Energy consumption optimization management system and method thereof

Abstract

The invention discloses an energy consumption optimization management system and method, and particularly relates to the field of energy management, and the system comprises an intelligent information collection terminal collection module, an energy data preprocessing module, an intelligent prediction module, a multi-target optimization scheduling module, a real-time feedback module, an abnormity alarm module, and a scheduling scheme output module. According to the energy consumption optimization management system and the method thereof, data are acquired through the intelligent information acquisition terminal acquisition module by means of various intelligent information acquisition terminals, so that efficient management of energy consumption is realized; dynamic prediction, optimal scheduling and efficient management of energy consumption are realized through the intelligent prediction module, and the accuracy of energy consumption trend prediction is improved. And a multi-target optimization scheduling scheme is obtained by applying a multi-target optimization algorithm through the multi-target optimization scheduling module, so that the scientificity and rationality of energy scheduling are improved, and energy waste caused by unreasonable scheduling is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of energy management, and more specifically, to an energy consumption optimization management system and method thereof. Background Art

[0002] With the continuous growth of global energy consumption and the complexity of energy structure, the intelligence and refinement of energy management have become particularly important; the industrial field generally faces the problems of high investment, high consumption and high emissions, the energy efficiency is not ideal, and the energy cost remains high. Promoting the greening and intelligence of industry and reducing industrial energy consumption have become an inevitable trend in my country's industrial development.

[0003] The existing energy management technology solutions are usually composed of basic energy metering instruments, manual meter reading and recording systems, and simple data analysis software. Energy consumption data is collected manually on a regular basis, and then simple data statistics and comparative analysis are performed using software to generate some regular reports. Based on these reports, energy usage can be roughly judged and links with high energy consumption can be checked.

[0004] However, in actual use, it still has some shortcomings, such as the lack of automated control means, which makes it impossible to adjust energy-consuming equipment and thus achieve immediate optimization of energy consumption; energy scheduling lacks flexibility and cannot respond to changes in the external environment in real time, making it difficult to accurately grasp the inherent laws of energy consumption changes under different working conditions and at different time periods; manual meter reading is cumbersome and inefficient, and when faced with large-scale, multi-regional energy metering work, meter reading errors and omissions are prone to occur, resulting in deviations in the analysis results. Summary of the invention

[0005] In order to overcome the above defects of the prior art, the present invention provides an energy consumption optimization management system and method thereof, which solves the problems raised in the above background technology through the following scheme.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The energy consumption optimization management system is characterized by comprising a system operation database, a system central processing module and a user information terminal, and also comprising:

[0008] Intelligent information collection terminal collection module: responds to the intelligent information collection terminal deployed in the energy consumption detection area to obtain real-time energy consumption data;

[0009] Energy data preprocessing module: used to preprocess the real-time energy consumption data transmitted by the intelligent information collection terminal collection module, and the preprocessing operation is used to obtain the key energy calibration group corresponding to the real-time energy consumption data;

[0010] Intelligent prediction module: used to obtain an intelligent prediction model, and obtain energy consumption characteristics according to the key energy calibration group transmitted by the energy data preprocessing module through the intelligent prediction model;

[0011] Multi-objective optimization scheduling module: Based on the real-time energy consumption data transmitted by the intelligent information collection terminal acquisition module and the energy consumption characteristics transmitted by the intelligent prediction module, a multi-objective optimization scheduling scheme is obtained through a multi-objective optimization algorithm;

[0012] Real-time feedback module: In response to the intelligent information collection terminal, it monitors the energy-consuming equipment according to the multi-objective optimization scheduling scheme transmitted by the multi-objective optimization scheduling module to obtain energy optimization feedback data;

[0013] Abnormal alarm module: used to perform abnormal judgment operation on the energy optimization feedback data transmitted by the real-time feedback module. The abnormal judgment operation is used to warn of abnormal situations corresponding to the energy optimization feedback data;

[0014] Scheduling scheme output module: based on the energy consumption characteristics transmitted by the intelligent prediction module, the multi-objective optimization scheduling scheme transmitted by the multi-objective optimization scheduling module, and the energy optimization feedback data transmitted by the real-time feedback module, the energy consumption optimization report is obtained and output to the user through a visual interface;

[0015] The system operation database includes all data texts of the energy consumption optimization management system, and collects the information text output by each module in real time. The system central processing module is used for the information text instructions output by each module in the central control system, and the user information terminal is an information output device for receiving the energy consumption optimization management system.

[0016] Preferably, the energy data preprocessing module obtains the key energy calibration group corresponding to the real-time energy consumption data, specifically including:

[0017] Perform feature extraction and correlation analysis on real-time energy consumption data, and extract key energy information based on correlation coefficient thresholds;

[0018] The feature extraction and association analysis operations are used to obtain energy consumption indicators, statistical indicators, and correlation coefficients corresponding to real-time energy consumption data.

[0019] Preferably, the energy data preprocessing module, feature extraction and association analysis operations specifically include:

[0020] Energy consumption indicators are key performance parameters corresponding to different types of energy consumption data;

[0021] Statistical indicators are a combination of different types of energy data;

[0022] The statistical indicators include the corresponding mean, variance, maximum value, and minimum value between each energy consumption indicator combination;

[0023] The correlation coefficient threshold is a pre-set benchmark for screening key energy information.

[0024] Preferably, the intelligent prediction module obtains energy consumption characteristics, specifically including:

[0025] According to the key energy calibration group, a prediction operation is performed on the key energy information to obtain key energy consumption characteristics corresponding to the key energy information;

[0026] The key energy consumption characteristics are the target consumption trends corresponding to the key energy information;

[0027] In the preset system operation database, an intelligent prediction model corresponding to the key energy consumption characteristics is obtained. The preset system operation database is used to save the corresponding relationship between the key energy consumption characteristics and the intelligent prediction model.

[0028] Preferably, the intelligent prediction module obtains energy consumption characteristics, specifically including:

[0029] The difference between the predicted energy consumption value and the actual energy consumption data is measured by the loss function corresponding to the intelligent prediction model;

[0030] An evaluation operation is performed on the predicted energy consumption value corresponding to the intelligent prediction model according to the difference degree obtained by calculating the loss function;

[0031] Based on the difference degree obtained by calculating the loss function, the prediction parameters corresponding to the intelligent prediction model are iteratively updated through the optimization algorithm.

[0032] Preferably, the intelligent prediction module obtains energy consumption characteristics, specifically including:

[0033] The evaluation operation is to obtain the evaluation index corresponding to the prediction parameter of the optimized intelligent prediction model;

[0034] The prediction parameters are the weights and bias parameters that are dynamically adjusted in the intelligent optimization model.

[0035] Preferably, the multi-objective optimization scheduling module obtains a multi-objective scheduling solution, specifically including:

[0036] B1: Encode the target scheduling index;

[0037] B2: Randomly generate an initial scheduling solution set;

[0038] B3: Define multi-objective comprehensive function;

[0039] B4: Perform differential mutation operations on each scheduling scheme in the initial scheduling scheme set based on the mutation probability;

[0040] The mutation probability is the difference of each scheduling scheme in the initial scheduling scheme set;

[0041] The differential mutation operation is used to modify the current scheduling plan, which is a scheduling plan randomly selected from the initial scheduling plan set to generate a random scheduling plan;

[0042] B5: Cross-operate the random scheduling scheme with the current scheduling scheme to generate a test scheduling scheme;

[0043] The crossover operation is used to select the scheduling method corresponding to the random scheduling scheme and the current scheduling scheme from the random scheduling scheme and the current scheduling scheme according to the uniform crossover rule;

[0044] B6: Compare the test scheduling scheme with the current scheduling scheme through a roulette wheel selection operation to generate a comprehensive scheduling scheme;

[0045] The roulette wheel selection operation is used to extract the scheduling method whose multi-objective comprehensive function value corresponding to the scheduling method in the test scheduling scheme is greater than the multi-objective comprehensive function value corresponding to the all-objective scheduling method in the current scheduling scheme;

[0046] B7: Update the initial scheduling solution set based on the comprehensive scheduling solution, and update the multi-objective comprehensive function value of each scheduling solution in the initial scheduling solution set;

[0047] B8: Repeat steps B4 to B7 until the termination condition is met. The multi-objective comprehensive function value of the termination condition comprehensive scheduling plan is maximized to generate a multi-objective optimization scheduling plan.

[0048] Preferably, the multi-objective optimization scheduling module obtains a multi-objective scheduling solution, specifically including:

[0049] Based on the actual energy allocation corresponding to the optimization target f j The expected energy allocation corresponding to the optimization target f j ‘ , calculate the relative deviation d between the actual energy allocation and the expected energy allocation corresponding to the jth optimization goal j , specifically expressed as:

[0050]

[0051] Among them, j represents the index of the optimization target;

[0052] The relative deviation d between the actual energy allocation and the expected energy allocation corresponding to the jth optimization goal j , the actual energy allocation corresponding to the jth optimization goal f j, and the expected energy allocation corresponding to the jth optimization goal f j ‘ , calculate the multi-objective comprehensive function value F corresponding to all optimization objectives, which is specifically expressed as:

[0053]

[0054] Among them, k represents the total number of optimization objectives, and j represents the index of the target scheduling indicator.

[0055] To achieve the above object, the present invention provides the following technical solution: an energy consumption optimization management method, implementing the above energy consumption optimization management system, comprising:

[0056] S1: Intelligent information collection terminal collects data: In response to the intelligent information collection terminal deployed in the energy consumption detection area, real-time energy consumption data is obtained;

[0057] S2: Preprocessing energy data: performing a preprocessing operation on the real-time energy consumption data, the preprocessing operation is used to obtain a key energy calibration group corresponding to the real-time energy consumption data;

[0058] S3: Intelligent prediction: Obtain an intelligent prediction model, and obtain energy consumption characteristics through a key energy calibration group based on the intelligent prediction model;

[0059] S4: Multi-objective optimization scheduling: Based on real-time energy consumption data and energy consumption characteristics, combined with the optimization objectives, a multi-objective optimization algorithm is used to obtain a multi-objective optimization scheduling solution;

[0060] S5: Real-time feedback: In response to the intelligent information collection terminal deployed in the energy consumption detection area, the energy consumption equipment is monitored according to the multi-objective optimization scheduling plan to obtain energy optimization feedback data;

[0061] S6: abnormal alarm: abnormal judgment operation is performed on the energy optimization feedback data, and the abnormal judgment operation is used to warn of abnormal situations corresponding to the energy optimization feedback data;

[0062] S7: Output scheduling plan: Obtain energy consumption optimization report based on energy consumption characteristics, multi-objective optimization scheduling plan, and energy optimization feedback data, and output it to the user through a visual interface.

[0063] Technical effects and advantages of the present invention:

[0064] 1. The present invention uses a variety of intelligent information collection terminals to acquire data through intelligent information collection terminal collection modules, thereby achieving efficient management of energy consumption and reducing operating costs;

[0065] 2. The present invention realizes dynamic prediction, optimized scheduling and efficient management of energy consumption through an intelligent prediction module, which increases the accuracy of energy consumption trend prediction and reduces the blindness of energy management decision-making caused by the inability to accurately predict;

[0066] 3. The present invention uses a multi-objective optimization algorithm through a multi-objective optimization scheduling module to obtain a multi-objective optimization scheduling solution, which increases the scientificity and rationality of energy scheduling and reduces energy waste caused by unreasonable scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 It is a system flow chart of the present invention.

[0068] Figure 2 This is a multi-scale LSTM model flow chart corresponding to the intelligent prediction model in the intelligent prediction module of the present invention.

[0069] Figure 3 This is the optimization scheduling flow chart corresponding to the multi-objective optimization scheduling module of the present invention. DETAILED DESCRIPTION

[0070] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0071] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification of the present application, the singular expressions "one", "a kind of", "said", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to and includes any or all possible combinations of one or more listed items.

[0072] In the following, the terms "first", "second", and "third" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", and "third" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0073] As attached Figure 1The energy consumption optimization management system shown includes a system operation database, a system central processing module and a user information terminal, and also includes an intelligent information collection terminal collection module, an energy data preprocessing module, an intelligent prediction module, a multi-objective optimization scheduling module, a real-time feedback module, an abnormal alarm module, and a scheduling plan output module.

[0074] The system operation database includes all data texts of the energy consumption optimization management system, and collects the information text output by each module in real time. The system central processing module is used for the information text instructions output by each module in the central control system, and the user information terminal is an information output device for receiving the energy consumption optimization management system.

[0075] The intelligent information collection terminal collection module responds to the intelligent information collection terminal deployed in the energy consumption detection area to obtain real-time energy consumption data.

[0076] Specifically, the energy consumption optimization management system responds to multiple intelligent information collection terminals pre-set in the energy consumption detection area, collects information from multiple energy consumption devices corresponding to the target energy consumption monitoring area, and the intelligent information collection terminals include but are not limited to current transformers, voltage sensors, water flow sensors, thermal gas mass flow meters, etc.; converts the data information obtained by the intelligent information collection terminals into digital signals; and transmits the digital signals by building a data transmission link with the help of LoRa wireless communication technology.

[0077] In this embodiment, for the incoming lines of the main distribution room in the industrial plant and the outgoing lines of each distribution cabinet, current transformers and voltage sensors with various ranges and accuracies are selected according to the rated current and voltage level of the line; for the line with a rated current of 1000A, a current transformer with a range of 0-1200A and an accuracy level of 0.5 is selected, and the wire is installed by passing through the center hole of the transformer in a through-hole manner; for the line with a voltage level of 10kV, a voltage sensor with a range of 0-12kV and an accuracy of 0.2 is selected.

[0078] In one possible implementation, the real-time energy consumption data includes the rated power of each energy-consuming device, the actual operating power of each energy-consuming device, the ratio of the production capacity to the energy consumption of each energy-consuming device, the temperature of the external environment in which each energy-consuming device is located, the humidity of the external environment in which each energy-consuming device is located, the wind speed of the external environment in which each energy-consuming device is located, the light intensity of the external environment in which each energy-consuming device is located, and the real-time market price of the energy used by each energy-consuming device.

[0079] The energy data preprocessing module is used to perform preprocessing operations on the real-time energy consumption data transmitted by the intelligent information collection terminal collection module, and the preprocessing operation is used to obtain the key energy calibration group corresponding to the real-time energy consumption data.

[0080] Specifically, after acquiring the real-time energy consumption data transmitted by the collection module of the intelligent information collection terminal, the energy consumption optimization management system preprocesses the real-time energy consumption data to obtain multiple key energy information corresponding to the real-time energy consumption data, namely, the key energy calibration group.

[0081] In a possible implementation, the preprocessing step includes:

[0082] A1: Data cleaning: Set multiple energy consumption parameter ranges for the received real-time energy consumption data. The energy consumption parameter ranges are pre-set normal range rules corresponding to various types of energy consumption data. Judge various types of real-time energy consumption data based on the energy consumption parameter ranges.

[0083] In this embodiment, for the voltage data corresponding to the power consumption, the normal operating voltage range corresponding to the 10kV line is set to 9kV-11kV. When the voltage value exceeds this range, it is determined as abnormal data and is removed;

[0084] A2: Data completion: Perform data completion operations based on missing real-time energy consumption data. The completion operation is used to complete the real-time energy consumption data that is not an outlier.

[0085] In this embodiment, when the real-time energy consumption data is missing in the power consumption data within the time period, and there are valid data points before and after the missing part, linear interpolation is used to complete it. The power at time t1 is 100kW, and the power at time t3 is 120kW. The missing data corresponding to time t2, the completion value is (100+(120-100) / (t3-t1)*(t2-t1))kW;

[0086] A3: Feature extraction and association analysis: Feature extraction and association analysis operations are performed on the cleaned and completed real-time energy consumption data. The feature extraction and association analysis operations are used to obtain energy consumption indicators, statistical indicators, and correlation coefficients corresponding to the real-time energy consumption data. Energy consumption indicators are key performance parameters corresponding to different types of energy consumption data. Statistical indicators are combinations of different types of energy data. Statistical indicators include the mean, variance, maximum value, and minimum value corresponding to each energy consumption indicator combination. Key energy information is extracted based on a preset correlation coefficient threshold. The correlation coefficient threshold is a pre-set benchmark for screening key energy information.

[0087] In this embodiment, multiple energy consumption indicators corresponding to each energy-consuming equipment in each working area are calculated in the industrial plant, and the multiple energy consumption indicators include but are not limited to active power, reactive power, power factor, etc., and the statistical indicators and correlation coefficients of each energy consumption indicator combination in unit time are calculated. When the absolute value of the correlation coefficient between each energy consumption indicator combination is greater than 0.7, the energy consumption indicator corresponding to the energy consumption indicator combination is extracted as key energy information;

[0088] A4: Data standardization processing: The extracted key energy information is processed using the Z-score standardization method to generate a key energy calibration group.

[0089] The intelligent prediction module is used to obtain an intelligent prediction model, and obtain energy consumption characteristics according to the key energy calibration group transmitted by the intelligent prediction model through the energy data preprocessing module.

[0090] Specifically, the intelligent prediction model is a pre-constructed learning model. By inputting the key energy calibration group transmitted by the energy data preprocessing module into the intelligent prediction model, the intelligent prediction model obtains energy consumption characteristics according to the key energy calibration group transmitted by the energy data preprocessing module.

[0091] In one possible implementation, obtaining energy consumption characteristics includes: performing a prediction operation on key energy information according to a key energy calibration group to obtain key energy consumption characteristics corresponding to the key energy information, the key energy consumption characteristics being a target consumption trend corresponding to the key energy information, the target consumption trends including rising energy consumption, stable energy consumption, and falling energy consumption; obtaining an intelligent prediction model corresponding to the energy consumption characteristics in a preset system operation database, the preset system operation database being used to store the correspondence between the energy consumption characteristics and the intelligent prediction model.

[0092] Specifically, in industrial plants, the intelligent prediction model adopts a deep learning multi-scale LSTM algorithm, performs time sorting based on key energy information, and obtains energy consumption trends through multiple LSTM models; obtains multiple time scales, where the time scale is multiple unit time periods corresponding to the prediction time; divides multiple LSTM models based on the time scale, and obtains multiple energy consumption features corresponding to the multiple LSTM models; uses a weighted average method to fuse multiple LSTM models of different time scales, uses multiple energy consumption features as fusion vectors according to weighted results, and obtains energy consumption features through activation functions. The energy consumption features include predicted energy consumption values, energy consumption trends, and evaluation indicators corresponding to the intelligent evaluation model.

[0093] In one possible implementation, obtaining energy consumption characteristics includes: measuring the degree of difference between the predicted energy consumption value and the actual energy consumption data through a loss function corresponding to the intelligent prediction model, the loss function includes mean square error and mean absolute error; based on the degree of difference calculated by the loss function, the prediction parameters corresponding to the intelligent prediction model are iteratively updated through an optimization algorithm, the optimization algorithm includes but is not limited to stochastic gradient descent, Adam, etc., and the prediction parameters are weights and bias parameters dynamically adjusted in the intelligent optimization model.

[0094] Specifically, the multi-scale LSTM model captures time series information at different levels by using LSTM models of multiple time scales at the same time. According to the importance of time scale to energy consumption characteristics, the prediction parameters corresponding to the LSTM models of multiple time scales are determined through the fully connected layer; based on the i-th predicted energy consumption value Ea i and the actual energy consumption value of the ith Calculate the root mean square error coefficient ER1 between the predicted energy consumption value and the actual energy consumption data, which is specifically expressed as:

[0095]

[0096] Where N represents the total number of test points in the test evaluation set for the intelligent prediction model, and i represents the index of the test point in the test evaluation set;

[0097] Based on the i-th predicted energy consumption value Ea i , the actual energy consumption value of the i-th And the observed average energy consumption Ea m , calculate the coefficient of determination ER2 between the predicted energy consumption value and the actual energy consumption data, which is specifically expressed as:

[0098]

[0099] Wherein, N represents the total number of test points in the test evaluation set for the intelligent prediction model, and i represents the index of the test point in the test evaluation set.

[0100] In a possible implementation, obtaining energy consumption characteristics includes: performing an evaluation operation on the predicted energy consumption value corresponding to the intelligent prediction model according to the degree of difference obtained by calculating multiple loss functions, and the evaluation operation is to obtain evaluation indicators corresponding to the prediction parameters of the optimized intelligent prediction model.

[0101] Specifically, when the root mean square error coefficient ER1 between the predicted energy consumption value and the actual energy consumption data is lower than the preset error threshold, the preset error threshold is set to 0.1, and the determination coefficient ER2 between the predicted energy consumption value and the actual energy consumption data belongs to the high-level standard, and the specific value corresponding to the high-level standard is 0.8, then the prediction result of the intelligent prediction model is evaluated to be highly accurate, and the prediction parameters corresponding to the intelligent prediction model do not need to be optimized through the optimization algorithm.

[0102] The multi-objective optimization scheduling module is based on the real-time energy consumption data transmitted by the intelligent information collection terminal acquisition module and the energy consumption characteristics transmitted by the intelligent prediction module. It combines the optimization objectives through the multi-objective optimization algorithm to obtain the multi-objective optimization scheduling plan.

[0103] Specifically, multiple optimization objectives are obtained, including but not limited to minimizing total energy consumption, reducing costs, reducing carbon emissions, etc.; the optimization objectives, real-time energy consumption data, and energy consumption characteristics are energy scheduled through a multi-objective optimization algorithm, and the multi-objective optimization algorithm includes a genetic algorithm, a particle swarm algorithm, and a simulated annealing algorithm to coordinate the energy consumption equipment corresponding to the energy consumption detection area and generate target optimization scheduling plans corresponding to multiple optimization objectives.

[0104] In a possible implementation, obtaining a multi-objective optimization scheduling solution includes:

[0105] B1: Encode the target scheduling index, which is the numerical value mapped to the energy-consuming equipment corresponding to multiple optimization targets. The target scheduling index includes the actual energy allocation of different types consumed by each energy-consuming equipment during operation, the actual start and stop time corresponding to each energy-consuming equipment, the expected energy allocation of different types consumed by each energy-consuming equipment during operation, and the actual start and stop time corresponding to each energy-consuming equipment;

[0106] B2: Randomly generate an initial scheduling plan set, which consists of multiple scheduling plans, and the scheduling plan contains D target scheduling indicators;

[0107] B3: define a multi-objective comprehensive function, which is used to evaluate the scheduling methods corresponding to each scheduling scheme in the initial scheduling scheme set;

[0108] In a possible implementation, defining a multi-objective comprehensive function includes: based on the actual energy allocation corresponding to the optimization target f j The expected energy allocation corresponding to the optimization target f j ‘ , calculate the relative deviation d between the actual energy allocation and the expected energy allocation corresponding to the jth optimization goal j , specifically expressed as:

[0109]

[0110] Among them, j represents the index of the optimization target;

[0111] The relative deviation d between the actual energy allocation and the expected energy allocation corresponding to the jth optimization goal j , the actual energy allocation corresponding to the jth optimization goal f j , and the expected energy allocation corresponding to the jth optimization goal f j ‘ , calculate the multi-objective comprehensive function value F corresponding to all optimization objectives, which is specifically expressed as:

[0112]

[0113] Among them, k represents the total number of optimization targets, and j represents the index of the target scheduling indicator;

[0114] B4: Perform differential mutation operation on each scheduling scheme in the initial scheduling scheme set based on mutation probability. The mutation probability is the difference of each scheduling scheme in the initial scheduling scheme set. The differential mutation operation is used to modify the current scheduling scheme. The current scheduling scheme is a scheduling scheme randomly selected from the initial scheduling scheme set to generate a random scheduling scheme.

[0115] B5: performing a cross operation on the random scheduling scheme and the current scheduling scheme. The cross operation is used to select a scheduling method corresponding to the random scheduling scheme and the current scheduling scheme according to a uniform cross rule from the random scheduling scheme and the current scheduling scheme to combine and generate a test scheduling scheme;

[0116] B6: Compare the test scheduling scheme with the current scheduling scheme through a roulette wheel selection operation, wherein the roulette wheel selection operation is used to extract a scheduling method whose multi-objective comprehensive function value corresponding to the scheduling method in the test scheduling scheme is greater than the multi-objective comprehensive function value corresponding to the full-objective partial scheduling method in the current scheduling scheme, so as to generate a comprehensive scheduling scheme;

[0117] B7: Update the initial scheduling solution set based on the comprehensive scheduling solution, and update the multi-objective comprehensive function value of each scheduling solution in the initial scheduling solution set;

[0118] B8: Repeat steps B4 to B7 until the termination condition is met. The multi-objective comprehensive function value of the termination condition comprehensive scheduling plan is maximized to generate a multi-objective optimization scheduling plan.

[0119] The real-time feedback module responds to the intelligent information collection terminal deployed in the energy consumption detection area, monitors the energy consumption equipment according to the multi-objective optimization scheduling scheme transmitted by the multi-objective optimization scheduling module, and obtains energy optimization feedback data.

[0120] Specifically, when it is detected that the multi-objective optimization scheduling scheme has begun to be implemented, the intelligent information collection terminal deployed in the energy consumption detection area will immediately respond to collect the energy consumption data corresponding to each energy-consuming device in the multi-objective optimization scheduling scheme in real time, and compare the energy consumption data with the expected data corresponding to the multi-objective optimization scheduling scheme, mark the deviation data, and organize it by timestamp and device identification to obtain energy optimization feedback data.

[0121] It should be noted that the energy optimization feedback data includes the deviation between the actual energy consumption and the expected consumption of each energy-consuming equipment presented in a tabular form, the preliminary judgment results on the causes of the deviation, and the energy optimization direction suggestions based on the preliminary judgment results on the causes of the deviation and the actual situation; the energy optimization direction suggestions based on the preliminary judgment results on the causes of the deviation and the actual situation are obtained by comparing the historical energy optimization direction suggestions stored in the system operation database.

[0122] The abnormal alarm module is used to perform abnormal judgment operations on the energy optimization feedback data transmitted by the real-time feedback module, and the abnormal judgment operation is used to warn of abnormal situations corresponding to the energy optimization feedback data.

[0123] Specifically, based on the historical data in the system operation database, an energy consumption deviation percentage threshold is set for the energy consumption deviation of each energy-consuming device corresponding to the energy optimization feedback data, and the energy consumption deviation percentage threshold is transmitted to the real-time feedback module and the multi-objective optimization scheduling module; at the same time, the time of occurrence of the abnormality, the name and number of the energy-consuming equipment, the energy consumption area and specific location, and the type of abnormality corresponding to the energy optimization feedback data are recorded, and early warning information is generated and released to the person in charge of the energy management department and the equipment maintenance personnel.

[0124] The scheduling plan output module obtains the energy consumption optimization report based on the energy consumption characteristics transmitted by the intelligent prediction module, the multi-objective optimization scheduling plan transmitted by the multi-objective optimization scheduling module, and the energy optimization feedback data transmitted by the real-time feedback module, and outputs it to the user through a visual interface.

[0125] As attached Figure 2 The energy consumption optimization management method shown includes: S1: intelligent information collection terminal collects data, S2: pre-processes energy data, S3: intelligent prediction, S4: multi-objective optimization scheduling, S5: real-time feedback, S6: abnormal alarm, and S7: output scheduling plan.

[0126] S1: Intelligent information collection terminal collects data: In response to the intelligent information collection terminal deployed in the energy consumption detection area, real-time energy consumption data is obtained;

[0127] S2: Preprocessing energy data: performing a preprocessing operation on the real-time energy consumption data, the preprocessing operation is used to obtain a key energy calibration group corresponding to the real-time energy consumption data;

[0128] S3: Intelligent prediction: Obtain an intelligent prediction model, and obtain energy consumption characteristics through a key energy calibration group based on the intelligent prediction model;

[0129] S4: Multi-objective optimization scheduling: Based on real-time energy consumption data and energy consumption characteristics, combined with the optimization objectives, a multi-objective optimization algorithm is used to obtain a multi-objective optimization scheduling solution;

[0130] S5: Real-time feedback: In response to the intelligent information collection terminal deployed in the energy consumption detection area, the energy consumption equipment is monitored according to the multi-objective optimization scheduling plan to obtain energy optimization feedback data;

[0131] S6: abnormal alarm: abnormal judgment operation is performed on the energy optimization feedback data, and the abnormal judgment operation is used to warn of abnormal situations corresponding to the energy optimization feedback data;

[0132] S7: Output scheduling plan: Obtain energy consumption optimization report based on energy consumption characteristics, multi-objective optimization scheduling plan, and energy optimization feedback data, and output it to the user through a visual interface.

[0133] Secondly: In the drawings of the embodiments disclosed in the present invention, only the structures related to the embodiments disclosed in the present invention are involved, and other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of the present invention can be combined with each other;

[0134] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. Energy consumption optimization management system, including system operation database, system central processing module and user information terminal, characterized in that: Also includes: Intelligent information collection terminal collection module: responds to the intelligent information collection terminal deployed in the energy consumption detection area to obtain real-time energy consumption data; Energy data preprocessing module: used to preprocess the real-time energy consumption data transmitted by the intelligent information collection terminal collection module, and the preprocessing operation is used to obtain the key energy calibration group corresponding to the real-time energy consumption data; Intelligent prediction module: used to obtain an intelligent prediction model, and obtain energy consumption characteristics according to the key energy calibration group transmitted by the energy data preprocessing module through the intelligent prediction model; Multi-objective optimization scheduling module: Based on the real-time energy consumption data transmitted by the intelligent information collection terminal acquisition module and the energy consumption characteristics transmitted by the intelligent prediction module, a multi-objective optimization scheduling scheme is obtained through a multi-objective optimization algorithm; Real-time feedback module: In response to the intelligent information collection terminal, it monitors the energy-consuming equipment according to the multi-objective optimization scheduling scheme transmitted by the multi-objective optimization scheduling module to obtain energy optimization feedback data; Abnormal alarm module: used to perform abnormal judgment operation on the energy optimization feedback data transmitted by the real-time feedback module. The abnormal judgment operation is used to warn of abnormal situations corresponding to the energy optimization feedback data; Scheduling scheme output module: based on the energy consumption characteristics transmitted by the intelligent prediction module, the multi-objective optimization scheduling scheme transmitted by the multi-objective optimization scheduling module, and the energy optimization feedback data transmitted by the real-time feedback module, the energy consumption optimization report is obtained and output to the user through a visual interface; The system operation database includes all data texts of the energy consumption optimization management system, and collects the information text output by each module in real time. The system central processing module is used for the information text instructions output by each module in the central control system, and the user information terminal is an information output device for receiving the energy consumption optimization management system.

2. The energy consumption optimization management system according to claim 1, characterized in that: The energy data preprocessing module obtains the key energy calibration group corresponding to the real-time energy consumption data, specifically including: Perform feature extraction and correlation analysis on real-time energy consumption data, and extract key energy information based on correlation coefficient thresholds; The feature extraction and association analysis operations are used to obtain energy consumption indicators, statistical indicators, and correlation coefficients corresponding to real-time energy consumption data.

3. The energy consumption optimization management system according to claim 2, characterized in that: The energy data preprocessing module performs feature extraction and association analysis operations, specifically including: Energy consumption indicators are key performance parameters corresponding to different types of energy consumption data; Statistical indicators are a combination of different types of energy data; The statistical indicators include the corresponding mean, variance, maximum value, and minimum value between each energy consumption indicator combination; The correlation coefficient threshold is a pre-set benchmark for screening key energy information.

4. The energy consumption optimization management system according to claim 1, characterized in that: The intelligent prediction module obtains energy consumption characteristics, specifically including: According to the key energy calibration group, a prediction operation is performed on the key energy information to obtain key energy consumption characteristics corresponding to the key energy information; The key energy consumption characteristics are the target consumption trends corresponding to the key energy information; In the preset system operation database, an intelligent prediction model corresponding to the key energy consumption characteristics is obtained. The preset system operation database is used to save the corresponding relationship between the key energy consumption characteristics and the intelligent prediction model.

5. The energy consumption optimization management system according to claim 4, characterized in that: The intelligent prediction module obtains energy consumption characteristics, specifically including: The difference between the predicted energy consumption value and the actual energy consumption data is measured by the loss function corresponding to the intelligent prediction model; An evaluation operation is performed on the predicted energy consumption value corresponding to the intelligent prediction model according to the difference degree obtained by calculating the loss function; Based on the difference degree obtained by calculating the loss function, the prediction parameters corresponding to the intelligent prediction model are iteratively updated through the optimization algorithm.

6. The energy consumption optimization management system according to claim 4, characterized in that: The intelligent prediction module obtains energy consumption characteristics, specifically including: The evaluation operation is to obtain the evaluation index corresponding to the prediction parameter of the optimized intelligent prediction model; The prediction parameters are the weights and bias parameters that are dynamically adjusted in the intelligent optimization model.

7. The energy consumption optimization management system according to claim 1, characterized in that: The multi-objective optimization scheduling module obtains a multi-objective scheduling solution, specifically including: B1: Encode the target scheduling index; B2: Randomly generate an initial scheduling solution set; B3: Define multi-objective comprehensive function; B4: Perform differential mutation operations on each scheduling scheme in the initial scheduling scheme set based on the mutation probability; The mutation probability is the difference of each scheduling scheme in the initial scheduling scheme set; The differential mutation operation is used to modify the current scheduling plan, which is a scheduling plan randomly selected from the initial scheduling plan set to generate a random scheduling plan; B5: Cross-operate the random scheduling scheme with the current scheduling scheme to generate a test scheduling scheme; The crossover operation is used to select the scheduling method corresponding to the random scheduling scheme and the current scheduling scheme according to the uniform crossover rule from the random scheduling scheme and the current scheduling scheme; B6: Compare the test scheduling scheme with the current scheduling scheme through a roulette wheel selection operation to generate a comprehensive scheduling scheme; The roulette selection operation is used to extract the scheduling method whose multi-objective comprehensive function value corresponding to the scheduling method in the test scheduling scheme is greater than the multi-objective comprehensive function value corresponding to the all-objective scheduling method in the current scheduling scheme; B7: Update the initial scheduling solution set based on the comprehensive scheduling solution, and update the multi-objective comprehensive function value of each scheduling solution in the initial scheduling solution set; B8: Repeat steps B4 to B7 until the termination condition is met. The multi-objective comprehensive function value of the termination condition comprehensive scheduling plan is maximized to generate a multi-objective optimization scheduling plan.

8. The energy consumption optimization management system according to claim 7, characterized in that: The multi-objective optimization scheduling module obtains a multi-objective scheduling solution, specifically including: Based on the actual energy allocation corresponding to the optimization target f j The expected energy allocation corresponding to the optimization target f j ‘ , calculate the relative deviation d between the actual energy allocation and the expected energy allocation corresponding to the jth optimization goal j , specifically expressed as: Among them, j represents the index of the optimization target; The relative deviation d between the actual energy allocation and the expected energy allocation corresponding to the jth optimization goal j , the actual energy allocation corresponding to the jth optimization goal f j , and the expected energy allocation corresponding to the jth optimization goal f j ‘ , calculate the multi-objective comprehensive function value F corresponding to all optimization objectives, which is specifically expressed as: Among them, k represents the total number of optimization objectives, and j represents the index of the target scheduling indicator.

9. A method for optimizing energy consumption management, according to any one of claims 1 to 8, for use in an energy consumption optimization management system, characterized in that: include: S1: Intelligent information collection terminal collects data: In response to the intelligent information collection terminal deployed in the energy consumption detection area, real-time energy consumption data is obtained; S2: Preprocessing energy data: performing a preprocessing operation on the real-time energy consumption data, the preprocessing operation is used to obtain a key energy calibration group corresponding to the real-time energy consumption data; S3: Intelligent prediction: Obtain an intelligent prediction model, and obtain energy consumption characteristics through a key energy calibration group based on the intelligent prediction model; S4: Multi-objective optimization scheduling: Based on real-time energy consumption data and energy consumption characteristics, combined with the optimization objectives, a multi-objective optimization algorithm is used to obtain a multi-objective optimization scheduling solution; S5: Real-time feedback: In response to the intelligent information collection terminal deployed in the energy consumption detection area, the energy consumption equipment is monitored according to the multi-objective optimization scheduling plan to obtain energy optimization feedback data; S6: abnormal alarm: abnormal judgment operation is performed on the energy optimization feedback data, and the abnormal judgment operation is used to warn of abnormal situations corresponding to the energy optimization feedback data; S7: Output scheduling plan: Obtain energy consumption optimization report based on energy consumption characteristics, multi-objective optimization scheduling plan, and energy optimization feedback data, and output it to the user through a visual interface.

Images