CN117764438A - Comprehensive energy equipment response value quantification method and system - Google Patents
Comprehensive energy equipment response value quantification method and system Download PDFInfo
- Publication number
- CN117764438A CN117764438A CN202311634448.4A CN202311634448A CN117764438A CN 117764438 A CN117764438 A CN 117764438A CN 202311634448 A CN202311634448 A CN 202311634448A CN 117764438 A CN117764438 A CN 117764438A
- Authority
- CN
- China
- Prior art keywords
- energy
- comprehensive
- model
- equipment
- demand response
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000004044 response Effects 0.000 title claims abstract description 167
- 238000000034 method Methods 0.000 title claims abstract description 67
- 238000011002 quantification Methods 0.000 title claims description 17
- 238000009826 distribution Methods 0.000 claims abstract description 48
- 238000005457 optimization Methods 0.000 claims abstract description 38
- 230000035945 sensitivity Effects 0.000 claims abstract description 27
- 238000011156 evaluation Methods 0.000 claims abstract description 25
- 238000013528 artificial neural network Methods 0.000 claims abstract description 24
- 238000005070 sampling Methods 0.000 claims abstract description 14
- 238000006243 chemical reaction Methods 0.000 claims description 31
- 230000005611 electricity Effects 0.000 claims description 26
- 230000006870 function Effects 0.000 claims description 21
- 238000004364 calculation method Methods 0.000 claims description 16
- 230000008859 change Effects 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 16
- 238000000354 decomposition reaction Methods 0.000 claims description 15
- 230000014509 gene expression Effects 0.000 claims description 14
- 238000005315 distribution function Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 9
- 230000003993 interaction Effects 0.000 claims description 9
- 238000004422 calculation algorithm Methods 0.000 claims description 8
- 230000001186 cumulative effect Effects 0.000 claims description 8
- 238000003860 storage Methods 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 5
- 239000007789 gas Substances 0.000 description 29
- 238000004146 energy storage Methods 0.000 description 14
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 12
- 239000011159 matrix material Substances 0.000 description 11
- 239000003795 chemical substances by application Substances 0.000 description 10
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 10
- 230000008878 coupling Effects 0.000 description 8
- 238000010168 coupling process Methods 0.000 description 8
- 238000005859 coupling reaction Methods 0.000 description 8
- 238000005265 energy consumption Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 8
- 238000006467 substitution reaction Methods 0.000 description 8
- 229910002092 carbon dioxide Inorganic materials 0.000 description 6
- 239000001569 carbon dioxide Substances 0.000 description 6
- 238000001816 cooling Methods 0.000 description 6
- 230000009194 climbing Effects 0.000 description 5
- 238000007599 discharging Methods 0.000 description 5
- 239000003345 natural gas Substances 0.000 description 5
- 238000010206 sensitivity analysis Methods 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000002829 reductive effect Effects 0.000 description 4
- 230000001276 controlling effect Effects 0.000 description 3
- 238000013178 mathematical model Methods 0.000 description 3
- 239000002245 particle Substances 0.000 description 3
- 230000009467 reduction Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000000903 blocking effect Effects 0.000 description 2
- 229910052799 carbon Inorganic materials 0.000 description 2
- 230000000295 complement effect Effects 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- AMXOYNBUYSYVKV-UHFFFAOYSA-M lithium bromide Chemical compound [Li+].[Br-] AMXOYNBUYSYVKV-UHFFFAOYSA-M 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 238000010248 power generation Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000011084 recovery Methods 0.000 description 2
- 238000009827 uniform distribution Methods 0.000 description 2
- 239000002918 waste heat Substances 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 238000000342 Monte Carlo simulation Methods 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000012983 electrochemical energy storage Methods 0.000 description 1
- 238000005430 electron energy loss spectroscopy Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000007717 exclusion Effects 0.000 description 1
- 239000000446 fuel Substances 0.000 description 1
- 229910021389 graphene Inorganic materials 0.000 description 1
- 238000005338 heat storage Methods 0.000 description 1
- 230000003116 impacting effect Effects 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000036961 partial effect Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000011158 quantitative evaluation Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 238000005057 refrigeration Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a system for quantifying response value of comprehensive energy equipment, wherein the method comprises the following steps: acquiring cost data of equipment operation, and establishing an energy hub optimization operation model of a comprehensive energy system taking comprehensive demand response into consideration; simulating the distribution of the key parameters by adopting random sampling, and solving an operation model; using a PSO-BP neural network proxy model to replace the comprehensive energy system energy hub optimization operation model considering the comprehensive demand response; calculating the evaluation index of each demand response value by using the output result of the agent model, calculating the global sensitivity index of key influence factors of each equipment participating in the demand response to the evaluation index of the demand response value of the multi-energy equipment, and obtaining the quantitative result of the response value of the multi-dimensional comprehensive energy equipment according to the probability distribution model. Technical support is provided for the classification method for establishing the comprehensive application value of the multifunctional equipment of different types of comprehensive energy systems such as the industry, the commerce and the like, and more accurate strategy support is provided for comprehensive energy service providers.
Description
Technical Field
The invention relates to the technical field of response value quantification of energy equipment, in particular to a comprehensive response value quantification method and system of energy equipment.
Background
The power demand response is a flexible and quick response means, can influence the power price and the system operation in a short period, and can improve the demand elasticity, smooth load curve and the like in a short period. The method has the characteristics of shifting peaks and filling valleys, reducing the power demand in peak time, and improving the stability and efficiency of the operation of the power grid; the operation mode of the power plant can be optimized, the capacity of the power grid for absorbing more intermittent distributed energy sources is enhanced, and the interaction level of the power grid and power users is improved. The power demand response technology has become the power supply protection measure adopted by each level of power authorities and power grid enterprises in the power marketing environment. With the development of energy technology, limitations of traditional power demand response are gradually revealed. On the one hand, the time shift of the load not only reduces the comfort level of the user to a certain extent, but also influences the output value and the satisfaction of the user, and is greatly influenced by the wish of the user; on the other hand, along with the acceleration of comprehensive energy technology and engineering construction, the trend of energy diversification is remarkable, the 'multi-energy complementation' characteristic of the comprehensive energy system can realize the conversion between electric power and other energy, and the overall external adjustable characteristic of the system is more obvious. The comprehensive demand response is generated as an extension of the traditional power demand response, and the comprehensive energy demand response not only has the characteristic of the power demand response, but also enables a user to reform the energy supply mode when the system reliability is threatened or the electricity price is high in peak period, so that the range of the power adjustable load equipment library is greatly expanded, and the power adjustable load equipment library is a beneficial supplement of the traditional power demand side management and demand response technology.
The main objective of the comprehensive energy demand response is to optimize the supply and demand relation of the comprehensive energy system, solve the contradiction between supply and demand, realize the goals of peak clipping and valley filling of electric power, frequency modulation and voltage regulation, new energy elimination, relief of line heavy load and the like, but at the present stage, the comprehensive energy demand response still has some problems, such as complex physical coupling of each unit of the comprehensive energy system, large response characteristic difference of multi-energy equipment, undefined coupling response mechanism of multi-type equipment, and incapability of effectively carrying out efficient regulation on equipment under regional energy sources such as workers/suppliers/people. Therefore, it is necessary to provide quantitative application values of multi-energy utilization in various scenes, so as to provide technical support for classification methods of multi-energy equipment comprehensive application values of different types of comprehensive energy systems such as operators and the like, and also provide more accurate decision support for comprehensive energy service providers.
Disclosure of Invention
The present invention has been made in view of the above-described problems.
Therefore, the technical problems solved by the invention are as follows: the existing energy equipment response value quantity method has optimization problems that the influence of key factors on modeling is not considered in a model, quantitative evaluation on the value of park equipment cannot be carried out, and the like.
In order to solve the technical problems, the invention provides the following technical scheme: a method for quantifying response value of comprehensive energy equipment comprises the following steps:
acquiring cost data of equipment operation, and establishing an energy hub optimization operation model of a comprehensive energy system taking comprehensive demand response into consideration;
simulating the distribution of the key parameters by adopting random sampling, and solving an operation model;
using a PSO-BP neural network proxy model to replace the comprehensive energy system energy hub optimization operation model considering the comprehensive demand response;
calculating the evaluation index of each demand response value by using the output result of the agent model, calculating the global sensitivity index of key influence factors of each equipment participating in the demand response to the evaluation index of the demand response value of the multi-energy equipment, and obtaining the quantitative result of the response value of the multi-dimensional comprehensive energy equipment according to the probability distribution model of the global sensitivity index.
As a preferable scheme of the comprehensive energy equipment response value quantification method, the invention comprises the following steps: the comprehensive energy system energy hub optimization operation model considering the comprehensive demand response comprises the steps of determining an objective function of the model:
min C＝min{1000·[C G +C P +C IDR ]}
wherein C is G In order to purchase the gas at a cost, For the unit price of purchasing gas, the unit is Yuan/m 3; c (C) P The electricity purchasing cost is realized; price of electricity implemented for the grid ∈ ->The unit is Yuan/(kW.h) for electricity purchasing price; p (P) t PG Electric power is purchased for IES in MW; c (C) IDR Is IDR cost;the unit is meta/(kW.h) for IDR compensation unit price; />The unit is MW for the change in power consumption of the ith AC response IDR.
As a preferable scheme of the comprehensive energy equipment response value quantification method, the invention comprises the following steps: determining constraints of the integrated energy system energy hub optimization run model that take into account the integrated demand response includes,
wherein P is c,t Representing the output of the CHP unit on the electric load supply side, W t Representing the output of the wind power electric load supply side, P pv,t Indicating the output of the photovoltaic load supply side, L e,t Represent DeltaL e,t Expressed, L h,t Represent DeltaL h,t Representation, H c,t Representing the output of the heat load supply side of the CHP unit, H GB,t Indicating the output of the heat load supply side of the gas turbine, H ET,t Representing the output of the electric boiler on the heat load supply side, P representing the system input,representing the upper limit of the input quantity of the system,Prepresents the lower limit of the input quantity of the system, g (P c,t ，H ct ,，H o,t ) Less than or equal to 0 represents the operation constraint of the CHP unit, the electric boiler and the hot spring equipment, and h (delta L e,t ，ΔL h,t ) And less than or equal to 0 represents the constraint condition that the active response of electric power and heat is satisfied.
As a preferable scheme of the comprehensive energy equipment response value quantification method, the invention comprises the following steps: modeling the key parameters of equipment in the model, simulating the key parameter distribution by adopting random sampling, and solving the model;
energy conversion equipment efficiency criticality modeling: comprehensive synthesisThere are multiple comprehensive energy devices in the energy system, and under the operating condition of considering the variable working condition, the efficiency is subjected to distribution:
modeling the critical parameters of the equipment capacity: there are a plurality of comprehensive energy devices in the comprehensive energy system, and the capacity critically and uniformly describes:
wherein,representing the maximum capacity of the r-th device, +.>The lower limit of the expression is S r The upper limit is->Is the electrical efficiency of the multi-functional device, +.>The distribution of the parameters μ, δ;
randomly sampling the critical parameters to generate a random number sequence with the size of N;
and after N groups of critical samples are obtained, substituting the N groups of critical samples into the comprehensive energy system energy hub optimization operation model considering the comprehensive demand response to solve.
As a preferable scheme of the comprehensive energy equipment response value quantification method, the invention comprises the following steps: the PSO-BP neural network comprises the steps that the key parameters are used as input variables of training samples, and solving results are used as output variables of the training samples by substituting the input variables into the comprehensive energy system energy hub optimization operation model considering comprehensive demand response;
Training a PSO-BP neural network agent model, wherein in the BP neural network, a PSO optimization algorithm optimizes the weight and bias of the neural network, and combines the PSO optimization algorithm to construct the PSO-BP neural network agent model;
predicting output variables by proxy model:
generating a large number of new input samples according to the probability distribution of the input variables; a large amount of new sample data is taken as input, thereby obtaining a prediction result.
As a preferable scheme of the comprehensive energy equipment response value quantification method, the invention comprises the following steps: the value of the multi-energy equipment demand response value evaluation index is as follows:
y＝f(x),x∈K d
wherein x represents a space defined in K d Input variables of (a); d represents K d The collection contains the number of elements; y represents the output response of the system, eta 1 ,η 2 ,η 3 … represents the energy conversion efficiency of the multiple energy in the system,representing the capacity of the multi-energy device in the system;
and (3) performing Sobol decomposition on the output response y of the system:
wherein f 0 Represent constant term, f i Representing a polynomial determined by 1 input variable i, f ij Representing a polynomial determined by two input variables ij, f 1,...,d A polynomial representing a decision of d input variables; x is x i Representing the i-th input variable; x is x j Represents the j-th input variable; x is x d Represents the d-th input variable;
The global sensitivity index theoretical calculation includes,
calculating the variance of the output response y=f (x):
the variances of all the steps in the Sobol decomposition are called as all the order variances, and the total variance of the output response y is decomposed into the sum of all the order variances:
the s-order bias is expressed as integral:
wherein,represents the ith s Input variables->Representing the decomposition of Sobol by i s The square of the sub-formula determined by the individual input variables. V (V) ij Representing the sub-variance, i.e. the second order variance, V, of the Sobol decomposition determined by the ith and jth input variables together 1,2,...d The sub-variances determined by the 1,2,3, …, d input variables in the Sobol decomposition are d-order variances; d represents the set K d The number of the elements is included;
defining Sobol indexes of each order based on the calculation result of the variance:
wherein,for second order global sensitivity, the ith is represented 1 And (i) 2 Influence of the interaction of the random variables on the variance of the output response, and so on,/o>Represents the ith 1 ，i 2 ，…,i s The effect of the interaction of the individual random variables on the variance of the output response; defining an ith input variable X based on expected and variance calculations of the variables i The contribution to the output variance is noted as the global sensitivity index S Ti ：
Wherein X is ～i Represents dividing X i The set of all variables except for Y represents that the output response y=f (x) is considered as a random variable.
As a preferable scheme of the comprehensive energy equipment response value quantification method, the invention comprises the following steps: calculating the probability distribution model, wherein the obtaining of the multi-dimensional comprehensive energy device response value quantification result comprises taking each evaluation index value as one-dimensional data, and d sample data are obtained: s is S T1 、S T2 、S T3 …S Td ；
Let the cumulative distribution function of the sample data be F (S), the probability density function be F (S):
wherein h represents width, S Ti-1 Previous sample data representing the ith sample data;
an empirical distribution function incorporating a cumulative distribution function representing the sum of the discrete variables S Ti In other words, the sum of occurrence probabilities of all values equal to or less than α:
wherein d represents the number of sample data;
using S in d observations Ti The ratio of the number of occurrences of alpha to d is used to approximate the probability; substituting the empirical distribution function into f (S i ) The method comprises the following steps:
after determining the bandwidth of h, the expression of f (S):
wherein S is Ti Representing the ith sample data; s is S Tj The representation size is between S Ti -h and S Ti Sample data of +h;
after the f (S) expression is obtained, the distribution condition of each evaluation index is obtained, so that the multidimensional response value judgment of the comprehensive energy equipment is obtained.
The invention relates to a comprehensive energy equipment response value quantification system adopting the method, which is characterized in that:
the acquisition unit is used for acquiring cost data of equipment operation and establishing an energy hub optimization operation model of the comprehensive energy system taking comprehensive demand response into consideration;
the solving unit adopts random sampling to simulate the distribution of the key parameters and solve the operation model;
the training unit uses a PSO-BP neural network agent model to replace the comprehensive energy system energy hub optimization operation model considering the comprehensive demand response;
the computing unit is used for computing the evaluation index of each demand response value by using the output result of the agent model, computing the global sensitivity index of the key influence factors of each equipment participating in the demand response to the evaluation index of the demand response value of the multi-energy equipment, and obtaining the quantitative result of the response value of the multi-dimensional comprehensive energy equipment according to the probability distribution model.
A computer device, comprising: a memory and a processor; the memory stores a computer program characterized in that: the processor, when executing the computer program, implements the steps of the method of any of the present invention.
A computer-readable storage medium having stored thereon a computer program, characterized by: the computer program, when executed by a processor, implements the steps of the method of any of the present invention.
The invention has the beneficial effects that: according to the comprehensive energy equipment response value quantification method provided by the invention, a refined comprehensive energy system energy hub optimization operation model considering comprehensive demand response is constructed, then key parameters concerned in the model are modeled and sampled, and the model is solved through MATLAB based on a sampling result. And finally, introducing global sensitivity analysis, determining input variables and output variables in the global sensitivity analysis, solving the process by utilizing MATLAB and CPLEX to obtain a multi-energy equipment response value quantification result, providing technical support for a classification method for preparing the multi-energy equipment comprehensive application values of different types of comprehensive energy systems such as industry and commerce and providing more accurate strategy support for comprehensive energy service providers.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. Wherein:
FIG. 1 is a general flow chart of a method for quantifying the response value of an integrated energy device according to a first embodiment of the present invention;
FIG. 2 is a diagram of an integrated energy system architecture for a method for quantifying a response value of an integrated energy device according to a second embodiment of the present invention;
FIG. 3 is a topology structure diagram of a neural network model of a method for quantifying the response value of a comprehensive energy device according to a second embodiment of the present invention;
fig. 4 is a cumulative probability distribution diagram of a method for quantifying response value of an integrated energy device according to a second embodiment of the present invention.
Detailed Description
So that the manner in which the above recited objects, features and advantages of the present invention can be understood in detail, a more particular description of the invention, briefly summarized above, may be had by reference to the embodiments, some of which are illustrated in the appended drawings. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
Example 1
Referring to fig. 1, for one embodiment of the present invention, there is provided a method for quantifying a response value of an integrated energy device, including:
S1: and constructing an energy hub optimization operation model of the comprehensive energy system considering the comprehensive demand response.
Determining an objective function of the model, as shown in the following formula:
min C＝min{1000·[C G +C P +C IDR ]}(1)
wherein: c (C) G In order to purchase the gas at a cost,for the unit price of purchasing gas, the unit is Yuan/m 3; c (C) P The electricity purchasing cost is realized; price of electricity implemented for the grid ∈ ->The unit is Yuan/(kW.h) for electricity purchasing price; p (P) t PG Electric power is purchased for IES in MW; c (C) IDR Is IDR cost;the unit is meta/(kW.h) for IDR compensation unit price; />The unit is MW for the change in power consumption of the ith AC response IDR.
Further, determining constraint conditions of the model, including the following constraints:
1. electric, thermal, gas load integrated demand response constraints:
a. electrical load demand response constraints:
considering that the movable electric load and the interruptible electric load are taken as regulation means to participate in the power demand response, the mathematical expression is as follows:
P t L* ＝P t L +P t move -P t cut +ΔL t e.AL (5)
wherein P is t L 、P t L* The total amount of electric loads before and after the demand response is respectively counted for a system t period; p (P) t move 、P t cut The time-variable electric load and the interruption-variable electric load of the system t period are respectively; ΔL t e.AL And (5) replacing the total change quantity of the load after the demand response for the moment t of the electric load.
The time-shifting load refers to the load that reaches the reduced or shifted energy usage period by temporarily changing the user's energy usage habit during the peak load period. The method is characterized in that the energy utilization time is flexible, the energy utilization total amount is constant, and the method can be expressed by the condition shown in a formula (6).
Wherein P is t move.max The maximum time-shifting load of the t time period is set to be 5% of the total amount of the electric load of the t time period for the maximum time-shifting load of the t time period of the system.
The interruptible load means that the partial load can be reduced under the premise of not influencing normal life and work of the energy utilization side, and the condition is described as shown in a formula (7).
0≤P t cut ≤P t cut.max (7)
Wherein P is t cut.max The maximum interruptible electrical load of the t period is set to be 5% of the total electrical load of the t period for the maximum interruptible electrical load of the t period of the system.
b. Gas load demand response constraints:
natural gas and electricity are the primary energy sources of IES, both of which have similar market attributes. Considering the time-variable load as a regulating measure to participate in the gas load demand response, the mathematical expression is as follows:
V t load* ＝V t load +V t move +ΔL t g.AL ,0≤V t move ≤V t move.max (8)
wherein V is t load 、V t load* The total gas load before and after the demand response is calculated for the system t period; v (V) t move 、V t move.max The time-shifting air load and the time-shifting air load maximum value of the system in the time t period are respectively set, and the time-shifting air load of the system in the time t period is set to be 10% of the total air load of the time t period; ΔL t g.AL Load total variation after demand response for gas load t moment substitution typeAmount of conversion.
c. Thermal load demand response constraints:
because the human body has certain ambiguity on the perception of the temperature comfort and the heat supply has delay, the indoor temperature is adjusted within a certain range, so that the comfort of a user is not affected. The time-shifting heat load is taken as a regulation measure to participate in heat load demand response, and the total heat load and the time-shifting load amount can be expressed as:
In the method, in the process of the invention,respectively calculating the total heat load before and after the demand response for the system t period; />Setting the maximum time-shifting heat load of the time period t as 10% of the total heat load of the time period t for the time period t of the system respectively as the time-shifting heat load and the maximum time-shifting heat load of the time period t of the system; />And (5) responding to the total change of the post-load for the replacement type demand at the moment of the thermal load t.
d. Alternative demand response:
considering the mutual substitution of electric, gas and heat loads, the user can select the energy types meeting the same energy and quality requirements according to the price relative relation among the energy sources. As shown in the substitution type demand response mathematical model, when constructing the multi-energy substitution type demand response mathematical model, the substitution direction between energy sources needs to be considered, and the following substitution directions are set as positive directions: the electrical load is replaced by a gas load and the thermal load is replaced by a gas load; in addition, the conversion coefficient of the effective heat value between the energy sources is based on the replaced energy source.
ΔL t e.AL ＝-θ eg ΔL t eg -θ eh ΔL t eh
ΔL t g.AL ＝θ eg μ eg ΔL t eg +θ hg μ hg ΔL t hg
ΔL t h.AL ＝θ eh μ eh ΔL t eh -θ hg ΔL t hg
0≤ΔL t ij ≤ΔL ij t,max ,ΔL t ij ＞0
0≤|μ ij ΔL t ij |≤ΔL ij t,max ',ΔL t ij ＜0
0≤|ΔL t i.AL |≤ΔL i.AL t,max (10)
Wherein i, j is { e, g, h }, i is not equal to j, e, g, h respectively refer to three types of energy of electricity, gas and heat;the total load change quantity and the upper limit of the total load change quantity after t time substitution type demand response are respectively; />The load substitution quantity between two energy sources at the moment t is positive, which represents that the former energy source is replaced by the latter energy source, and the value is negative, otherwise; θ eg 、θ eh 、θ hg Representing the replacement state between two energy sources, taking 1 when the load replacement direction is positive, otherwise taking-1; mu (mu) eg 、μ eh 、μ hg Taking 1.5, 2.8 and 0.625 as effective heat value conversion coefficients among energy sources respectively; />Is the load replaced amount and the upper limit of the replaced amount.
Furthermore, the total amount of load involved in IDR should be within a certain interval:
in the method, in the process of the invention,expressed as t time and the amount of load after IDR; />The upper and lower limits of the load after IDR are respectively calculated, and the electric and gas loads are taken to be +.>Heat load taking->
2. Input-output constraints:
the energy hub is essentially a functional relation describing input to output of the multi-energy system, and since only devices such as energy transmission, conversion and storage exist in the system, a coupling matrix (coupling matrix) can be used to describe the relation between the two, and the relation is denoted by C, and then:
wherein L is n The output of the nth energy hub; p (P) m The input of the mth energy hub; c nm The coupling factor is the ratio of the n-th form of energy output to the m-th form of energy input.
The input to output of various forms of energy can be divided into two steps: energy distribution and energy transmission or conversion. Energy distribution refers to the distribution of various energy sources to different energy transmission or conversion devices in a certain proportion. The energy transmission or conversion means that the energy is converted through mechanical, chemical and other ways after being input into the equipment, and the energy has certain conversion efficiency. The coupling matrix in equation (12) can be further decomposed:
In the formula, v nm As a distribution factor, representing the proportion of the energy input of the nth form to be distributed to the energy input of the mth form for conversion, and corresponding N is a distribution matrix; η (eta) nm The energy is an efficiency factor, the efficiency of converting the energy of the nth form into the energy of the mth form is represented, and the corresponding eta is an efficiency matrix; p'. m Representing P m The energy after conversion; l (L) n Representing P' m And (5) energy after conversion.
3. Other constraints:
each energy conversion device needs to meet the upper and lower limit constraints and the climbing rate constraint. For the energy storage device, besides the upper limit constraint, the lower limit constraint and the climbing rate constraint of power, capacity constraint and mutual exclusion constraint are also required to be met, and the storage battery is also required to meet the charge and discharge frequency constraint, which is specifically as follows:
wherein: j represents the kind of each energy device,an upper limit of output power of the energy conversion device a; />The operation state flag bit is the operation state flag bit of the energy source equipment a; />And->Respectively->Minimum and maximum climbing rate of (2); k represents the type of each energy storage device; />And->Respectively charging and discharging power of the energy storage device k; />Is the capacity state of the energy storage device k;and->The upper limit and the lower limit of the energy charging and discharging power of the energy storage device k are respectively set; />And->The minimum and maximum values of the capacity of the energy storage device k are respectively; / >And->The capacity states of the energy storage device k at initial and final moments in the period are respectively; />And-> Respectively charging and discharging the energy storage equipment k, and climbing the upper limit and the lower limit of the slope rate; />And->Respectively charging and discharging power state zone bits of the energy storage device k; t is the maximum charge and discharge frequency number of the storage battery.
4. Electric power balance constraint:
P t d +P t EB +P t L* ＝V t MT η MT LH gas +P t WT +V t CHP η CHP LH gas +P t PV +P t R (16)
wherein eta is MT 、η CHP The gas-electricity conversion coefficients of the gas turbine and the CHP are respectively; v (V) t MT 、V t CHP The volume of natural gas consumed by the gas turbine and the cogeneration unit in the t time period is respectively; LH (LH) gas The value of the natural gas is 9.7kW.h/m 3.
5. Thermal power balance constraint:
wherein mu is MT 、μ CHP The gas-heat conversion efficiency of the gas turbine and the cogeneration unit is respectively; mu (mu) EB Is the electrothermal conversion coefficient of the electric boiler.
6. User satisfaction constraints:
when guiding a user to perform energy utilization adjustment, the energy utilization satisfaction degree of the user needs to be considered, and the more the energy utilization adjustment behaviors of the user are, the lower the energy utilization satisfaction degree of the user is, the mathematical model is as follows:
in zeta SEU Satisfaction of the user;a lower limit of satisfaction for the user; n represents the set of all energy categories.
In summary, the basic form of the established energy hub optimization operation model of the comprehensive energy system considering the comprehensive demand response is shown as formula (19):
In the formula, the optimization target of the model is that the cost is minimized in min C; the decision variables of the model comprise input quantity of each energy source, distribution condition of the energy source among different devices, start-stop condition of the devices and the like;
P c,t +W t +P pv,t +P FC,t the output representing the electric load supply side consists of a CHP unit, wind power, photovoltaic and a fuel cell; l (L) e,t +ΔL e,t 、L h,t +ΔL h,t Representing the load amount after the integrated demand response is taken into account; ΔL e,t 、ΔL h,t Is the active response of the load; h c,t +H GB,t +H ET,t The output representing the heat load supply side consists of a CHP unit, a gas turbine and an electric boiler; l=cp is the operation constraint of the energy hub itself, reflecting the coupling relationship of the energy input and output of each form;representing system input quantity constraint, namely energy system network constraint, equipment operation constraint and the like; g (P) c,t ,H c,t ,H o,t ) Less than or equal to 0 represents the operation constraint of the CHP unit, the electric boiler, the heat pump and other equipment, the constraint is inequality constraint, the feasible operation interval of the CHP unit is described, and a flexible adjustment space is created for the operation of the regional multi-energy system; h (DeltaL) e,t ,ΔL h,t ) Less than or equal to 0 meansThe active response of electric power and thermal power needs to meet certain constraint conditions, such as element characteristic constraint, upper and lower limit constraint of response and the like.
S2: and simulating the distribution of the key parameters by adopting random sampling, and solving an operation model.
The efficiency of the energy conversion equipment and the capacity of the comprehensive energy equipment are key to influence the comprehensive response value of the comprehensive energy conversion equipment and the output power feasible region of the comprehensive energy system, and the key parameters are used as key factors for sampling modeling to study the influence of the key parameters on the operation of the comprehensive energy system.
Various comprehensive energy devices exist in the comprehensive energy system, such as a central air conditioner, a lithium bromide refrigerator, a waste heat boiler, a heat storage pipe, electrochemical energy storage, a gas boiler, a gas turbine and the like, and the efficiency can be considered to follow a certain distribution under the condition of considering variable working conditions.
Where eta is the electrical efficiency of the multi-functional device,the parameters are the distribution of other parameters such as mu, delta and the like.
Modeling the critical parameters of the equipment capacity:
various comprehensive energy devices exist in the comprehensive energy system, such as energy storage devices (electricity, heat and gas), air conditioners, absorption refrigerators, CHP units and the like, and the capacity criticality of the comprehensive energy system can be uniformly described as follows:
in the method, in the process of the invention,for maximum capacity of the r-th device, +.>Is of the lower limit ofS r The upper limit is->Is a uniform distribution of (c).
The key parameters are sampled, and functions such as built-in rand, randn, wblrnd and the like can be called in the MATLAB platform to randomly sample. The function is a random number generation function based on uniform distribution of a pseudo-random number generator, and the generation process comprises the steps of generating a random number seed, generating a random number sequence, normalizing the random number sequence and the like to generate the needed corresponding distribution. In the present invention, a random number sequence of size N is generated using the above function, where N is the sample size.
S3: and training the PSO-BP neural network proxy model by taking the input-output variables as training samples. The PSO-BP neural network agent model can replace a comprehensive energy system energy hub optimization operation model considering comprehensive demand response, and the function of quickly obtaining an output variable from an input variable is realized.
The BP neural network proxy model is a model for approximating complex systems. The main objective is to simulate a complex process with a relatively simple model to enable faster calculation and analysis. The neural network agent model is trained by taking an input-output sample as a training set, taking the key parameter as an input variable of the training sample, and substituting the key parameter into the comprehensive energy system energy hub optimization operation model considering the comprehensive demand response to solve the result as an output variable of the training sample.
Particle swarm optimization (ParticleSwarmOptimization, PSO) is a swarm intelligence-based optimization algorithm that is generally used to solve continuous space optimization problems. In the BP neural network, a particle swarm optimization algorithm can be used for optimizing the weight and bias of the neural network and improving the performance of the neural network. The built-in functions such as net, train and the like can be called in the MATLAB platform, and a PSO-BP neural network proxy model can be built by combining a PSO optimization algorithm. Parameters such as the number of hidden layers and neurons, learning rate, maximum iteration times, target errors and the like need to be configured in the process of constructing the proxy model.
Based on the Monte Carlo simulation method, a large number of new input samples are generated according to the probability distribution of the input variables. On the MATLAB platform, a sim function can be called, and a large amount of new sample data is taken as input, so that a prediction result is obtained.
S4: calculating the evaluation index of each demand response value by using the output result of the agent model, calculating the global sensitivity index of key influence factors of each equipment participating in the demand response to the evaluation index of the demand response value of the multi-energy equipment, and obtaining the quantitative result of the response value of the multi-dimensional comprehensive energy equipment according to the probability distribution model.
The calculation of the multi-energy equipment demand response value evaluation index comprises the following steps:
user plane:
1) Satisfaction of the user in the power consumption mode.
When the terminal user does not participate in the demand response service, the satisfaction degree of the power utilization mode is highest, ζ SEU =1. After participating in the demand response service, zeta is the larger the change of the new load curve and the original load curve due to the change of the power utilization mode and the generation of the new load curve SEU The lower the value.
In zeta SEU Satisfaction of the user; n represents the set of all energy source types, T represents the energy source types in [0-T ]]And calculating the satisfaction degree of the energy consumption in the time period.
S 1 ∈[0,1]And the satisfaction degree of the power utilization mode of the user is indicated.
2) Satisfaction of the user's electricity fee expenditure.
When an end user participates in a demand response service, the change condition of the electric charge payment of the end user can influence the satisfaction level of the user. When the expenditure expense of the user exceeds the expectation of the user, the satisfaction level of the user may be at risk of decreasing, and the enthusiasm of the user for participating in the demand response service next time is affected, so the electricity expense change is also one of important indexes for evaluating the demand response service, and the expression mode is as follows:
/>
wherein S is 2 ∈[0,1]The satisfaction degree of the electricity expense of the user is represented; in the middle ofThe electricity charge and gas charge change quantity paid by the user in each time period before and after the user can be adjusted respectively; />The electricity fee and the gas fee of each period before the user can be adjusted respectively. S is S 2 The larger indicates that the more the user has reduced energy expenditure, the higher the user energy expenditure satisfaction level. T is represented by [0-T ]]The energy expenditure satisfaction is calculated in a time period.
3) The user synthesizes satisfaction.
The comprehensive satisfaction of the user is expressed by a weighted average of the satisfaction of the electricity consumption mode and the satisfaction of the electricity fee expenditure, and a measurement model of the comprehensive satisfaction is shown as follows:
S＝α 1 S 1 +α 2 S 2 (24)
α 1 +α 2 ＝1 (25)
Wherein S represents the comprehensive satisfaction of the user; alpha 1 、α 2 And respectively represent the weight coefficients of S1 and S2.
When the user pays attention to different power utilization modes and power charge expenditure, alpha with different values can be generated by reflecting the power utilization modes and the power charge expenditure on the weight coefficient 1 、α 2 . For example, a part of resident users pay more attention to electricity fee than the electricity utilization mode of the users, at this time alpha 2 Is larger. For some loads with severe working procedures and small variability of the power utilization mode, the user will pay more attention to the power utilization mode, and at this time alpha 1 Is larger.
Grid level:
1) Comprehensive energy utilization rate.
The comprehensive energy utilization rate refers to the ratio of the total energy produced by the system to the non-renewable energy consumption of the system, and is also called energy utilization rate or comprehensive energy efficiency. Because the input and output of the regional comprehensive energy system are mostly different energy sources, the different energy sources are converted to the same energy level by utilizing the effective heat value conversion coefficient among the energy sources in order to consider the difference among the different energy sources and the conversion efficiency of the energy conversion link. The conversion coefficient of the effective heat value among the energy sources reflects the power-applying capacity of the energy sources, and the conversion coefficient of the effective heat value among the energy sources of different energy sources is a fixed value under the same standard environment.
Wherein eta is ex The comprehensive energy utilization rate is achieved; omega shape out 、Ω in The energy types are respectively output and input; e's' i,out 、E i,in The total amount of energy to be output and input respectively; lambda (lambda) i Is the conversion coefficient of the ith energy;is the basic load quantity of the park; mu (mu) t,i Is an IDR participation status; />The IDR participation power of the ith load at the moment t; lambda (lambda) e 、λ g The electric power and gas utilization efficiency is achieved; />For electrical, gas input to the network.
2) Carbon dioxide emission reduction rate.
The carbon dioxide emission reduction rate refers to the ratio of the carbon dioxide emission amount reduced by using renewable energy sources such as solar energy, wind energy, geothermal energy and the like to the total carbon dioxide emission amount of a park. Mainly refers to carbon dioxide emission reduction caused by photovoltaic power generation, wind power generation and ground source heat pump heat supply.
Wherein C is b 、C′ b The carbon dioxide emission rates before and after the response, respectively.
3) Grid security.
Before the demand response, the standby of the system is only satisfied by a conventional unit; after the demand response, the interruptible load and the conventional unit jointly meet the standby demand, and the interruptible load cuts down the power consumption according to the demand so as to meet the standby capacity of the system.
In the method, in the process of the invention,spare capacity before and after demand response, respectively; r is R t The upper limit of the climbing rate of the unit is set; n (N) G The total number of the units; r is R G,i,t The spare capacity provided for the unit i at the time t; />Spare capacity provided for adjustable load.
Without loss of generality, the input-output model function is shown in a formula (31), the input x is the key factor set in the step 3, and the output is the value of the multi-energy equipment demand response value evaluation index:
y＝f(x),x∈K d (32)
wherein x is defined in space K d Input variables of (a); d is K d The collection contains the number of elements; y is the output response of the system, eta 1 ,η 2 ,η 3 ,...,The energy conversion efficiency and capacity of the multi-energy devices in the system are generally referred to; and (3) performing Sobol decomposition on the output response y of the system:
wherein f 0 As constant terms, the remaining sub-terms represent polynomials determined from 2 input variables, 3 input variables, and so on to d input variables; x is x i Representing the i-th input variable; x is x j Represents the j-th input variable; x is x d Representing the d-th input variable.
Global sensitivity index theory calculation:
the variance of the output response y=f (x) can be expressed by integration as:
the variances of all the steps in the Sobol decomposition are called as all the order variances, and the total variance of the output response y is decomposed into the sum of all the order variances:
where the variance can be expressed as:
wherein,represents the ith s Input variables- >Representing the decomposition of Sobol by i s The square of the sub-formula determined by the individual input variables. V (V) ij Representing the sub-variance, i.e. the second order variance, V, of the Sobol decomposition determined by the ith and jth input variables together 1,2,...d The sub-variances, i.e., d-order variances, determined by the 1,2,3, …, d input variables together in the Sobol decomposition are shown. d represents the set K d Contains the number of elements.
The following Sobol index is defined:
wherein,for second order global sensitivity, the ith is represented 1 And (i) 2 The effect of the interaction of the individual random variables on the variance of the output response; and so on->Represents the ith 1 ，i 2 ，…,i s The effect of the interaction of the individual random variables on the variance of the output response; to characterize a single input variable X i The influence of the randomness of (2) on the randomness of the output response, defining an input variable X based on the expectations and variances of the variables i The contribution to the output variance is noted as the global sensitivity index S Ti The method comprises the following steps:
wherein X is ～i Represents dividing X i A set of all variables except. Y denotes that the output response y=f (x) is regarded as a random variable.
The index simultaneously reflects the input random variable X i And the proportion of interactions with other variables in the variance of the function output response.
First-order global sensitivity index S i And an overall global sensitivity index S Ti Is involved in complex integral operations and is generally difficult to obtain analytical expressions, so that auxiliary calculations by means of MCS-based numerical integration are required, the expressions being as follows:
first a sample matrix of dimension N x 2d is generated, and the sample point values should obey the probability distribution of the input variables. Let the first d columns of the matrix constitute a blocking matrix a and the remaining d columns constitute a blocking matrix B.The matrix a is represented by replacing the column vector of the i-th column with the i-th column of the matrix B.
Auxiliary calculation of global sensitivity index (S) of all input random variables according to equations (40), (41) i ，S Ti ) I=1, …, d, together requiring a solution of N sim The output response of the original calculation model/function at sample number N (d+2). Because obtaining accurate computation results through the MCS generally requires a large number of samples N, when the sample size N is large enough, the sample function values of the random input variables will converge probability-wise to the current function expression values.
And calculating a probability distribution model to obtain a multi-dimensional comprehensive energy equipment response value quantification result. Based on the rolling optimization result, the distribution histogram of each evaluation index value is easy to obtain, the kernel density estimation, namely Kernel DensityEstimation, is used for deducing the distribution of the overall data based on limited samples, therefore, the result of the kernel density estimation is the probability density function estimation of the samples, and according to the probability density function of the estimation, some properties of the data distribution, such as the aggregation area of the data, can be obtained.
Considering each evaluation index value as one-dimensional data, there are d pieces of sample data: s is S T1 、S T2 、S T3 …S Td 。
Assuming that the cumulative distribution function of the sample data is F (S) and the probability density function is F (S), there are:
an empirical distribution function incorporating a cumulative distribution function:
using S in d observations Ti The ratio of the number of occurrences of alpha to d is used to approximate the probability; substituting the empirical distribution function into f (S i ) The method comprises the following steps:
according to this formula, in actual calculation, an h value must be given, which cannot be too large or too small, and which cannot satisfy the condition that h approaches 0, too few sample data points are used too small, and the error is large, so there are many studies on the selection of the h value, which is also called bandwidth in kernel density estimation. After determining the bandwidth, the expression for f (S) can be written:
wherein S is Ti Representing the ith sample data; s is S Tj The representation size is between S Ti -h and S Ti Sample data of +h.
After the f (S) expression is obtained, the distribution condition of each evaluation index is obtained, so that the multidimensional response value judgment of the comprehensive energy equipment is obtained.
On the other hand, a comprehensive energy equipment response value quantization system is characterized in that: the acquisition unit is used for acquiring cost data of equipment operation and establishing an energy hub optimization operation model of the comprehensive energy system taking comprehensive demand response into consideration; the solving unit adopts random sampling to simulate the distribution of the key parameters and solve the operation model; the training unit uses a PSO-BP neural network agent model to replace the comprehensive energy system energy hub optimization operation model considering the comprehensive demand response; the computing unit is used for computing the evaluation index of each demand response value by using the output result of the agent model, computing the global sensitivity index of the key influence factors of each equipment participating in the demand response to the evaluation index of the demand response value of the multi-energy equipment, and obtaining the quantitative result of the response value of the multi-dimensional comprehensive energy equipment according to the probability distribution model.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the various embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile memory may include read only memory, magnetic tape, floppy disk, flash memory, optical memory, high density embedded nonvolatile memory, resistive memory, magnetic memory, ferroelectric memory, phase change memory, graphene memory, and the like. Volatile memory can include random access memory, external cache memory, or the like. By way of illustration, and not limitation, RAM can take many forms, such as static random access memory or dynamic random access memory. The databases referred to in the various embodiments provided herein may include at least one of relational databases and non-relational databases. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like.
The processors referred to in the embodiments provided herein may be general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic units, quantum computing-based data processing logic units, etc., without being limited thereto.
Example 2
Referring to fig. 2-4, for one embodiment of the present invention, a method for quantifying the response value of an integrated energy device is provided, and in order to verify the beneficial effects of the present invention, scientific demonstration is performed through economic benefit calculation and simulation experiments.
By using the method of the invention, an example analysis is carried out by taking a multi-purpose commercial park in the Wu district of Changzhou city in Jiangsu province as an example.
(1) Basic data.
Constructing a comprehensive energy system architecture shown in fig. 2, wherein natural gas input into the energy hub is supplied to GT and GB, and a power supply bus absorbs power from a power grid and PV, CCHP, EES to supply basic power load, AC load and EES charging load; the heat supply bus absorbs the heat power from the CCHP and the GB and supplies the heat load; the cold supply bus absorbs cold power from the AC and LBR and supplies the building refrigeration load. The user, the CCHP, the electric energy storage and the AC receive the time-of-use electricity price and the IDR compensation price information through the park EH dispatching center to adjust the self energy consumption and the capacity behavior.
The price of the natural gas is 2.83 yuan/m 3, the electricity price is the time-sharing electricity price of common commercial electricity, namely 1.3782 yuan/(kW.h) in peak time periods (10:00-15:00, 18:00-21:00), 0.3658 yuan/(kW.h) in valley time periods (0:00-7:00, 23:00), 0.8595 yuan/(kW.h) in normal time periods (the rest of 1 day), and the IDR compensation price implemented in a garden is 0.9 yuan/(kW.h).
The global sensitivity index of each device to the above index is calculated by the proposed global sensitivity analysis method based on the SOBOL' method using the random efficiency of the energy conversion device and the energy storage device AC, LBR, GB, HS, ESS, HRB in the campus shown in fig. 2 as input variables, the daily operation cost minimum as target solution model, and the indexes such as cost, energy consumption satisfaction, energy consumption expenditure satisfaction, comprehensive satisfaction and comprehensive energy utilization efficiency as output variables. In order to improve the calculation efficiency of the global sensitivity index, the input variable dimension of the introduced PSO-BP model is 6 (AC, LBR, GB, HS, ESS, HRB random efficiency), the output variable dimension is 3 (minimum daily operation cost, comprehensive user satisfaction and comprehensive energy utilization efficiency), 10 hidden layers (the number of neurons is 3) are arranged, the population size is set to 20 by a particle swarm algorithm, and the population update times are 50. The training sample set of the PSO-BP model consists of 1000 groups of input-output samples, and the mean square error of the model finally obtained through 22 times of generation training is 8.69 multiplied by 10 < -7 >, so that the required precision requirement is met. When the global sensitivity index is calculated, the sampling times of the three output variables are calculated by adopting MCS to be 50000.
And obtaining random efficiency input samples of 6 devices with sample capacity of 50000 through MCS, calling a mature commercial solver CPLEX to solve an IESS economic optimization model based on GAMS, and analyzing the sensitivity of each input variable to different output results.
(2) Global sensitivity analysis results.
Minimum daily operating cost:
how to reduce the running costs is a primary issue for the campus operators that they should consider, so it is necessary to analyze the global sensitivity analysis of each input variable for daily running costs. AC. Specific values of the Total Sensitivity Coefficient (TSC) and the first order sensitivity coefficient (FSC) of LBR, GB, HS, ESS and HRB for the minimum daily operation cost are shown in table 1.
TABLE 1 FSC and TSC input variables versus minimum daily operating cost
| Apparatus and method for controlling the operation of a device | FSC | TSC |
| AC | 0.1267 | 0.1215 |
| LBR | 0.2975 | 0.3214 |
| GB | 0.0025 | 0.0027 |
| HS | 0.0257 | 0.0206 |
| ESS | 0.2695 | 0.2690 |
| HRB | 0.2766 | 0.2862 |
Among the 6 selected facilities, the minimum daily operational cost of the campus is most affected by the operational efficiency of the LBR, followed by HRB, followed by AC, all less affected by GB, HS, ESS. This is because the system cooling load is primarily supplied by the LBR in the CCHP system, AC assisted cooling, under the energy hub parameters set forth herein, and thus the campus operating costs are most sensitive to the efficiency variations of the LBR. Secondly, HRB is used as an important waste heat recovery device in CCHP, and the recovery efficiency of the HRB not only directly affects the heat energy supply, but also indirectly affects the cold energy supply through LBR, so that the HRB has higher contribution degree to the park operation cost. Next, AC is one of the direct cooling devices, whose COP will have a direct impact on the cooling energy supply, and although "multi-energy complementation" can be achieved by the LBR participating in the cooling, AC is still needed to participate in the energy supply during the peak period of cooling energy, so the COP variation of AC also has a certain impact on the cost. ESS acts as the only electrical energy storage in the campus, which can be used as a flexible resource to reduce costs by discharging during peak power and charging during off-peak power, thus significantly impacting costs. In the case of GB, in the integrated energy system provided herein, GB functions as a backup heat source, and only participates in supplying heat energy when CCHP is insufficient in heat supply, so that the influence on running cost is weak. As an energy storage device, HS is limited in its capacity and is not a primary energy supply device, while its impact on the minimum operational cost of the campus is minimal due to the "multi-energy complementary" effect in the network which further weakens its impact on the operational cost of the campus.
From table 1, it can be seen that the sum of FSCs of the 6 selected devices is 0.9829, very close to 1, for the minimum daily operating cost. Meanwhile, for 3 devices with larger influence degree, the TSC is slightly larger than the FSC, so that the influence of interaction among multiple devices on the minimum daily operation cost is small, and the probability distribution of the minimum daily operation cost of a park is mainly determined by the independent action of each input variable. Because FSC and TSC have very small phase difference, the invention mainly uses FSC as an index for identifying key factors of a system.
(2) The user synthesizes satisfaction.
When the user adjusts the energy consumption according to the energy price, the energy consumption experience level of the user and the satisfaction degree of the user should be considered. It is contemplated herein that the user aggregate satisfaction consists of user satisfaction and energy expenditure satisfaction.
TABLE 2 FSC and TSC for input variables to user Integrated satisfaction
| Apparatus and method for controlling the operation of a device | FSC | TSC |
| AC | 0.0561 | 0.0651 |
| LBR | 0.0529 | 0.0929 |
| GB | 0.0037 | 0.0093 |
| HS | 0.0050 | 0.0223 |
| ESS | 0.0006 | 0.0023 |
| HRB | 0.8329 | 0.8684 |
FSC and TSC for the case of integrated energy plant efficiency bias for the park user integrated satisfaction a=0.6, b=0.4 are shown in table 2. As can be seen from the table, the overall satisfaction of the campus subscribers is affected by the uncertainty of the deviation of the efficiency of the integrated energy facility, from large to small, HRB, LBR, AC, HS, GB and ESS, respectively. The comprehensive satisfaction of the user is mainly affected by the HRB to the greatest extent, and the TSC value is 0.8684.
Further, b=0.2 for a=0.8, respectively; the user satisfaction for both cases a=0.2, b=0.8, and the FSC of the efficiency deviation for each device is shown in table 3. Comparing the FSC of the comprehensive energy equipment efficiency deviation with the FSC of the a=0.6 and the b=0.4 cases and other cases, it is known that the comprehensive satisfaction degree of users with different compositions is affected by the equipment in the comprehensive energy system. The HRB is an important component device of the CCHP of the main multi-energy supply unit, and compared with other multi-energy complementary devices capable of participating in coupling energy supply, the HRB has a single energy supply mode, the supply cost cannot be complementarily changed through a plurality of energy supply modes, the user is sensitive to the energy supply price, the satisfaction degree of the user for participating in demand response can be greatly influenced by the change of the energy supply price, and the energy consumption expense of the user is further influenced, so that the influence of the HRB on the comprehensive satisfaction degree is the greatest.
TABLE 3 TSC of input variable to user Integrated satisfaction with different compositions
(3) Comprehensive energy utilization efficiency
Under the double-carbon target, a clean low-carbon energy system needs to be constructed, and comprehensive energy utilization efficiency indexes are introduced for the system. For this index, the FSC of the efficiency deviation of each device is shown in table 4.
TABLE 4 FSC and TSC input variable versus comprehensive energy utilization efficiency
| Apparatus and method for controlling the operation of a device | FSC | TSC |
| AC | 0.1893 | 0.1944 |
| LBR | 0.4333 | 0.4840 |
| GB | 0.0124 | 0.0245 |
| HS | 0.0430 | 0.0452 |
| ESS | 0.0108 | 0.0049 |
| HRB | 0.2877 | 0.2928 |
It is clear from the table that the comprehensive energy utilization efficiency is greatly influenced by LBR, HRB and AC and is less influenced by the efficiency change of other devices. The sum of the TSCs of LBR, HRB and AC is 0.9712, while the sum of the TSCs of GB, HS, ESS is only 0.07, which affects to a negligible extent compared to the former three.
Further, the following 3 scenarios were set to verify the validity of the Sobol' method-based FSC for identifying key factors of the system. Scene 1: the efficiency of the 6-class comprehensive energy equipment generates random deviation; scene 2: only the key equipment LBR, HRB and AC generate random deviation, and the other equipment respectively take the average value of probability distribution; scene 3: the critical devices LBR, HRB and AC average their probability distributions, whereas the non-critical device GB, HS, ESS generates random bias. For the above 3 cases, the cumulative probability distribution diagram of the comprehensive energy utilization efficiency is obtained as shown in fig. 4. It is known that when only important input variables (scenario 2) are considered, the result is substantially consistent with CDF taking all random input variables (scenario 1) into account; on the contrary, when only unimportant random input variables (scene 3) are considered, the CDF is very different from scene 1, and the comprehensive energy utilization efficiency change range of scene 1 and scene 2 is far larger than that of scene 3. Therefore, the FSC calculated based on the Sobol' method can effectively identify important input random variables influencing the probability distribution of the output variables of the system, thereby being beneficial to reducing the dimension of the input variables of the random problem of the comprehensive energy system and improving the calculation efficiency.
It should be noted that the above embodiments are only for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention may be modified or substituted without departing from the spirit and scope of the technical solution of the present invention, which is intended to be covered in the scope of the claims of the present invention.
Claims (10)
1. The method for quantifying the response value of the comprehensive energy equipment is characterized by comprising the following steps of:
acquiring cost data of equipment operation, and establishing an energy hub optimization operation model of a comprehensive energy system taking comprehensive demand response into consideration;
simulating the distribution of the key parameters by adopting random sampling, and solving an operation model;
using a PSO-BP neural network proxy model to replace the comprehensive energy system energy hub optimization operation model considering the comprehensive demand response;
calculating the evaluation index of each demand response value by using the output result of the agent model, calculating the global sensitivity index of key influence factors of each equipment participating in the demand response to the evaluation index of the demand response value of the multi-energy equipment, and obtaining the quantitative result of the response value of the multi-dimensional comprehensive energy equipment according to the probability distribution model of the global sensitivity index.
2. The method for quantifying the response value of an integrated energy device according to claim 1, wherein: the comprehensive energy system energy hub optimization operation model considering the comprehensive demand response comprises the steps of determining an objective function of the model:
min C＝min{1000·[C G +C P +C IDR ]}
wherein C is G Lambda is the cost of purchasing gas t G For the unit price of purchasing gas, the unit is Yuan/m 3; c (C) P The electricity purchasing cost is realized; for the electricity prices implemented by the power grid,the unit is Yuan/(kW.h) for electricity purchasing price; p (P) t PG Electric power is purchased for IES in MW; c (C) IDR Is IDR cost; lambda (lambda) t IDR The unit is meta/(kW.h) for IDR compensation unit price; />The unit is MW for the change in power consumption of the ith AC response IDR.
3. The method for quantifying the response value of an integrated energy device according to claim 2, wherein: determining constraints of the integrated energy system energy hub optimization run model that take into account the integrated demand response includes,
wherein P is c,t Representing the output of the CHP unit on the electric load supply side, W t Representing the output of the wind power electric load supply side, P pv,t Indicating the output of the photovoltaic load supply side, L e,t Represent DeltaL e,t Expressed, L h,t Represent DeltaL h,t Representation, H c,t Representing the output of the heat load supply side of the CHP unit, H GB,t Indicating the output of the heat load supply side of the gas turbine, H ET,t Representing the output of the electric boiler on the heat load supply side, P representing the system input, Representing the upper limit of the input quantity of the system,Prepresents the lower limit of the input quantity of the system, g (P c,t ，H ct ,，H o,t ) Less than or equal to 0 represents the operation constraint of the CHP unit, the electric boiler and the hot spring equipment, and h (delta L e,t ，ΔL h,t ) And less than or equal to 0 represents the constraint condition that the active response of electric power and heat is satisfied.
4. The method for quantifying the response value of an integrated energy device according to claim 3, wherein: modeling the key parameters of equipment in the model, simulating the key parameter distribution by adopting random sampling, and solving the model;
energy conversion equipment efficiency criticality modeling: various comprehensive energy devices exist in the comprehensive energy system, and under the operating condition of considering variable working conditions, the efficiency is subjected to distribution:
modeling the critical parameters of the equipment capacity: comprehensive energy systemThere are a plurality of comprehensive energy devices in the system, and the capacity criticality is uniformly described:
wherein,representing the maximum capacity of the r-th device, +.>The lower limit of the expression is S r The upper limit is->Is the electrical efficiency of the multi-functional device, +.>The distribution of the parameters μ, δ;
randomly sampling the critical parameters to generate a random number sequence with the size of N;
and after N groups of critical samples are obtained, substituting the N groups of critical samples into the comprehensive energy system energy hub optimization operation model considering the comprehensive demand response to solve.
5. The method for quantifying the response value of an integrated energy device according to claim 4, wherein: the PSO-BP neural network comprises the steps that the key parameters are used as input variables of training samples, and solving results are used as output variables of the training samples by substituting the input variables into the comprehensive energy system energy hub optimization operation model considering comprehensive demand response;
training a PSO-BP neural network agent model, wherein in the BP neural network, a PSO optimization algorithm optimizes the weight and bias of the neural network, and combines the PSO optimization algorithm to construct the PSO-BP neural network agent model;
predicting output variables by proxy model:
generating a large number of new input samples according to the probability distribution of the input variables; a large amount of new sample data is taken as input, thereby obtaining a prediction result.
6. The method for quantifying the response value of an integrated energy device according to claim 5, wherein: the value of the multi-energy equipment demand response value evaluation index is as follows:
y＝f(x),x∈K d
wherein x represents a space defined in K d Input variables of (a); d represents K d The collection contains the number of elements; y represents the output response of the system, eta 1 ,η 2 ,η 3 … represents the energy conversion efficiency of the multiple energy in the system,representing the capacity of the multi-energy device in the system;
And (3) performing Sobol decomposition on the output response y of the system:
wherein f 0 Represent constant term, f i Representing a polynomial determined by 1 input variable i, f ij Representing a polynomial determined by two input variables ij, f 1,...,d A polynomial representing a decision of d input variables; x is x i Representing the i-th input variable; x is x j Represents the j-th input variable; x is x d Represents the d-th input variable;
the global sensitivity index theoretical calculation includes,
calculating the variance of the output response y=f (x):
the variances of all the steps in the Sobol decomposition are called as all the order variances, and the total variance of the output response y is decomposed into the sum of all the order variances:
the s-order bias is expressed as integral:
wherein,represents the ith s A number of input variables; />Representing the decomposition of Sobol by i s Squaring the sub-formula determined by the individual input variables; v (V) ij The sub-variance determined by the ith input variable and the jth input variable in the Sobol decomposition is represented as a second-order variance; v (V) 1,2,...d The sub-variances determined by the 1,2,3, …, d input variables in the Sobol decomposition are d-order variances; d represents the set K d The number of the elements is included;
defining Sobol indexes of each order based on the calculation result of the variance:
wherein,for second order global sensitivity, the ith is represented 1 And (i) 2 Influence of the interaction of the random variables on the variance of the output response, and so on,/o>Represents the ith 1 ，i 2 ，…,i s The effect of the interaction of the individual random variables on the variance of the output response; defining an ith input variable X based on expected and variance calculations of the variables i The contribution to the output variance is noted as the global sensitivity index S Ti ：
Wherein X is ～i Represents dividing X i The set of all variables except for Y represents that the output response y=f (x) is considered as a random variable.
7. The method for quantifying the response value of an integrated energy device according to claim 6, wherein: calculating the probability distribution model, wherein the obtaining of the multi-dimensional comprehensive energy device response value quantification result comprises taking each evaluation index value as one-dimensional data, and d sample data are obtained: s is S T1 、S T2 、S T3 …S Td ；
Let the cumulative distribution function of the sample data be F (S), the probability density function be F (S):
wherein h represents width, S Ti-1 Representing the ith samplePrevious sample data of the present data;
an empirical distribution function incorporating a cumulative distribution function representing the sum of the discrete variables S Ti In other words, the sum of occurrence probabilities of all values equal to or less than α:
wherein d represents the number of sample data;
using S in d observations Ti The ratio of the number of occurrences of alpha to d is used to approximate the probability; substituting the empirical distribution function into f (S i ) The method comprises the following steps:
after determining the bandwidth of h, the expression of f (S):
wherein S is Ti Representing the ith sample data; s is S Tj The representation size is between S Ti -h and S Ti Sample data of +h;
after the f (S) expression is obtained, the distribution condition of each evaluation index is obtained, so that the multidimensional response value judgment of the comprehensive energy equipment is obtained.
8. A comprehensive energy device response value quantification system employing the method of any of claims 1-7, wherein:
the acquisition unit is used for acquiring cost data of equipment operation and establishing an energy hub optimization operation model of the comprehensive energy system taking comprehensive demand response into consideration;
the solving unit adopts random sampling to simulate the distribution of the key parameters and solve the operation model;
the training unit uses a PSO-BP neural network agent model to replace the comprehensive energy system energy hub optimization operation model considering the comprehensive demand response;
the computing unit is used for computing the evaluation index of each demand response value by using the output result of the agent model, computing the global sensitivity index of the key influence factors of each equipment participating in the demand response to the evaluation index of the demand response value of the multi-energy equipment, and obtaining the quantitative result of the response value of the multi-dimensional comprehensive energy equipment according to the probability distribution model.
9. A computer device, comprising: a memory and a processor; the memory stores a computer program characterized in that: the processor, when executing the computer program, implements the steps of the method of any one of claims 1 to 7.
10. A computer-readable storage medium having stored thereon a computer program, characterized by: the computer program implementing the steps of the method of any of claims 1 to 7 when executed by a processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311634448.4A CN117764438A (en) | 2023-12-01 | 2023-12-01 | Comprehensive energy equipment response value quantification method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311634448.4A CN117764438A (en) | 2023-12-01 | 2023-12-01 | Comprehensive energy equipment response value quantification method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117764438A true CN117764438A (en) | 2024-03-26 |
Family
ID=90324769
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311634448.4A Pending CN117764438A (en) | 2023-12-01 | 2023-12-01 | Comprehensive energy equipment response value quantification method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117764438A (en) |
-
2023
- 2023-12-01 CN CN202311634448.4A patent/CN117764438A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Data-driven distributionally robust scheduling of community integrated energy systems with uncertain renewable generations considering integrated demand response | |
| Wang et al. | Multi-objective planning of regional integrated energy system aiming at exergy efficiency and economy | |
| James et al. | Optimal V2G scheduling of electric vehicles and unit commitment using chemical reaction optimization | |
| CN105809349A (en) | Scheduling method considering incoming water correlation cascade hydropower stations | |
| Kong et al. | Refined peak shaving potential assessment and differentiated decision-making method for user load in virtual power plants | |
| CN113592133A (en) | Energy hub optimal configuration method and system | |
| Lasemi et al. | Stochastic multi-objective scheduling of a wind farm integrated with high-temperature heat and power storage in energy market | |
| Zhang et al. | Research on AGC performance during wind power ramping based on deep reinforcement learning | |
| Bagheri et al. | Stochastic optimization and scenario generation for peak load shaving in Smart District microgrid: sizing and operation | |
| CN116861776A (en) | Near real-time carbon emission calculation method based on electric-energy-carbon correlation model | |
| Zhang et al. | Deep reinforcement learning based bi-layer optimal scheduling for microgrid considering flexible load control | |
| CN104578160A (en) | Micro network energy control method | |
| Ersoz et al. | A stochastic evaluation of investments in combined cooling, heat, and power systems | |
| Kang et al. | Feed-forward active operation optimization for CCHP system considering thermal load forecasting | |
| CN115733178A (en) | Optical storage charging station capacity configuration method and system based on cost and risk multiple targets | |
| Peng et al. | Sequential coalition formation for wind-thermal combined bidding | |
| Wen et al. | ELCC-based capacity value estimation of combined wind-storage system using IPSO algorithm | |
| Li et al. | Optimal design and operation method of integrated energy system based on stochastic scenario iteration and energy storage attribute evaluation | |
| CN114418160A (en) | Park multi-energy system optimal scheduling method based on comprehensive evaluation system | |
| CN105514988A (en) | Micro power grid power supply planning scheme optimal selection method taking dynamic time-space features into consideration | |
| Cui et al. | Demand response method considering multiple types of flexible loads in industrial parks | |
| Wu et al. | Design of A battery energy management system for capacity charge reduction | |
| Su et al. | Optimal placement and capacity sizing of energy storage systems via NSGA-II in active distribution network | |
| CN117764438A (en) | Comprehensive energy equipment response value quantification method and system | |
| CN116823008A (en) | Park energy utilization efficiency evaluation method, system, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |