priorities from large to small to obtain the priority sequence YXL;

[0062] The estimated energy consumption values of the region ID corresponding to the highest value in the priority sequence are transmitted; the highest priority value of the region is deleted from the priority sequence YXL to adjust the priority sequence;

[0063] Calculating a number of energy regulation consumption values according to a regulation priority sequence;

[0064] A plurality of energy transmissions are performed to the corresponding area ID according to a plurality of energy adjustment consumption values, and an energy supply reduction alarm signal is generated.

[0065] Furthermore, the calculation of several energy regulation consumption values according to the regulation priority sequence includes:

[0066] Get some energy estimate differences NYC j , adjust the priority sequence corresponding to several regional priorities QY k , as well as the regional ID and its corresponding estimated energy consumption values NYX k,j ;

[0067] Based on energy estimate difference NYC j and regional priority QY k Calculate the energy reduction QNJ for several regions k,j ; The regional energy reduction QNJ k,j Satisfies the following formula:

[0068]

[0069] A number of energy adjustment consumption values are obtained by performing difference calculation between a number of estimated energy consumption values and a number of regional energy reduction amounts; wherein k represents the region number in the adjustment priority sequence.

[0070] Through the above steps, when the supply is insufficient to meet the consumption demand, the present application prioritizes ensuring that the area with the highest regional priority obtains the full energy supply; at the same time, for other areas, appropriate energy reductions are implemented in some areas based on regional priority to ensure that each area can obtain a certain amount of energy supply; this not only guarantees the energy supply ratio of high-priority areas, but also further improves the overall efficiency of the energy management system.

[0071] Compared with the prior art, the present invention has the following advantages:

[0072] 1. This application generates regional priorities based on regional data; generates dynamic time windows based on energy data; generates predicted energy data based on the dynamic time windows; generates adjustment plans based on the predicted energy data and regional priorities, and implements energy management scheduling based on the adjustment plans. The size of the dynamic time window is adjusted based on the error of the prediction results to make the prediction results more accurate. At the same time, when supply cannot meet demand, energy is dynamically allocated according to regional priorities, thereby improving the accuracy and efficiency of the energy management system.

[0073] 2. This application calculates the errors between the predicted data and the corresponding actual data, and adjusts the dynamic time window during prediction based on historical errors, so that the predicted data is more in line with the actual situation. At the same time, it considers avoiding the situation where the time window is too short or too long when the time window is dynamically updated, thereby improving the prediction accuracy and stability of the energy management system.

[0074] 3. This application ensures that each area has energy supply by supplying all the energy to the area with the highest regional priority when the supply cannot meet the consumption. At the same time, in the remaining areas, energy is reduced in several areas according to regional priority, so that each area has energy supply, ensuring the energy supply ratio of high-priority areas and improving the efficiency of the energy management system. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0076] Figure 1 This is a schematic diagram of the principle of a distributed energy management system based on a multi-energy microgrid of this application;

[0077] Figure 2 This is a flow chart of the distributed energy management method based on multi-energy microgrid of this application;

[0078] Figure 3 Generate a flow chart for the mediation solution for this application. DETAILED DESCRIPTION

[0079] The following will clearly and completely describe the technical solutions of this application in conjunction with the embodiments. Obviously, the embodiments described are only a part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0080] See also Figure 1-Figure 2 In a first aspect, an embodiment of the present application provides a distributed energy management system based on a multi-energy microgrid, comprising: a data acquisition module, a data analysis module, an early warning module, and a database; the data acquisition module is electrically and / or communicatively connected to the data analysis module; the data analysis module is electrically and / or communicatively connected to the early warning module; the database is electrically and / or communicatively connected to the data acquisition module, the data analysis module, and the early warning module, respectively;

[0081] Data acquisition module: acquires regional data and energy data through data acquisition equipment; data acquisition equipment includes various sensors, etc.

[0082] Data analysis module: Generates regional priorities based on regional data. Regional priorities refer to the priority levels corresponding to several regional IDs. Generates dynamic time windows based on energy data. Dynamic time windows refer to the size of the prediction sequence input each time a prediction is made. Generates predicted energy data based on the dynamic time windows. Predicted energy data refers to predicted energy production values and predicted energy consumption values. Generates adjustment plans based on predicted energy data and regional priorities, and implements energy management scheduling based on the adjustment plans.

[0083] Early warning module: issues prompts based on alarm signals and contacts management personnel; alarm signals include energy supply reduction alarm signals, etc.

[0084] The database is used to store data collected by data acquisition devices and store historical data required for training models.

[0085] In this embodiment, the regional priority QY i Satisfies the following formula:

[0086]

[0087] Among them, α1 is an exponential coefficient, α1>1, and the specific value is set based on experience. The setting of α1 is to make the priority of the area corresponding to the critical facility level higher when calculating the regional priority, amplifying the influence of high-level facilities. For example, the priority of hospitals is higher than that of residential areas. In this embodiment, α1=1.5, and the critical facility level is represented as a discrete value from 1 to 5; α2 and α3 are proportional coefficients, α2 and α3∈(0,1), and the specific values are set based on experience. In this embodiment, α2 and α3 are set to 0.8 and 0.5. When the population density is 1, it means that the population density of the area is the maximum among all areas, and the maximum gain of the regional priority affected by population density is 80%. By synergizing with the critical facility level, it reflects the superposition risk of high security level and high population density; α3 is set to adjust the influence of the economic efficiency item; RMG i It represents the value after normalization of the i-th region; max() and min() represent the maximum and minimum value operations; i represents the number corresponding to the region ID, GD represents the critical facility level, RM represents the population density, and GDP represents the unit energy consumption; when the critical facility level, population density and unit energy consumption in a region gradually increase, the regional priority of the region will also increase accordingly.

[0088] This embodiment obtains data from several regions and calculates regional priorities using multi-source data. It not only considers the impact of each data on regional priority, but also considers the synergy between multiple data, reflecting the overlapping risks of high security levels and high population density, so as to provide accurate data support for subsequent energy management and improve the accuracy of the energy management system.

[0089] Generating a dynamic time window based on energy data in this embodiment includes:

[0090] Real-time acquisition of regional data, energy data, and their corresponding historical forecast energy data; energy data includes several actual energy values; historical forecast energy data includes several historical energy forecast values;

[0091] Calculate the energy error NW based on the energy forecast value NYZ and the actual energy value NSZ i,j,t ; The energy error NW i,j,t Satisfies the following formula:

[0092] Where, j represents the energy number and t represents the time number. The larger the difference between the historical energy forecast value and the actual value of the energy, the worse the forecast effect is, which will increase the error of the energy.

[0093] Calculate the average energy error PJNW based on the energy errors NW of the last N times i,j; The average energy error PJNW i,j Satisfies the following formula:

[0094] Where n represents the number of the most recent N energy errors; that is, the most recent N energy errors are selected in sequence. If the most recent N energy errors are all large, the average energy error will also increase. N is an integer, N>1, and the specific value is set based on experience. In this embodiment, N is set to 3. The setting of N is to smooth the error effect caused by instantaneous fluctuations.

[0095] Determine whether the average energy error is within the error range WF; if yes, do nothing;

[0096] No, generate dynamic time windows based on the average energy error.

[0097] In this embodiment, generating a dynamic time window according to the average energy error includes:

[0098] Get regional data, average energy error PJNW i,j and error range WF; the error range includes the maximum error range WFD and the minimum error range WFX;

[0099] Based on the average energy error PJNW i,j The error deviation degree WP is calculated based on the error range WF; the error deviation degree WP satisfies the following formula:

[0100] WP represents the ratio of the error exceeding the threshold. When the system is in steady state, that is, the average energy error is within the error range, the window length change is 0, avoiding meaningless time window fluctuations.

[0101] Based on the average energy error PJNW i,j The sign function F is calculated by summing the error range WF; the sign function F satisfies the following formula:

[0102] Through F, dynamic time windows can be applied to scenarios where error fluctuations are large or frequent adjustments need to be avoided;

[0103] Calculate the dynamic time window W based on the sign function F and the error deviation degree WP i,t+1 ; The dynamic time window W i,t+1 Satisfies the following formula:

[0104] W i，t+1 =clip(W i，t +ΔW×F×|WP| β , W min , W max); When the average energy error exceeds the maximum value of the error range, more historical data needs to be considered when considering the prediction again, that is, the dynamic time window will increase accordingly, so that the prediction result is more accurate; Among them, clip() is to control the dynamic time window to the maximum value of the time window W max and the minimum value of the time window W min Between; W max With W min is the value range of the preset time window, and the specific value is set according to experience. In this embodiment, W max With W min It is set to 72 hours and 6 hours to avoid the situation where the time window is too short or too long when the time window is dynamically updated; ΔW represents the basic adjustment step, and the specific value is set according to experience. In this embodiment, ΔW is set to 1 hour; β is the exponential coefficient, β>0, and the specific value is set according to experience. In this embodiment, β is set to 1.5. The setting of β is to control the sensitivity of the adjustment amplitude. When β=1, the greater the deviation of the average energy error, the greater the adjustment amplitude of the dynamic time window; when β>1, the adjustment amplitude of the dynamic time window is more sensitive to large deviations of the average energy error; when β<1, the adjustment amplitude of the dynamic time window is smoother.

[0105] This embodiment calculates the error between predicted and actual data, using historical errors as a benchmark to adjust the dynamic time window during prediction, ensuring that the predicted data more closely matches actual conditions. Furthermore, to prevent the time window from being too short or too long during dynamic updates, the adjustment process is optimized, thereby improving the energy management system's forecast accuracy and stability.

[0106] In this embodiment, generating predicted energy data according to a dynamic time window includes:

[0107] Acquire regional data, device data, user data, dynamic time windows, and their corresponding energy and environmental data; energy data includes several energy production values and several energy consumption values; environmental data includes temperature, humidity, wind speed, and weather forecast data; regional data includes region ID and region priority; and device data includes several energy production devices, such as production power and device status.

[0108] Integrate dynamic time windows and their corresponding energy production values, equipment data, and environmental data into several energy production forecast sequences. Energy production forecasts consider environmental data, historical energy production data, and equipment data. This differs from energy consumption forecasts, which consider environmental data, user data, and historical consumption patterns in different regions.

[0109] Inputting a number of energy production forecast sequences into an energy production estimation model to obtain a number of energy production estimates; the energy production estimation model is constructed using a machine learning model;

[0110] Integrate the region ID and its corresponding dynamic time window, environmental data, user data, and several energy consumption values into several energy consumption prediction sequences;

[0111] Input a number of energy consumption prediction sequences into an energy consumption estimation model to obtain a number of energy consumption estimation values; the energy consumption estimation model is constructed through a machine learning model.

[0112] This embodiment adjusts the dynamic time window, adaptively selects the prediction sequence, and uses several corresponding pre-trained prediction models to perform energy production and energy consumption predictions, thereby improving the accuracy of the predictions and providing accurate data support for subsequent energy management.

[0113] The energy production estimation model in this embodiment is constructed using a machine learning model, including:

[0114] Obtain several historical dynamic time windows and their corresponding equipment data and environmental data as well as several historical energy production values;

[0115] Integrate several historical dynamic time windows and their corresponding equipment data and environmental data as well as several historical energy production values into several historical energy production forecast sequences;

[0116] Several historical energy production forecast series are divided into training set, test set and validation set; the ratio between training set, test set and validation set is 7:2:1;

[0117] Select a machine learning model as the base model; machine learning models include LSTM models, etc.

[0118] Train the basic model using the training set, and adjust the learning rate or other hyperparameters on the validation set to obtain a pre-trained model;

[0119] By verifying the pre-trained model on the test set, we finally obtain an energy production estimation model whose input is several energy production prediction sequences and output is several energy estimated production values.

[0120] The energy consumption estimation model is constructed using a machine learning model, including:

[0121] Obtain several historical area IDs and their corresponding dynamic time windows, environmental data, user data, and several historical energy consumption values;

[0122] Integrate several historical area IDs and their corresponding dynamic time windows, environmental data, user data, and several historical energy consumption values into several historical energy consumption prediction sequences;

[0123] Several historical energy consumption forecast series are divided into training set, test set and validation set; the ratio between training set, test set and validation set is 7:2:1;

[0124] Select a machine learning model as the base model; machine learning models include LSTM models, etc.

[0125] Train the basic model using the training set, and adjust the learning rate or other hyperparameters on the validation set to obtain a pre-trained model;

[0126] By verifying the pre-trained model on the test set, we finally obtain an energy consumption estimation model whose input is a number of energy consumption prediction sequences and whose output is a number of energy consumption estimated values corresponding to a number of regional IDs.

[0127] See also Figure 3 In this embodiment, the adjustment plan is generated based on the predicted energy data and regional priorities, including:

[0128] Obtaining predicted energy data and regional data; the predicted energy data includes a number of estimated energy production values, a number of estimated energy consumption values, and a number of energy storage values; the regional data includes a region ID and a region priority;

[0129] A plurality of overall estimated energy consumption values are obtained by summing a plurality of estimated energy consumption values of a plurality of regions;

[0130] A plurality of estimated energy supply values are obtained by summing a plurality of estimated energy production values and a plurality of energy storage values;

[0131] determining whether a number of estimated overall energy consumption values are greater than a number of estimated energy supply values;

[0132] Yes, generating an adjustment plan based on a number of estimated overall energy consumption values and a number of estimated energy supply values;

[0133] No, do nothing.

[0134] In this embodiment, the adjustment scheme is generated based on a number of estimated overall energy consumption values and a number of estimated energy supply values, including:

[0135] Obtaining a number of overall energy estimated consumption values, a number of energy estimated supply values, and regional priorities;

[0136] A plurality of energy estimated difference values are obtained by performing difference calculation between a plurality of overall energy estimated consumption values and a plurality of energy estimated supply values;

[0137] Sort the regional priorities from large to small to obtain the priority sequence YXL;

[0138] The estimated energy consumption values of the region ID corresponding to the highest value in the priority sequence are transmitted; the highest priority value of the region is deleted from the priority sequence YXL to adjust the priority sequence; the region corresponding to the highest priority is fully supplied to ensure that the energy consumption of the region can meet the demand;

[0139] Calculating a number of energy regulation consumption values according to a regulation priority sequence;

[0140] A plurality of energy transmissions are performed to the corresponding area ID according to a plurality of energy adjustment consumption values, and an energy supply reduction alarm signal is generated.

[0141] Several energy regulation consumption values are calculated based on the regulation priority sequence, including:

[0142] Get some energy estimate differences NYC j , adjust the priority sequence corresponding to several regional priorities QY k , as well as the regional ID and its corresponding estimated energy consumption values NYX k,j ;

[0143] Based on energy estimate difference NYC j and regional priority QY k Calculate the energy reduction QNJ for several regions k,j ; The regional energy reduction QNJ k,j Satisfies the following formula:

[0144] Energy reduction is performed on several regions according to their regional priorities. The region with the highest regional priority has the lowest reduction, while the region with the lowest regional priority has the highest reduction. The reduction amount for each region is equal to the difference in energy estimates.

[0145] A number of energy adjustment consumption values are obtained by performing difference calculation between a number of estimated energy consumption values and a number of regional energy reduction amounts; wherein k represents the region number in the adjustment priority sequence.

[0146] Some of the data in the above formula are calculated by removing the dimensions and taking their numerical values. The formula is a formula that is closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0147] The working principle of this application is: by acquiring regional data and energy data; generating regional priorities based on regional data; generating dynamic time windows based on energy data; generating predicted energy data based on the dynamic time windows; generating adjustment plans based on predicted energy data and regional priorities, and realizing energy management scheduling based on the adjustment plans; making prompts based on alarm signals and contacting management personnel, adjusting the size of the dynamic time window based on the error of the prediction results, so that the prediction results are more accurate, and at the same time, when supply cannot meet demand, energy is dynamically allocated according to regional priorities, thereby improving the accuracy and efficiency of the energy management system, avoiding the problem that the existing technology lacks consideration of the deviation between the prediction results and the actual results, resulting in low accuracy of the prediction data, and lacks consideration of effective solutions when actual production demand cannot meet actual consumption demand, resulting in low accuracy and efficiency of the energy management system.

[0148] The above embodiments are only used to illustrate the technical method of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present application.

Claims

1. A distributed energy management system based on multi-energy microgrid, characterized by: include: Interconnected data acquisition module and data analysis module; The data acquisition module is used to acquire regional data and energy data through data acquisition equipment; The data analysis module generates regional priorities based on regional data and generates dynamic time windows based on energy data; Generate predicted energy data based on dynamic time windows; generate adjustment plans based on predicted energy data and regional priorities, and determine energy management scheduling based on the adjustment plans.

2. The distributed energy management system based on multi-energy microgrid according to claim 1 is characterized in that: The regional priority QY i Satisfies the following formula: Among them, α1 is the exponential coefficient, α1>1; α2 and α3 are proportional coefficients, α2 and α3∈(0,1); RMG i It represents the value after normalization of the i-th region; max() and min() represent the maximum and minimum value operations; i represents the number corresponding to the region ID, GD represents the critical facility level, RM represents the population density, and GDP represents the unit energy consumption.

3. The distributed energy management system based on multi-energy microgrid according to claim 1 is characterized in that: Generating a dynamic time window according to energy data includes: Real-time acquisition of regional data, energy data, and corresponding historical forecast energy data; the energy data includes a number of actual energy values; the historical forecast energy data includes a number of historical energy forecast values; Calculate the energy error NW based on the energy forecast value NYZ and the actual energy value NSZ i,j,t ; The energy error NW i,j,t Satisfies the following formula: Among them, j represents the energy number and t represents the time number; Calculate the average energy error PJNW based on the energy errors NW of the last N times i,j ; The average energy error PJNW i,j Satisfies the following formula: Where n represents the number of the most recent N energy errors; N is an integer, N>1; Determine whether the average energy error is within the error range WF; if yes, do nothing; No, generate dynamic time windows based on the average energy error.

4. The distributed energy management system based on multi-energy microgrid according to claim 3 is characterized in that: Generating a dynamic time window according to the average energy error includes: Get regional data, average energy error PJNW i,j and an error range WF; the error range includes a maximum error range WFD and a minimum error range WFX; Based on the average energy error PJNW i,j The error deviation degree WP is calculated based on the error range WF; the error deviation degree WP satisfies the following formula: Based on the average energy error PJNW i,j The sign function F is calculated by summing the error range WF; the sign function F satisfies the following formula: Calculate the dynamic time window W based on the sign function F and the error deviation degree WP i,t+1 ; The dynamic time window W i,t+1 Satisfies the following formula: W i，t+1 =clip(W i，t +ΔW×F×|WP| β , W min , W max ); Among them, clip() is to control the dynamic time window to the maximum value of the time window W max and the minimum value of the time window W min ΔW is the basic adjustment step, β is the exponential coefficient, and β>0.

5. The distributed energy management system based on multi-energy microgrid according to claim 1 is characterized in that: The generating of predicted energy data according to the dynamic time window includes: Acquire regional data, device data, user data, dynamic time windows, and their corresponding energy data and environmental data; the energy data includes several energy production values and several energy consumption values; the environmental data includes temperature, humidity, wind speed, and weather forecast data; the regional data includes regional ID and regional priority; Integrate the dynamic time windows and their corresponding energy production values, equipment data, and environmental data into several energy production forecast sequences; Inputting a plurality of energy production forecast sequences into an energy production estimation model to obtain a plurality of energy estimated production values; the energy production estimation model is constructed using a machine learning model; Integrate the region ID and its corresponding dynamic time window, environmental data, user data, and several energy consumption values into several energy consumption prediction sequences; Inputting a plurality of energy consumption prediction sequences into an energy consumption estimation model to obtain a plurality of energy consumption estimation values; the energy consumption estimation model is constructed by a machine learning model.

6. The distributed energy management system based on multi-energy microgrid according to claim 5 is characterized in that: The energy production estimation model is constructed using a machine learning model, including: Obtain several historical dynamic time windows and their corresponding equipment data and environmental data as well as several historical energy production values; Integrate several historical dynamic time windows and their corresponding equipment data and environmental data as well as several historical energy production values into several historical energy production forecast sequences; Several historical energy production forecast series are divided into training set, test set and validation set; Select a machine learning model as the base model; Train the basic model using the training set, and adjust the learning rate or other hyperparameters on the validation set to obtain a pre-trained model; By verifying the pre-trained model on the test set, we finally obtain an energy production estimation model whose input is several energy production prediction sequences and output is several energy estimated production values.

7. The distributed energy management system based on multi-energy microgrid according to claim 5 is characterized in that: The energy consumption estimation model is constructed through a machine learning model, including: Obtain several historical area IDs and their corresponding dynamic time windows, environmental data, user data, and several historical energy consumption values; Integrate several historical area IDs and their corresponding dynamic time windows, environmental data, user data, and several historical energy consumption values into several historical energy consumption prediction sequences; Several historical energy consumption forecast series are divided into training set, test set and validation set; Select a machine learning model as the base model; Train the basic model using the training set, and adjust the learning rate or other hyperparameters on the validation set to obtain a pre-trained model; By verifying the pre-trained model on the test set, we finally obtain an energy consumption estimation model whose input is a number of energy consumption prediction sequences and whose output is a number of energy consumption estimated values corresponding to a number of regional IDs.

8. The distributed energy management system based on multi-energy microgrid according to claim 1 is characterized in that: The generation of a regulation plan based on predicted energy data and regional priorities includes: Obtaining predicted energy data and regional data; the predicted energy data includes a number of estimated energy production values, a number of estimated energy consumption values, and a number of energy storage values; the regional data includes a region ID and a region priority; A plurality of overall estimated energy consumption values are obtained by summing a plurality of estimated energy consumption values of a plurality of regions; A plurality of estimated energy supply values are obtained by summing a plurality of estimated energy production values and a plurality of energy storage values; determining whether a number of estimated overall energy consumption values are greater than a number of estimated energy supply values; Yes, generating an adjustment plan based on a number of estimated overall energy consumption values and a number of estimated energy supply values; No, do nothing.

9. The distributed energy management system based on multi-energy microgrid according to claim 8 is characterized in that: The generating of the adjustment scheme according to the plurality of estimated overall energy consumption values and the plurality of estimated energy supply values includes: Obtaining a number of overall energy estimated consumption values, a number of energy estimated supply values, and regional priorities; A plurality of energy estimated difference values are obtained by performing difference calculation between a plurality of overall energy estimated consumption values and a plurality of energy estimated supply values; Sort the regional priorities from large to small to obtain the priority sequence YXL; The estimated energy consumption values of the region ID corresponding to the highest value in the priority sequence are transmitted; the highest priority value of the region is deleted from the priority sequence YXL to adjust the priority sequence; Calculating a number of energy regulation consumption values according to a regulation priority sequence; A plurality of energy transmissions are performed to the corresponding area ID according to a plurality of energy adjustment consumption values, and an energy supply reduction alarm signal is generated.

10. The distributed energy management system based on multi-energy microgrid according to claim 9 is characterized in that: The calculation of a plurality of energy regulation consumption values according to the regulation priority sequence includes: Get some energy estimate differences NYC j , adjust the priority sequence corresponding to several regional priorities QY k , as well as the regional ID and its corresponding estimated energy consumption values NYX k,j ; Based on energy estimate difference NYC j and regional priority QY k Calculate the energy reduction QNJ for several regions k,j ; The regional energy reduction QNJ k,j Satisfies the following formula: A number of energy adjustment consumption values are obtained by performing difference calculation between a number of estimated energy consumption values and a number of regional energy reduction amounts; wherein k represents the region number in the adjustment priority sequence.

Images