Intelligent energy consumption distribution method for multi-source energy system

Abstract

The application discloses an intelligent energy consumption distribution method of a multi-source energy system, which comprises energy supply data fusion, double-channel joint prediction, energy response lag correction, multi-objective optimization and intelligent energy consumption distribution. The application relates to the technical field of data processing of power management and resource scheduling. Through unified collection and archiving of operation control data and environmental information of power grid energy, distributed photovoltaic energy and energy storage systems, an energy supply fusion data set is constructed. A joint supply and demand prediction method of double-domain attention decoupling fusion is adopted to extract energy availability and user load characteristics respectively, so that double-channel joint modeling is realized. In combination with a dynamic response curve reconstruction method, the response lag of different energies is modeled, estimated and corrected. A layered target penalty function containing real-time response error, economic cost, carbon emission and response penalty term is further constructed and embedded into a particle swarm optimization algorithm to obtain an energy distribution strategy under multi-objective optimization.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of energy system regulation and intelligent optimization, and particularly relates to an intelligent energy consumption distribution method of a multi-source energy system. BACKGROUND

[0002] The intelligent energy consumption distribution method of the multi-source energy system is a dynamic energy scheduling technology based on an optimization algorithm, aiming to realize the optimal balance of economy, reliability and sustainability by real-time coordinating the energy supply proportions of renewable energy, energy storage systems and traditional power grids and the like while meeting user demand. The method dynamically optimizes the distribution amounts of various energies by taking load demand as a constraint, combining energy cost, response delay penalty (such as imposing a quadratic penalty on energy for overtime response) and device physical limitations (such as energy storage SOC and power grid power upper limit), thereby reducing energy consumption cost, improving response speed, enhancing system stability, and promoting efficient use of clean energy, and is suitable for microgrids, smart buildings and industrial energy management scenarios, and is a core means for promoting the intelligentization and low carbonization of energy systems.

[0003] However, in the existing intelligent energy consumption distribution method of the multi-source energy system, the current intelligent energy consumption distribution method of the multi-source energy system generally adopts a static priority, fixed rule or simple linear rule guided regulation mode, lacks systematic modeling of the heterogeneous characteristics (such as response speed, availability fluctuation and scheduling delay) among multiple sources of energy, resulting in the technical problems of frequent energy switching, unbalanced supply and demand matching and out-of-control response lag in the system under dynamic load changes and complex weather disturbances; in the existing dual-channel joint prediction method, the traditional dual-channel prediction method usually directly concatenates energy availability and user load as two input features and inputs them into a unified network model, without effectively identifying the differences in information structure and the independence of driving logic between the two, resulting in the technical problems of serious feature coupling interference, weak key factor identification ability and high inaccuracy rate of the prediction model in extreme scenarios; in the existing energy response lag correction method, the existing energy scheduling system usually only uses a static offset method or an empirical lag correction means for response delay compensation, without fully modeling the dynamic response curves of devices under different load conditions, resulting in the technical problems of large lag estimation deviation, invalidation of strategy execution in advance or delay. SUMMARY

[0004] In view of the above, in order to overcome the defects of the prior art, the present application provides an intelligent energy consumption distribution method of a multi-source energy system, which comprises the following steps:

[0005] Step S1: energy supply data fusion;

[0006] Step S2: dual-channel joint prediction;

[0007] Step S3: Energy response lag correction;

[0008] Step S4: Multi-objective optimization;

[0009] Step S5: Intelligent energy consumption allocation.

[0010] Further, in step S1, the energy supply data fusion is used to collect and integrate multi-energy and environmental information data, specifically to uniformly collect, classify and structure process the grid energy, distributed photovoltaic energy, energy storage system energy and related environmental variables, and for each energy, to collect structured operation data and control records, obtaining an energy supply fusion data set;

[0011] The energy supply fusion data set specifically includes operation control data and external environment data.

[0012] Further, in step S2, the dual-channel joint prediction is used for dual-channel modeling of energy availability and user load, specifically, according to the energy supply fusion data set, a dual-domain attention decoupling fusion joint supply and demand prediction method is used for dual-channel joint prediction, obtaining joint energy supply and demand change prediction data, including the following steps:

[0013] Step S21: Dual-channel data construction, specifically, according to the energy supply fusion data set, external energy availability data is extracted as first channel raw sequence data input, and user load record data is extracted as second channel raw sequence data input, by performing normalization processing on the first channel raw sequence data input and the second channel raw sequence data input respectively, and introducing a time embedding vector based on the sine and cosine function, dual-channel input data is constructed;

[0014] Step S22: Dual-channel feature extraction network construction, specifically, according to the dual-channel input data, a dual-channel feature extraction network is constructed, and by using the dual-channel feature extraction network, dual-channel feature extraction is performed, obtaining energy availability feature data and user historical load feature data;

[0015] The dual-channel feature extraction network includes an energy availability channel and a user load channel;

[0016] Step S23: Decoupling attention mechanism modeling, specifically, self-attention modules are constructed for the energy availability feature data and the user historical load feature data respectively, obtaining decoupling attention feature data; the decoupling attention feature data specifically includes energy availability decoupling attention features and user historical load decoupling attention features;

[0017] Step S24: dual-domain fusion residual correction, specifically, the energy availability decoupling attention feature and the user historical load decoupling attention feature are spliced and input into a multi-layer perception module to obtain a fused dual-domain feature, and a cross-channel residual correction function is introduced to correct the fused dual-domain feature to obtain a corrected fusion feature data;

[0018] Step S25: dual-channel joint prediction, specifically, decoding prediction is performed according to the corrected fusion feature data to obtain joint supply and demand prediction data, and data subdivision is performed on the joint supply and demand prediction data to obtain joint energy supply and demand change prediction data;

[0019] The joint energy supply and demand change prediction data specifically includes a power grid energy availability prediction value, a distributed photovoltaic energy availability prediction value, an energy storage system energy availability prediction value, a user load demand prediction value, and a supply and demand change prediction difference value.

[0020] Further, in step S3, the energy response lag correction is used to model and correct the response time and efficiency of different energies, specifically, a multi-source energy lag correction method combined with a dynamic response curve reconstruction is used to correct the energy response lag according to the energy supply fusion data set and the joint energy supply and demand change prediction data, to obtain energy correction prediction data, including the following steps:

[0021] Step S31: response curve modeling, specifically, the joint energy supply and demand change prediction data is used to collect the historical control instruction sequence of each energy and the actual output power at the corresponding time, and a response curve model fitting is performed to obtain energy response curve data; the response curve model fitting specifically refers to fitting the power grid energy and the energy storage system energy by an improved first-order delay linear function; fitting the distributed photovoltaic energy by a saturated improved nonlinear exponential function;

[0022] Step S32: response time difference estimation, specifically, a lag time window search model is constructed, and the maximum correlation coefficient between the control signal and the response value of each energy is calculated by a sliding lag step, and the maximum correlation lag value parameter is calculated for the power grid energy and the energy storage system energy, and the maximum correlation lag value parameter is estimated for the distributed photovoltaic energy using a lag correlation peak positioning algorithm to obtain an optimal response time delay estimation parameter;

[0023] Step S33: lag alignment correction, specifically, time alignment processing is performed according to the energy response curve data and the optimal response time delay estimation parameter to obtain aligned energy prediction data; the time alignment processing is performed by reverse solving a linear response model for the power grid energy and the energy storage system energy; and the time alignment processing is performed by numerical interpolation approximation for the distributed photovoltaic energy;

[0024] Step S34: dynamic correction fusion, specifically, constructing an aligned comprehensive energy supply prediction vector containing grid energy, distributed photovoltaic energy and energy storage system energy according to the alignment energy supply prediction data, obtaining multi-source energy supply prediction matrix data;

[0025] Step S35: energy response lag correction, specifically, through the response curve modeling, the response time difference estimation, the lag alignment correction and the dynamic correction fusion, energy response lag correction is performed, the multi-source energy supply prediction matrix data is output as energy response lag corrected energy supply prediction data, and energy correction prediction data is obtained.

[0026] Further, in step S4, the multi-objective optimization is used to generate a multi-objective optimal energy distribution strategy, specifically, according to the energy correction prediction data, an energy distribution particle swarm optimization algorithm combined with a hierarchical target penalty function is used, an improved hierarchical target penalty function is constructed and embedded into the fitness evaluation link of particle swarm optimization, and optimization variables are defined, multi-objective optimization is performed, and energy distribution strategy reference data is obtained;

[0027] The improved hierarchical target penalty function is specifically constructed by adaptively weighting the real-time response error, economic cost and total carbon emission as optimization targets, and introducing an energy response lag penalty term;

[0028] The optimization variables specifically include total load demand distribution, energy supply upper limit and device physical constraints;

[0029] The embedding into the fitness evaluation link of particle swarm optimization is specifically for a group of energy distribution schemes represented by each particle, calculating the evaluation indexes of each energy distribution scheme in real-time response error, economic cost and total carbon emission, combining the response lag degree of each energy in the energy correction prediction data, using the improved hierarchical target penalty function to accurately score the fitness of the particle, obtaining the fitness parameter; by using the standard particle swarm optimization algorithm and according to the fitness parameter, guiding the particle to dynamically adjust the position and speed in the search space, gradually approaching the optimal energy consumption distribution strategy under multi-objective trade-off, obtaining energy distribution strategy reference data.

[0030] Further, in step S5, the intelligent energy consumption distribution is used to realize intelligent energy regulation and coordination, specifically, according to the energy distribution strategy reference data, the optimal energy distribution of various types of energy is analyzed, and the control data frame is formatted and generated by combining the running state parameters of various types of energy, obtaining intelligent energy consumption distribution control instructions;

[0031] The intelligent energy consumption allocation control instruction specifically comprises a timestamp, an energy type, a control power value, a control allowable range value, an instruction type, a delay hysteresis compensation parameter and a control priority level.

[0032] The above scheme has the following beneficial effects:

[0033] (1) In the existing intelligent energy consumption allocation method of the multi-source energy system, the intelligent energy consumption allocation method in the current multi-source energy system generally adopts a static priority, a fixed rule or a simple linear rule guided regulation mode, lacks systematic modeling of the heterogeneous characteristics (such as response speed, availability fluctuation and scheduling delay) among multi-source energy, and leads to the technical problems that under the conditions of dynamic load change and complex weather disturbance, the system is prone to frequent energy switching, unbalanced supply and demand matching, and out-of-control response lag. The present scheme creatively adopts a comprehensive intelligent method of double-channel joint prediction, hysteresis correction and multi-objective optimization, realizes dynamic adjustment of energy response matching strategy on the basis of accurate prediction of future energy supply and demand trend, and balances real-time performance, economy and carbon emission, thereby realizing a multi-source collaborative, dynamic matching and intelligent controllable energy consumption allocation mechanism;

[0034] (2) In the existing double-channel joint prediction method, the traditional double-channel prediction method usually directly splices the energy availability and user load as two input features and inputs them into a unified network model, fails to effectively identify the difference in information structure and independence of driving logic between the two, and leads to the technical problems that the prediction model has serious feature coupling interference, weak key factor identification ability and high inaccuracy rate in extreme scenarios. The present scheme creatively adopts a joint supply and demand prediction method of double-domain attention decoupling fusion, extracts key driving factors of energy and load by constructing independent time sequence modeling channels and self-attention modules, and introduces a residual compensation mechanism for feature fusion, thereby realizing multi-channel collaborative prediction effects of decoupled prediction features, enhanced fusion and improved supply and demand fitting accuracy;

[0035] (3) In the existing energy response hysteresis correction method, the existing energy scheduling system usually only adopts a static offset method or an empirical hysteresis correction means for response delay compensation, fails to fully model the dynamic response curve of the equipment under different load conditions, leads to large hysteresis estimation deviation, invalidation of strategy execution in advance or delay, and especially the technical problems are particularly significant in photovoltaic fluctuation and critical SOC interval of energy storage. The present scheme creatively adopts a multi-source energy hysteresis correction method combining dynamic response curve reconstruction to correct the energy response hysteresis, establishes multiple types of response behavior models for energy sources such as power grid, photovoltaic and energy storage, and dynamically corrects the prediction results by combining historical energy supply behavior and control signal backstepping hysteresis step, thereby realizing the goals of accurate response alignment, time effectiveness enhancement and hysteresis adaptive scheduling ability improvement;

[0036] (4) In the existing multi-objective optimization and control method, the current multi-objective optimization method often relies on the traditional weighting method or the Pareto solution set method for energy allocation, and fails to combine the response behavior and regulation delay characteristics of different energy types to construct an optimization target structure with real-time adaptability, resulting in that the optimization result is theoretically reasonable but poor in practical executability, and especially lacks the technical problem of strategy punishment constraint for response lag equipment. The scheme creatively adopts an energy distribution particle swarm optimization algorithm combined with a hierarchical target punishment function, realizes multi-objective optimization by constructing an improved hierarchical target punishment function and embedding it into the fitness evaluation link of particle swarm optimization, and defining optimization variables, and realizes a multi-source energy distribution optimization mechanism with multi-objective index fusion, response delay dynamic punishment and enhanced strategy executability. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 A flowchart of an intelligent agricultural machinery performance improvement method provided by the present application is shown in the figure.

[0038] Figure 2 A flowchart of the double-channel joint prediction of step S2 is shown in the figure.

[0039] Figure 3 A double-channel structure diagram of the double-channel joint prediction of step S2 is shown in the figure.

[0040] Figure 4 A flowchart of the energy response lag correction of step S3 is shown in the figure.

[0041] The accompanying drawings are used to provide a further understanding of the present application, and constitute a part of the specification, and are used to explain the present application together with embodiments of the present application, and do not constitute a limitation on the present application. DETAILED DESCRIPTION

[0042] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0043] In the description of the present application, it should be understood that the terms "upper", "lower", "front", "back", "left", "right", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application.

[0044] Embodiment one, refer to Figure 1 The application provides an intelligent agricultural machinery performance improvement method, which comprises the following steps:

[0045] Step S1: energy supply data fusion;

[0046] Step S2: double-channel joint prediction;

[0047] Step S3: energy response lag correction;

[0048] Step S4: multi-objective optimization;

[0049] Step S5: intelligent energy consumption distribution.

[0050] By performing the above operation, in the intelligent energy consumption distribution method of the existing multi-source energy system, the intelligent energy consumption distribution method in the current multi-source energy system generally adopts a static priority, a fixed rule or a simple linear rule guided regulation mode, lacks systematic modeling of the heterogeneous characteristics (such as response speed, availability fluctuation and scheduling delay) among multi-source energy, and leads to the technical problems that under the conditions of dynamic load change and complex weather disturbance, the system is prone to frequent energy switching, unbalanced supply and demand matching, and out-of-control response lag. The present scheme creatively adopts a comprehensive intelligent method of double-channel joint prediction, lag correction and multi-objective optimization, realizes dynamic adjustment of energy response matching strategy on the basis of accurate prediction of future energy supply and demand trend, and balances real-time performance, economy and carbon emission, and realizes a multi-source collaborative, dynamic matching and intelligent controllable energy consumption distribution mechanism.

[0051] Embodiment two, refer to Figure 1 and Figure 2 In step S1, the energy supply data fusion is used for collecting and integrating multi-energy and environmental information data, specifically, the power grid energy, distributed photovoltaic energy, energy storage system energy and related environmental variables are uniformly collected, classified and structured, and for each kind of energy, structured operation data and control record collection are performed to obtain an energy supply fusion data set;

[0052] The energy supply fusion data set specifically comprises operation control data and external environment data;

[0053] The operation control data specifically comprises current and historical energy supply power data, energy supply upper and lower limit data, start and stop control instruction data, electricity price data, energy storage state of charge data and control mode state data;

[0054] The external environment data specifically comprises illumination intensity, sunshine duration, wind speed, temperature and humidity, weather type, cloud cover index and air quality index.

[0055] Embodiment three, refer to Figure 1 ,Figure 2 and Figure 3 The embodiment is based on the above-mentioned embodiment, in step S2, the dual-channel joint prediction is used for dual-channel modeling of energy availability and user load, specifically, a dual-domain attention decoupling fusion joint supply and demand prediction method is used to perform dual-channel joint prediction according to the energy supply fusion data set, to obtain joint energy supply and demand change prediction data, including the following steps:

[0056] Step S21: dual-channel data construction, specifically, according to the energy supply fusion data set, external energy availability data is extracted as first channel original sequence data input, and user load record data is extracted as second channel original sequence data input, normalization processing is performed on the first channel original sequence data input and the second channel original sequence data input respectively, and a time embedding vector based on a sine and cosine function is introduced, to construct dual-channel input data;

[0057] Step S22: dual-channel feature extraction network construction, specifically, according to the dual-channel input data, a dual-channel feature extraction network is constructed, and dual-channel feature extraction is performed by using the dual-channel feature extraction network, to obtain energy availability feature data and user historical load feature data;

[0058] The dual-channel feature extraction network includes an energy availability channel and a user load channel;

[0059] The energy availability channel performs local feature enhancement through a one-dimensional convolutional network, and inputs the local features into a bidirectional long short-term memory network to obtain energy availability feature data, and the calculation formula is:

[0060] ;

[0061] In the formula, H s is the energy availability feature data, BiLSTM is the bidirectional long short-term memory network, Conv1D is the one-dimensional convolutional network, and X s is the first channel original sequence data input in the dual-channel input data;

[0062] The user load channel performs periodic fluctuation feature recognition enhancement through a gated recurrent unit network and introduces a periodic attention module, to obtain user historical load feature data, and the calculation formula is:

[0063] ;

[0064] In the formula, H d is the user historical load feature data, PACGRU is the gated recurrent unit network with the periodic attention module, and X d is the second channel original sequence data input in the dual-channel input data.

[0065] Step S23: decoupling attention mechanism modeling, specifically constructing a self-attention module for the energy availability feature data and the user historical load feature data respectively to obtain decoupled attention feature data; the decoupled attention feature data specifically includes energy availability decoupled attention features and user historical load decoupled attention features;

[0066] Step S24: dual-domain fusion residual correction, specifically inputting the energy availability decoupled attention features and the user historical load decoupled attention features after feature splicing to a multi-layer perception module to obtain fusion dual-domain features, and performing dual-domain fusion residual correction on the fusion dual-domain features by introducing a cross-channel residual correction function to obtain corrected fusion feature data;

[0067] The calculation formula of the fusion dual-domain features is:

[0068] ;

[0069] In the formula, is the fusion dual-domain features, MLP is a multi-layer perception module, Z s is the energy availability decoupled attention features, Z d is the user historical load decoupled attention features, and || is a feature splicing operator;

[0070] The calculation formula of the cross-channel residual correction function is:

[0071] ;

[0072] In the formula, Z fused is the corrected fusion feature data, is the fusion dual-domain features, is a residual adjustment factor, which is specifically obtained by automatic learning through training;

[0073] Step S25: dual-channel joint prediction, specifically decoding prediction according to the corrected fusion feature data to obtain joint supply and demand prediction data, and performing data subdivision on the joint supply and demand prediction data to obtain joint energy supply and demand change prediction data;

[0074] The joint energy supply and demand change prediction data specifically includes a power grid energy availability prediction value, a distributed photovoltaic energy availability prediction value, an energy storage system energy availability prediction value, a user load demand prediction value, and a supply and demand change prediction difference value.

[0075] By performing the above operation, for in the existing double-channel joint prediction method, there is a traditional double-channel prediction method usually directly splicing energy availability and user load as two input features into a unified network model, which fails to effectively identify the difference in information structure and the independence of driving logic between the two, resulting in the technical problems of serious feature coupling interference, weak key factor identification ability, and high inaccuracy rate in extreme scenarios. The present scheme creatively adopts a dual-domain attention decoupling fusion joint supply and demand prediction method, extracts key driving factors of energy and load through the construction of independent time series modeling channels and self-attention modules, and introduces a residual compensation mechanism for feature fusion, achieving the multi-channel collaborative prediction effect of decoupled, fused and enhanced prediction features and improved supply and demand fitting accuracy.

[0076] Embodiment four, refer to Figure 1 and Figure 4 This embodiment is based on the above-mentioned embodiments. In step S3, the energy response lag correction is used to model and correct the response time and efficiency of different energies. Specifically, according to the energy supply fusion data set and the joint energy supply and demand change prediction data, a multi-source energy lag correction method combined with a dynamic response curve reconstruction is used to correct the energy response lag, and energy correction prediction data is obtained, including the following steps:

[0077] Step S31: response curve modeling, specifically, according to the joint energy supply and demand change prediction data, the historical control instruction sequence of each energy and the actual output power at the corresponding time are collected to perform response curve model fitting to obtain energy response curve data; for the grid energy and the energy storage system energy, an improved first-order delay linear function is used for fitting; for the distributed photovoltaic energy, a saturated improved nonlinear exponential function is used for fitting;

[0078] The improved first-order delay linear function specifically introduces a dynamic response gain adjustment parameter and a control signal fluctuation suppression factor to improve the anti-disturbance ability, and the calculation formula is:

[0079] ;

[0080] In the formula, P(t) is the output power of the improved first-order delay linear function, K d is a dynamic response gain parameter, which is calculated automatically, t is the current time index, t0 is the control signal sending time index, is a lag time parameter, which is used to represent the time offset of the energy, is a system response time constant, is a fluctuation suppression factor, the default value is set to 0.1, and the maximum value is not higher than 0.3, and u(t) is the original input signal of the control signal, is the actual input signal of the system after the control signal at the current time t takes effect after the backtracking lag time;

[0081] The saturation-improved nonlinear exponential function is specifically improved in nonlinear change modeling capability by constructing a model with saturation gating and power upper limit fitting mechanisms, and the calculation formula is:

[0082] ;

[0083] In the formula, P PV (t) is a saturation-improved nonlinear exponential function, P max is the rated maximum power, The whole is a power upper limit fitting term, is a response steepness coefficient, I sun (t) is an illumination intensity parameter, is a lag time parameter, I th is a response threshold, is a saturation gating term;

[0084] Preferably, the calculation formula of the saturation gating term is:

[0085] ;

[0086] In the formula, t sunrise is the sunrise time, t sunset is the sunset time;

[0087] Step S32: response time difference estimation, specifically constructing a lag time window search model and calculating the maximum correlation coefficient between the control signal and the response value of each energy source through a sliding lag step, and calculating the maximum correlation lag value parameter for the grid energy and the energy storage system energy, and using a lag correlation peak positioning algorithm to estimate the maximum correlation lag value parameter for the distributed photovoltaic energy to obtain the optimal response time delay estimation parameter;

[0088] Step S33: lag alignment correction, specifically performing time alignment processing according to the energy response curve data and the optimal response time delay estimation parameter to obtain aligned energy supply prediction data; for the grid energy and the energy storage system energy, the time alignment processing is calculated by back-solving a linear response model; for the distributed photovoltaic energy, the time alignment processing is calculated by numerical interpolation approximation solution;

[0089] The linear response model specifically refers to a data model corresponding to the improved first-order delay linear function;

[0090] The numerical interpolation approximation solution specifically refers to dynamic time interpolation based on the saturation-improved nonlinear exponential function;

[0091] Step S34: dynamic correction fusion, specifically, constructing an aligned comprehensive energy supply prediction vector containing grid energy, distributed photovoltaic energy and energy storage system energy according to the aligned energy supply prediction data, to obtain a multi-source energy supply prediction matrix data;

[0092] The calculation formula of the multi-source energy supply prediction matrix is:

[0093] ;

[0094] In the formula, is a multi-source energy supply prediction matrix, is the aligned energy supply prediction data of the grid energy, wherein t is the time index, is the time offset of the grid energy, is the aligned energy supply prediction data of the distributed photovoltaic energy, is the time offset of the distributed photovoltaic energy, is the aligned energy supply prediction data of the energy storage system energy, is the time offset of the energy storage system energy;

[0095] Step S35: energy response lag correction, specifically, through the response curve modeling, the response time difference estimation, the lag alignment correction and the dynamic correction fusion, the energy response lag correction is performed, the multi-source energy supply prediction matrix data is output as the energy response lag corrected energy supply prediction data, to obtain the energy correction prediction data.

[0096] By performing the above operation, in the existing energy response lag correction method, there is a technical problem that in the existing energy dispatching system, only static offset method or empirical lag correction means is usually used for response delay compensation, the dynamic response curve of the equipment under different load conditions cannot be fully modeled, resulting in large lag estimation deviation, strategy execution failure in advance or delay, especially in the photovoltaic fluctuation or energy storage critical SOC interval. The present scheme creatively adopts a multi-source energy lag correction method combined with dynamic response curve reconstruction to correct the energy response lag, establishes a multi-type response behavior model for grid, photovoltaic and energy storage energy sources, and inversely calculates the lag step combined with historical energy supply behavior and control signal, dynamically corrects the prediction result, and achieves the goals of precise response alignment, time effectiveness enhancement and lag adaptive scheduling capability improvement.

[0097] Embodiment four, refer to Figure 1The embodiment is based on the above embodiment, and in step S4, the multi-objective optimization is used to generate a multi-objective optimal energy distribution strategy, specifically, an energy distribution particle swarm optimization algorithm combined with a hierarchical target penalty function is used to generate the energy distribution strategy reference data by constructing an improved hierarchical target penalty function and embedding it into the fitness evaluation link of the particle swarm optimization, and defining optimization variables, performing multi-objective optimization, and obtaining energy distribution strategy reference data;

[0098] The improved hierarchical target penalty function is specifically constructed by adaptively weighting the real-time response error, economic cost and total carbon emission as optimization targets, and introducing an energy response lag penalty term, and the calculation formula is:

[0099] ;

[0100] In the formula, F total is an improved hierarchical target penalty function, is a real-time response error weight, and F1 is a real-time response error target, is an economic cost weight, and F2 is an economic cost target, is a total carbon emission weight, F3 is a total carbon emission target, K is the total number of energy types, and k is the energy index, is the lag sensitivity weight of the kth energy, P delay (k) is an energy lag penalty term, and the specific calculation formula is , in which n k is the penalty coefficient of the kth energy, T d is the difference between the actual response time and the demand time, T e is the maximum allowed response time threshold;

[0101] The optimization variable specifically includes total load demand distribution, energy supply upper limit and device physical constraint;

[0102] The embedding into the fitness evaluation link of the particle swarm optimization is specifically that for each particle representing a group of energy distribution schemes, the evaluation indexes of each energy distribution scheme in real-time response error, economic cost and total carbon emission are calculated, and the response lag degree of each energy in the energy correction prediction data is combined, the improved hierarchical target penalty function is used to accurately score the fitness of the particle, and the fitness parameter is obtained; by using the standard particle swarm optimization algorithm and according to the fitness parameter, the position and speed of the particle in the search space are dynamically adjusted to gradually approach the optimal energy consumption distribution strategy under multi-objective trade-off, and the energy distribution strategy reference data is obtained.

[0103] By performing the above operation, for in the existing multi-objective optimization and control method, there is a current multi-objective optimization method often relies on the traditional weighted method or Pareto solution set method for energy deployment, unable to combine the response behavior and regulation delay characteristics of different energy types to construct an optimization target structure with real-time adaptability, resulting in an optimization result that is theoretically reasonable but poor in practical executability, especially lacking a strategy penalty constraint for response lag equipment, the scheme creatively adopts an energy distribution particle swarm optimization algorithm combined with a hierarchical target penalty function, by constructing an improved hierarchical target penalty function and embedding it into the fitness evaluation link of particle swarm optimization, and defining optimization variables, multi-objective optimization is performed, realizing a multi-source energy distribution optimization mechanism with multi-objective index fusion, response delay dynamic penalty and enhanced strategy executability.

[0104] In an embodiment six, referring to Figure 1 This embodiment is based on the above-mentioned embodiments, in step S5, the intelligent energy consumption distribution is used to realize intelligent energy regulation and coordination, specifically, the optimal energy supply distribution of various types of energy is analyzed according to the energy distribution strategy reference data, and control data frames are formatted by combining the operation state parameters of various types of energy, to obtain intelligent energy consumption distribution control instructions;

[0105] The intelligent energy consumption distribution control instructions specifically include timestamp, energy type, control power value, control allowable range value, instruction type, delay lag compensation parameter and control priority level.

[0106] It should be noted that in this paper, relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the term "includes", "contains" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device.

[0107] Although the embodiments of the present application have been shown and described, it can be understood by those of ordinary skill in the art that various changes, modifications, replacements and variations can be made to these embodiments without departing from the principles and spirits of the present application.

[0108] The above describes the present application and its embodiments, which are not limited, and the drawings only show one of the embodiments of the present application, and the actual structure is not limited thereto. In general, if a person skilled in the art is inspired thereby, without departing from the purpose of the present application, without creative design, similar structure and embodiments of the technical solution are not creative, and should belong to the protection scope of the present application.

Claims

1. An intelligent energy consumption allocation method for a multi-source energy system, characterized by: The method comprises the following steps: Step S1: energy supply data fusion to obtain an energy supply fusion data set; Step S2: dual-channel joint prediction is used for dual-channel modeling of energy availability and user load. Specifically, based on the energy supply fusion dataset, a dual-domain attention decoupling fusion joint supply and demand prediction method is used to perform dual-channel joint prediction to obtain joint energy supply and demand change prediction data, including the following steps: Step S21: Dual-channel data construction, specifically, based on the energy supply fusion dataset, extracting external energy availability data as first-channel raw sequence data input, and extracting user load record data as second-channel raw sequence data input, performing normalization processing on the first-channel raw sequence data input and the second-channel raw sequence data input, and introducing a time embedding vector based on a sine and cosine function, thereby constructing dual-channel input data; Step S22: constructing a dual-channel feature extraction network, specifically, constructing a dual-channel feature extraction network based on the dual-channel input data, and performing dual-channel feature extraction using the dual-channel feature extraction network to obtain energy availability feature data and user historical load feature data; The dual-channel feature extraction network includes an energy availability channel and a user load channel; Step S23: Decoupled attention mechanism modeling, specifically, constructing a self-attention module for each of the energy availability feature data and the user historical load feature data to obtain decoupled attention feature data; the decoupled attention feature data specifically includes energy availability decoupled attention features and user historical load decoupled attention features; Step S24: Dual-domain fusion residual correction, specifically, concatenating the energy availability decoupled attention feature and the user history load decoupled attention feature and inputting the concatenated features into a multi-layer perceptron module to obtain a fused dual-domain feature, and performing a dual-domain fusion residual correction on the fused dual-domain feature by introducing a cross-channel residual correction function to obtain corrected fusion feature data; Step S25: dual-channel joint prediction, specifically, decoding and predicting based on the modified fusion feature data to obtain joint supply and demand prediction data, and obtaining joint energy supply and demand change prediction data by subdividing the joint supply and demand prediction data; Step S3: Energy response lag correction is used to model and correct the response time and efficiency of different energy sources. Specifically, based on the energy supply fusion data set and the joint energy supply and demand change prediction data, a multi-source energy lag correction method combined with dynamic response curve reconstruction is used to perform energy response lag correction to obtain energy correction prediction data, including the following steps: Step S31: Response curve modeling, specifically, based on the combined energy supply and demand change prediction data, by collecting the historical control instruction sequence of each energy source and the actual output power at the corresponding time, performing response curve model fitting to obtain energy response curve data; the response curve model fitting specifically refers to fitting the grid energy and energy storage system energy using an improved first-order delayed linear function; and fitting the distributed photovoltaic energy using a saturation improved nonlinear exponential function; The improved first-order delayed linear function specifically introduces a dynamic response gain adjustment parameter and a control signal fluctuation suppression factor to improve the anti-winding capability; The saturation-improved nonlinear exponential function specifically improves the nonlinear change modeling capability by constructing a model with saturation gating and power upper limit fitting mechanisms; Step S32: Response time difference estimation, specifically, constructing a hysteresis time window search model and calculating the maximum correlation coefficient between the control signal and the response value of each energy source by sliding the hysteresis step, and calculating the maximum correlation hysteresis value parameter for the grid energy and the energy storage system energy. For the distributed photovoltaic energy, the maximum correlation hysteresis value parameter is estimated using the hysteresis correlation peak positioning algorithm to obtain the optimal response delay estimation parameter; Step S33: Lag alignment correction, specifically, performing time alignment processing based on the energy response curve data and the optimal response delay estimation parameter to obtain aligned energy supply prediction data; the time alignment processing is performed by calculating the grid energy and energy storage system energy by inverse linear response model; and the distributed photovoltaic energy is calculated by numerical interpolation approximation solution; The linear response model specifically refers to the data model corresponding to the improved first-order delayed linear function; The numerical interpolation approximation solution specifically refers to dynamic time interpolation based on the saturation-improved nonlinear exponential function; Step S34: Dynamic correction and fusion, specifically, constructing an aligned comprehensive energy supply prediction vector including grid energy, distributed photovoltaic energy, and energy storage system energy based on the aligned energy supply prediction data, to obtain multi-source energy supply prediction matrix data; Step S35: Energy response lag correction, specifically, performing energy response lag correction through the response curve modeling, the response time difference estimation, the lag alignment correction, and the dynamic correction fusion, outputting the multi-source energy supply prediction matrix data as energy supply prediction data after energy response lag correction, and obtaining energy correction prediction data; Step S4: Multi-objective optimization: Based on the energy correction prediction data, an energy allocation particle swarm optimization algorithm combined with a hierarchical objective penalty function is used. An improved hierarchical objective penalty function is constructed and embedded into the fitness evaluation link of the particle swarm optimization, and optimization variables are defined to perform multi-objective optimization to obtain energy allocation strategy reference data. The improved hierarchical objective penalty function is specifically constructed by adaptively weighting the real-time response error, economic cost, and total carbon emissions as optimization objectives, and introducing an energy response lag penalty term. The optimization variables specifically include the total load demand allocation, the energy supply upper limit, and the equipment physical constraints. Step S5: Intelligent energy consumption allocation, formatting and generating a control data frame to obtain an intelligent energy consumption allocation control instruction.

2. The intelligent energy consumption allocation method for a multi-source energy system according to claim 1, characterized in that: In step S1, the energy supply data fusion is used to collect and integrate multiple energy and environmental information data, specifically to uniformly collect, classify and structure grid energy, distributed photovoltaic energy, energy storage system energy and related environmental variables, and to collect structured operation data and control records for each energy source to obtain an energy supply fusion data set; The energy supply fusion data set specifically includes operation control data and external environment data.

3. The intelligent energy consumption allocation method for a multi-source energy system according to claim 2, characterized in that: In step S2, the combined energy supply and demand change prediction data specifically includes the grid energy available prediction value, the distributed photovoltaic energy available prediction value, the energy storage system energy available prediction value, the user load demand prediction value and the supply and demand change prediction difference.

4. The intelligent energy consumption allocation method for a multi-source energy system according to claim 3, characterized in that: In step S4, the multi-objective optimization is used to generate a multi-objective optimal energy allocation strategy. Specifically, based on the energy correction prediction data, an energy allocation particle swarm optimization algorithm combined with a hierarchical objective penalty function is adopted. By constructing an improved hierarchical objective penalty function and embedding it into the fitness evaluation link of the particle swarm optimization, and defining optimization variables, multi-objective optimization is performed to obtain energy allocation strategy reference data.

5. The intelligent energy consumption allocation method for a multi-source energy system according to claim 4, characterized in that: The fitness evaluation link embedded in the particle swarm optimization specifically targets a set of energy allocation schemes represented by each particle, calculates the evaluation indicators of each energy allocation scheme in terms of real-time response error, economic cost, and total carbon emissions, and combines the response lag degree of each energy source in the energy correction prediction data with the improved hierarchical objective penalty function to accurately score the fitness of the particles and obtain fitness parameters. By adopting the standard particle swarm optimization algorithm and based on the fitness parameters, the particles are guided to dynamically adjust their position and speed in the search space, gradually approaching the optimal energy consumption allocation strategy under multi-objective trade-offs, and obtaining energy allocation strategy reference data.

6. The intelligent energy consumption allocation method for a multi-source energy system according to claim 5, characterized in that: In step S5, the intelligent energy consumption allocation is used to realize intelligent energy regulation and coordination. Specifically, the optimal energy supply allocation of each type of energy is analyzed based on the energy allocation strategy reference data, and the control data frame is formatted and generated in combination with the operating status parameters of each type of energy to obtain the intelligent energy consumption allocation control instruction; The intelligent energy consumption allocation control instruction specifically includes a timestamp, an energy type, a control power value, a control allowable range value, an instruction type, a delay lag compensation parameter and a control priority level.

Images