CN112686571A - Comprehensive intelligent energy optimization scheduling method and system based on dynamic adaptive modeling - Google Patents
Comprehensive intelligent energy optimization scheduling method and system based on dynamic adaptive modeling Download PDFInfo
- Publication number
- CN112686571A CN112686571A CN202110038209.7A CN202110038209A CN112686571A CN 112686571 A CN112686571 A CN 112686571A CN 202110038209 A CN202110038209 A CN 202110038209A CN 112686571 A CN112686571 A CN 112686571A
- Authority
- CN
- China
- Prior art keywords
- energy
- model
- equipment
- energy efficiency
- comprehensive
- 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.)
- Granted
Links
Images
Abstract
The invention provides a dynamic self-adaptive modeling-based comprehensive intelligent energy optimization scheduling method and system, which can solve the problems of inaccurate optimization scheduling result, incapability of tracking the optimal operation state of a system, insufficient economy of system operation and the like caused by deviation of equipment energy efficiency parameters and actual operation working condition energy efficiency in the conventional comprehensive energy scheduling model, and enable the comprehensive energy system to operate more economically and efficiently.
Description
Technical Field
The invention belongs to the technical field of energy optimization scheduling, and particularly relates to a comprehensive intelligent energy optimization scheduling method and system based on dynamic adaptive modeling.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
With the continuous development of economic society in China, the energy production and consumption modes are greatly changed, and the development of comprehensive energy services meeting the requirements of diversified energy production and consumption becomes a necessary trend of energy development. The comprehensive intelligent energy is a novel regional energy system which integrates and applies equipment such as distributed gas power generation, photovoltaic, distributed wind power, heat pumps, refrigerators and energy storage, is close to user management, promotes management modes by advanced information technology and intelligent technology, and organically coordinates and optimizes links such as distribution, conversion, storage and consumption of various energy sources. The system breaks through the traditional mode of independent planning, independent design and independent operation of different energy varieties, provides an integrated solution for regional comprehensive energy, and has the advantages of high comprehensive energy utilization efficiency, capability of promoting on-site consumption of renewable energy, capability of meeting diversified energy utilization requirements and the like.
The optimization scheduling of the comprehensive energy is a key means for realizing the safe, stable, efficient and intelligent operation of the comprehensive energy system, and is also a hotspot problem of academic research in the field at present. The energy flow model and the energy efficiency parameter of each device in the system are required for the optimization, scheduling and modeling of the comprehensive energy system, the operation energy efficiency of each device directly influences the energy consumption and the operation cost of the system, and the accurate energy flow model and the accurate operation energy efficiency parameter are important bases for achieving the optimization and scheduling of the system operation. Currently, most of the comprehensive energy systems are optimized and scheduled by an equipment energy efficiency value or an energy efficiency curve with fixed parameters under a design working condition. However, the operation energy efficiency parameters of each device are affected by conditions such as seasons, weather, device aging state, external load change, installation location and the like, and change continuously along with the change of operation conditions, and the deviation of the actual operation conditions and the design conditions causes the deviation of a scheduling model and an actual system, so that the optimal operation state of the system cannot be tracked by an optimized scheduling result.
Disclosure of Invention
The invention provides a comprehensive intelligent energy optimization scheduling method and system based on dynamic self-adaptive modeling, and aims to solve the problems that the optimal scheduling result is inaccurate, the optimal operation state of a system cannot be tracked, the system operation is not economic enough and the like due to deviation of equipment energy efficiency parameters and the energy efficiency of the actual operation working condition in the conventional comprehensive energy scheduling model, so that the comprehensive energy system is more economic and efficient to operate.
According to some embodiments, the invention adopts the following technical scheme:
a comprehensive intelligent energy optimization scheduling method based on dynamic adaptive modeling comprises the following steps:
step 1: forecasting cold, heat and electric loads based on weather, air temperature, load historical data and weather forecast information to obtain a load forecasting curve of a dispatching time period;
step 2: forecasting distributed photovoltaic, distributed wind power and solar hot water hourly output in the system based on weather, temperature, output historical data and weather forecast information to obtain a wind and light output forecasting curve in a dispatching period;
step 3: constructing a self-adaptive energy efficiency model and an energy flow model of each device or subsystem in the system based on the current and historical operating data of the device or the system;
step 4: establishing an optimized dispatching model of the comprehensive energy system according to the structure of the comprehensive energy system and the operation parameters of each device, wherein a device self-adaptive energy efficiency model and an energy flow model are used as constraint conditions of the optimized dispatching model;
step 5: and solving the comprehensive energy system optimization scheduling model by using an optimization solving algorithm to obtain a system optimal scheduling scheme.
As an alternative embodiment, the specific process of constructing the adaptive energy efficiency model and the energy flow model of each device or subsystem in the system includes:
(1) determining equipment energy efficiency parameters and factors influencing equipment energy efficiency;
(2) selecting a self-adaptive learning sample based on clustering;
(3) and constructing an equipment self-adaptive energy efficiency model and an energy flow model.
As a further limited embodiment, for the energy conversion device, the determining and describing of the device energy efficiency parameters and the factors affecting the device energy efficiency specifically include: the fluence model is represented as:
Qout(t)＝ηeq(t)·Qin(t)，
in the formula Qout(t),Qin(t) represents the output and input energy of the device during the time period t, ηeq(t) represents the energy conversion efficiency of the equipment, i.e. the energy efficiency parameter, and the energy efficiency value of the equipment changes along with the internal and external conditions and the operating condition, so the energy efficiency model of the equipment is represented as:
ηeq(t)＝feq[θ1(t),θ2(t),…,θK(t)]，
wherein, theta1,θ2,…,θKTaking K factor values influencing the energy efficiency of the equipment as model input; f. ofeqAn adaptive model structure is constructed based on data mining.
As an alternative embodiment, for the energy conversion device, the specific process of determining the device energy efficiency parameter and the factor affecting the device energy efficiency includes: for air source heat pumps, ground source heat pumps and chiller plants, Qout(t),Qin(t),ηeq(t) respectively representing the output heat/cold quantity, the consumed electricity quantity and the COP heating/cold coefficient in the time period t;
the energy efficiency model of the air source heat pump is expressed as:
COP(t)＝f[PLR(t),Ti(t),To(t),Te(t),F(t)]，
wherein PLR (T), Ti(t),To(t),Te(t), F (t) respectively represent the load factor, the water inlet temperature, the water outlet temperature, the ambient air temperature and the wind speed of the heat pump at the time t, and for the air source heat pump capable of heating and refrigerating, the heating mode and the air source heat pumpThe cooling modes are modeled separately.
The energy efficiency model of the ground source heat pump is expressed as follows:
COP(t)＝f[PLR(t),Teo(t),Tei(t),Tgi(t),Tgo(t),Te(t)]，
wherein PLR (T), Teo(t),Tei(t),Tgi(t),Tgo(t),Te(t) respectively showing the load factor, the tail end side outlet water temperature, the tail end side inlet water temperature, the ground source side outlet water temperature and the ambient temperature of the ground source heat pump at the moment t; and the heating mode and the cooling mode are respectively modeled.
The energy efficiency model of the water chilling unit is expressed as:
COP(t)＝f[PLR(t),Teo(t),Tci(t),Me(t),Mc(t)]，
wherein PLR (T), Teo(t),Tci(t),Me(t),McAnd (t) respectively representing the cooling load rate, the chilled water supply temperature, the cooling water inlet temperature, the chilled water flow and the cooling water flow of the unit at the time t, and respectively modeling different working conditions of the double-working-condition water chilling unit.
As a further limited implementation, for the energy storage device, the determining and describing of the device energy efficiency parameters and the factors affecting the device energy efficiency specifically include: the fluence model is represented as:
Qst(t＝Qst(t-1)·(1-μloss(t))+Qst_in(t)·ηin(t)-Qst_out(t)/ηout(t)
in the formula, Qst(t) represents the energy stored in the energy storage device at time t, Qst_in(t)，Qst_out(t) represents the energy storage input and the energy release output energy of the energy storage device in the time period from the t-1 moment to the t moment respectively, and mulossIndicating the energy loss rate, eta, of the energy storage devicein，ηoutRespectively representing the energy storage efficiency and the energy release efficiency of the energy storage device. The energy efficiency model of the energy storage device may be expressed as
[μloss(t),ηin(t),ηout(t)]＝f[Sc(t-1),Qst_in(t),Qst_out(t),Te(t)]，
Can also be expressed in segments as
Energy storage working condition: [ mu ] ofloss(t),ηin(t)]＝fx[Sc(t-1),Qst_in(t),Te(t)]
Energy release working conditions are as follows: [ mu ] ofloss(t),ηout(t)]＝fs[Sc(t-1),Qst_out(t),Te(t)]
And (3) idle working condition: [ mu ] ofloss(t)]＝f0[Sc(t-1),Te(t)]
In the formula (I), the compound is shown in the specification,
representing the state of energy storage at time T-1, Te(t) is the ambient air temperature.
As an alternative embodiment, for the energy conversion device, the specific process of selecting the adaptive learning sample based on the load gradient clustering includes:
(1) clustering is carried out on historical data in a certain time range of a period to be scheduled according to load gradient, namely, a load interval is equally divided into R subintervals according to the upper limit and the lower limit of the load of equipment operation, and samples with equipment load rates in the same subinterval form a cluster;
(2) randomly sampling in each clustering subsample, performing secondary filtering on the clustering subsample based on the corresponding air temperature value to form n/R samples, and finally forming a training sample set containing the n samples;
(3) the sample data is normalized.
As an alternative embodiment, for the energy storage device, the specific process of selecting the adaptive learning sample based on the gradient cluster of the energy storage state includes:
(1) clustering is carried out on historical data in a certain time range of a period to be scheduled according to the energy storage state gradient, namely, the energy storage state interval is equally divided into R subintervals according to the upper and lower limits of the energy storage state of equipment operation, and samples of the energy storage state of the equipment in the same subinterval form a cluster;
(2) randomly sampling in each clustering subsample, performing secondary filtering on the clustering subsample based on the corresponding air temperature value to form n/R samples, and finally forming a training sample set containing the n samples;
(3) the sample data is normalized.
As an alternative embodiment, the specific process of constructing the device adaptive energy efficiency model based on the PSO-LSSVM includes:
(1) setting the particle swarm scale, the maximum iteration times, the learning factor and the inertia coefficient range, and initializing the primary particle swarm;
(2) constructing an LSSVM model for each particle, calculating each body fitness, and recording an individual extreme value and a global extreme value;
(3) updating the speed and position of each particle;
(4) calculating the fitness of each particle of the current generation, and updating the individual extreme value and the global extreme value;
(5) and (4) judging whether a termination condition is met, if so, outputting a corresponding LSSVM energy efficiency model according to the optimal value of the particles, and otherwise, turning to the step (3).
As an alternative embodiment, the integrated energy system optimization scheduling model includes an objective function and a constraint condition for system optimization, and the objective function is that the total operating cost of the integrated energy system is the lowest or the operating energy efficiency of the system is the highest.
As an alternative embodiment, the constraint conditions of the integrated energy system optimization scheduling model include an energy balance constraint, an equipment output ramp constraint, an energy storage constraint, a gateway constraint and a system flow network constraint.
As an alternative implementation, the specific process of solving the comprehensive energy system optimization scheduling model by using the genetic evolution algorithm to obtain the system optimal scheduling scheme includes:
(1) determining scheduling decision variables, including the starting and stopping states of each device, the load rate, the energy storage and release states of each energy storage device and the energy storage and release energy at each moment in a scheduling interval, and coding a scheduling scheme consisting of all the decision variables;
(2) initializing evolution algorithm parameters and generating an initial population, wherein the algorithm parameters comprise selection probability, variation probability and maximum iteration times;
(3) inputting the load rate and the energy storage quantity parameters of each individual in the population corresponding to the scheduling scheme at each moment and the predicted values of the external environment energy efficiency influence factors at the corresponding moment into the constructed equipment self-adaptive energy efficiency model to obtain the equipment energy efficiency at each moment, calculating the fitness of each individual in the population based on the optimized scheduling model, wherein the fitness function adopts the sum of a target function and a constraint unsatisfied penalty term;
(4) storing the current optimal individual, and executing selection, crossing or variation operation to generate a new generation of population;
(5) judging whether a termination condition is reached, and if so, outputting an optimal solution; if not, returning to the step (3).
An integrated intelligent energy optimization scheduling system based on dynamic adaptive modeling comprises:
the forecasting module is configured to forecast cold, heat and electric loads based on weather, air temperature, load historical data and weather forecast information to obtain a dispatching time period load forecasting curve, and forecast distributed photovoltaic, distributed wind power and solar hot water hourly output in the system based on the weather, air temperature, output historical data and the weather forecast information to obtain a dispatching time period wind and light output forecasting curve;
the energy flow model building module is configured to build an adaptive energy efficiency model and an energy flow model of each device or subsystem in the system based on current and historical operating data of the device or system;
the scheduling model building module is configured to build an optimized scheduling model of the comprehensive energy system according to the structure of the comprehensive energy system and the operation parameters of each device;
and the solving module is configured to solve the comprehensive energy system optimization scheduling model by using an optimization solving algorithm to obtain a system optimal scheduling scheme.
A computer readable storage medium, having stored therein a plurality of instructions adapted to be loaded by a processor of a terminal device and to execute the steps of the method for integrated intelligent energy optimized scheduling based on dynamic adaptive modeling.
A terminal device comprising a processor and a computer readable storage medium, the processor being configured to implement instructions; the computer readable storage medium is used for storing a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the steps of the comprehensive intelligent energy optimization scheduling method based on the dynamic adaptive modeling.
Compared with the prior art, the invention has the beneficial effects that:
the method is suitable for operation optimization scheduling of the comprehensive intelligent energy system, and can effectively reduce the deviation between the energy efficiency parameter of the equipment in the comprehensive energy optimization scheduling model and the energy efficiency of the actual operation working condition by performing dynamic self-adaptive energy efficiency modeling on each equipment based on the operation data, so that the scheduling result can better track the optimal operation state of the actual system, thereby improving the comprehensive utilization efficiency of energy, effectively reducing the operation cost and improving the economy of the comprehensive energy system.
In order to make the aforementioned and other objects, features and advantages of the invention more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
Fig. 1 is a schematic flow chart of the present embodiment.
The specific implementation mode is as follows:
the present invention will be further described with reference to the following examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
An optimal scheduling method based on dynamic self-adaptive energy efficiency modeling aims to solve the problems that in an existing comprehensive energy scheduling model, equipment energy efficiency parameters and actual operation working condition energy efficiency have deviation, so that an optimal scheduling result is inaccurate, the optimal operation state of a system cannot be tracked, the system is not economical enough, and the like, and the comprehensive energy system is more economical and efficient to operate.
The characteristics of the comprehensive intelligent energy system include: the energy supply system is provided with one or more energy storage devices of a gas internal combustion engine, a gas boiler or an electric boiler, a water chilling unit, an air source heat pump, a sewage source heat pump, distributed photovoltaic, distributed wind power, solar hot water, cold accumulation, heat accumulation and electric power storage, and can automatically optimize and switch operation modes according to system operation states, user side load conditions, environment and energy price information and the like in different periods such as cold supply seasons, hot supply seasons, transition seasons and the like, optimize corresponding equipment combination and load distribution, meet the requirements of various energy sources such as energy side cold, heat, electricity, water and the like, and maximize economic, energy efficiency and environmental protection benefits.
Specifically, as shown in fig. 1, the method comprises the following steps:
1. predicting cold, heat and electric loads based on information such as weather, temperature, load historical data, weather forecast and the like to obtain a load prediction curve of a scheduling period;
2. forecasting distributed photovoltaic, distributed wind power, solar hot water and the like in the system time by time based on information such as weather, temperature, output historical data, weather forecast and the like to obtain a wind and light output forecasting curve in a dispatching period;
3. constructing a self-adaptive energy efficiency model and an energy flow model of each device or subsystem in the system; for energy conversion devices, the energy flow model may be expressed as
Qout(t)＝ηeq(t)·Qin(t)，
In the formula Qout(t),Qin(t) indicates that the device is in the t periodOutput, input energy ofeq(t) represents the energy conversion efficiency of the equipment, i.e., the energy efficiency parameter, since the energy efficiency value of the equipment varies with the internal and external conditions and the operation conditions, the energy efficiency model of the equipment can be represented as
ηeq(t)＝feq[θ1(t),θ2(t),…,θK(t)]，
Wherein, theta1,θ2,…,θKTaking K factor values influencing the energy efficiency of the equipment as model input; f. ofeqThe self-adaptive model structure constructed based on data mining can be realized by system identification modeling methods such as a neural network and a support vector machine.
The concrete steps are as follows:
and 3.1, determining equipment energy efficiency parameters and factors influencing equipment energy efficiency.
For air source heat pump, ground source heat pump, water chilling unit and other equipment, Qout(t),Qin(t),ηeq(t) represents the output heat/cold amount, the consumed electricity amount, and the COP heating/cold coefficient of the time period t, respectively. The air source heat pump energy efficiency model can be expressed as
COP(t)＝f[PLR(t),Ti(t),To(t),Te(t),F(t)]，
Wherein PLR (T), Ti(t),To(t),TeAnd (t), F (t) respectively represent the load rate, the water inlet temperature, the water outlet temperature, the ambient air temperature and the wind speed of the heat pump at the moment t, and a heating mode and a cooling mode are respectively modeled for the air source heat pump capable of heating and cooling.
The energy efficiency model of the ground source heat pump can be expressed as
COP(t)＝f[PLR(t),Teo(t),Tei(t),Tgi(t),Tgo(t),Te(t)]，
Wherein PLR (T), Teo(t),Tei(t),Tgi(t),Tgo(t),Te(t) respectively showing the load factor, the tail end side outlet water temperature, the tail end side inlet water temperature, the ground source side outlet water temperature and the ambient air temperature of the ground source heat pump at the time t. And the heating mode and the cooling mode are respectively modeled.
The chiller energy efficiency model can be expressed as
COP(t)＝f[PLR(t),Teo(t),Tci(t),Me(t),Mc(t)]，
Wherein PLR (T), Teo(t),Tci(t),Me(t),McAnd (t) respectively representing the cooling load rate, the chilled water supply temperature, the cooling water inlet temperature, the chilled water flow and the cooling water flow of the unit at the time t, and respectively modeling different working conditions of the double-working-condition water chilling unit.
3.2 selecting adaptive learning samples based on load gradient clustering
(1) Clustering according to load gradient from historical data within the time range from 1 day to D days before a period to be scheduled, namely, equally dividing a load interval into R subintervals according to the upper and lower load limits of equipment operation, and forming a cluster by samples with equipment load rates in the same subinterval;
(2) randomly sampling in each cluster subsample, performing secondary filtering based on corresponding air temperature values to form n/R samples, and finally forming a training sample set containing n samples
The sampling secondary filtering method comprises the following steps: the air temperature change interval in the period to be scheduled is recorded as
If the sample corresponds to the air temperature
Then the process is passed; if it is
Then according to
Is rejected to pass; if it is
Then according to
Is rejected.
(3) According to the formula
Sample data is normalized, where K is 1,2, …, K,
respectively corresponding to the maximum value and the minimum value of the corresponding items of the sample data.
3.3 constructing an equipment self-adaptive energy efficiency model based on the PSO-LSSVM.
(1) Setting the particle swarm size m and the maximum iteration number
Learning factor c1,c2And range of inertia coefficients [ w ]min,wmax]Initializing the first generation particle swarm (gamma)l,σl),l＝1,2,…,m；
(2) For each particle (. gamma.)l,σl) Constructing an LSSVM model according to the following formula, calculating the fitness of each individual, and recording an individual extreme value plbestAnd a global extremum pgbest；
Fitness calculation formula:
wherein x isiCorresponding to in the ith sample
yiCorresponding to in the ith sample
(3) The velocity v of each particle is updated as followslAnd position plWherein, in the step (A),
pl＝[γl,σl]；
(4) calculating the fitness of each particle of the current generation according to the formula in the step (2), and updating the individual extreme value plbestAnd a global extremum pgbest；
(5) And (4) judging whether a termination condition is met, if so, outputting a corresponding LSSVM energy efficiency model according to the optimal value of (gamma, sigma), and otherwise, turning to the step (3).
For energy storage devices such as storage batteries and energy storage tanks, the energy flow model can be expressed as
Qst(t)＝Qst(t-1)·(1-μloss(t))+Qst_in(t)ηin(t)-Qst_out(t)/ηout(t)
In the formula, Qst(t) represents the energy stored in the energy storage device at time t, Qst_in(t)，Qst_out(t) represents the energy storage input and the energy release output energy of the energy storage device in the time period from the t-1 moment to the t moment respectively, and mulossIndicating the energy loss rate, eta, of the energy storage devicein，ηoutRespectively representing the energy storage efficiency and the energy release efficiency of the energy storage device. The energy efficiency model of the energy storage device may be expressed as
[μloss(t),ηin(t),ηout(t)]＝fpSc(t-1),Qst_in(t),Qst_out(t),Te(t)]，
Can also be expressed in segments as
Energy storage working condition: [ mu ] ofloss(t),ηin(t)]＝fx[Sc(t-1),Qst_in(t),Te(t)]
Energy release working conditions are as follows: [ mu ] ofloss(t),ηout(t)]＝fs[Sc(t-1),Qst_out(t),Te(t)]
And (3) idle working condition: [ mu ] ofloss(t)]＝f0[Sc(t-1),Te(t)]
In the formula (I), the compound is shown in the specification,
represents the state of energy storage (state of charge SOC for the battery), T, at time T-1e(t) is the ambient air temperature. The energy efficiency model of the energy storage device can also be realized by system identification modeling methods such as a neural network and a support vector machine, and a multi-input multi-output system identification method is required to be adopted unlike the modeling of equipment such as a heat pump.
Specific examples are as follows: modeling based on multi-output LSSVM:
constructing a self-adaptive learning sample by the same method as 3.2, and only clustering according to the load gradient, namely clustering according to the energy storage state gradient;
constructing an equipment self-adaptive energy efficiency model based on the PSO-LSSVM, wherein the method is the same as 3.3, and the LSSVM model in the step (2) needs to be changed into a multi-output model as follows:
αi＝[αi1,αi2,…,αir]
fitness calculation formula:
wherein r is the number of outputs, xiCorresponding to [ Sc (t-1), Q in the ith samplest_in(t),Qst_out(t),Te(t)]，yi＝[y11,y12,…,y1r]Corresponding to [ mu ] in the ith sampleloss(t),ηin(t),ηout(t)]。
And the dynamic energy efficiency modeling of equipment such as a gas internal combustion engine, a gas boiler and the like refers to the steps.
4. Establishing an optimized dispatching model of the comprehensive energy system according to the structure of the comprehensive energy system, the operation parameters of each device and the like;
the comprehensive energy system optimization scheduling model comprises an objective function and a constraint condition of system optimization, wherein the objective function is that the total running cost of the comprehensive energy system is lowest or the running energy efficiency of the system is highest, and the total running cost is taken as an example and is expressed as follows:
min fc＝Cgrid+Cfuel+Com，
wherein f iscRepresents the total cost of system operation, CgridRepresents the cost of purchasing electricity from outside, CfuelRepresents the fuel cost, ComRepresenting the equipment operation and maintenance cost;
respectively corresponding to the electricity purchasing price, the outsourcing electric quantity, the electricity selling price and the selling electric quantity in the time period t,
the price of the gas in the time period t and the amount of the natural gas consumed by the gas equipment i are respectively; xit、
For the on-off state and output power of device i during time period t,
the device start-stop maintenance cost coefficient and the power operation maintenance cost coefficient are respectively.
The constraint conditions comprise energy balance constraint, equipment output climbing constraint, energy storage constraint, gateway constraint, system process network constraint and the like.
The energy balance constraint is:
in the formula, Meo、Mei、MesRespectively, a collection of power generating, consuming and storing devices, M, within the systemco、Mci、McsAre respectively a systemSet of internal cold production, consumption and storage devices, Mho、Mhi、MhsRespectively a set of heat generating, heat consuming and heat storing equipment in the system,
the power required by the user end for electricity, cold and heat is t;
power is lost for the energy of electricity, cold and heat in the system;
the equipment output constraints are:
in the formula (I), the compound is shown in the specification,
for the minimum and maximum values of the operating output of the cooling device i,
the minimum value and the maximum value of the operation output of the heating equipment i,
the minimum and maximum values of the output are operated for the power supply device i.
The equipment output climbing constraint is as follows:
in the formula (I), the compound is shown in the specification,
for the lower limit and the upper limit of the variation of the operating output of the power supply equipment i in unit time (from time t to time t + 1), the output is increased to be positive, and the output is reduced to be negative;
the lower limit and the upper limit of the operating output of the heating device i changing in unit time,
the lower limit and the upper limit of the variation of the output force of the cooling device i in unit time are set.
The energy storage constraint is:
in the formula, Xit、X′itRespectively representing the energy release state and the energy storage state of the energy storage device i in the time t;
the maximum energy storage capacity of the energy storage device i.
The gateway constraints are:
in the formula (I), the compound is shown in the specification,
the capacity of the transformer gateway of the external power grid is obtained.
The plant energy flow models in the system flow network constraints employ the energy flow models generated in section 3.
5. Solving the model by adopting an improved genetic evolution algorithm to obtain an optimal scheduling scheme of the system;
(1) determining scheduling decision variables, including the starting and stopping states of each device, the load rate, the energy storage and release states of each energy storage device and the energy storage and release energy at each moment in a scheduling interval, and coding a scheduling scheme consisting of all the decision variables by adopting decimal floating point numbers;
(2) initializing evolution algorithm parameters and generating an initial population, wherein the algorithm parameters comprise selection probability, variation probability, maximum iteration times and the like;
(3) inputting parameters such as load rate and energy storage of each individual in the population corresponding to the scheduling scheme at each moment and predicted values of energy efficiency influence factors such as external environment at the corresponding moment into the equipment self-adaptive energy efficiency model constructed in the step3 to obtain the equipment energy efficiency at each moment, and then calculating the fitness of each individual in the population based on the optimized scheduling model, wherein the fitness function adopts the sum of a target function and a constraint unsatisfied penalty term;
(4) storing the current optimal individual, and executing selection, crossing or variation operation to generate a new generation of population;
(5) judging whether a termination condition is reached, and if so, outputting an optimal solution; if not, return to (3).
The following product examples are also provided:
an integrated intelligent energy optimization scheduling system based on dynamic adaptive modeling comprises:
the forecasting module is configured to forecast cold, heat and electric loads based on weather, air temperature, load historical data and weather forecast information to obtain a dispatching time period load forecasting curve, and forecast distributed photovoltaic, distributed wind power and solar hot water hourly output in the system based on the weather, air temperature, output historical data and the weather forecast information to obtain a dispatching time period wind and light output forecasting curve;
the energy flow model building module is configured to build an adaptive energy efficiency model and an energy flow model of each device or subsystem in the system based on current and historical operating data of the device or system;
the scheduling model building module is configured to build an optimized scheduling model of the comprehensive energy system according to the structure of the comprehensive energy system and the operation parameters of each device;
and the solving module is configured to solve the comprehensive energy system optimization scheduling model by using an optimization solving algorithm to obtain a system optimal scheduling scheme.
A computer readable storage medium, having stored therein a plurality of instructions adapted to be loaded by a processor of a terminal device and to execute the steps of the method for integrated intelligent energy optimized scheduling based on dynamic adaptive modeling.
A terminal device comprising a processor and a computer readable storage medium, the processor being configured to implement instructions; the computer readable storage medium is used for storing a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the steps of the comprehensive intelligent energy optimization scheduling method based on the dynamic adaptive modeling.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Although the present invention has been described with reference to the specific embodiments, it should be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention.
Claims (10)
1. A comprehensive intelligent energy optimization scheduling method based on dynamic adaptive modeling is characterized by comprising the following steps: the method comprises the following steps:
forecasting cold, heat and electric loads based on weather, air temperature, load historical data and weather forecast information to obtain a load forecasting curve of a dispatching time period;
forecasting distributed photovoltaic, distributed wind power and solar hot water hourly output in the system based on weather, temperature, output historical data and weather forecast information to obtain a wind and light output forecasting curve in a dispatching period;
constructing a self-adaptive energy efficiency model and an energy flow model of each device or subsystem in the system based on the current and historical operating data of the device or the system;
establishing an optimized dispatching model of the comprehensive energy system according to the structure of the comprehensive energy system and the operation parameters of each device, wherein a device self-adaptive energy efficiency model and an energy flow model are used as constraint conditions of the optimized dispatching model;
and solving the comprehensive energy system optimization scheduling model by using an optimization solving algorithm to obtain a system optimal scheduling scheme.
2. The method according to claim 1, wherein the comprehensive intelligent energy optimization scheduling method based on dynamic adaptive modeling comprises: the specific process for constructing the self-adaptive energy efficiency model and the energy flow model of each device or subsystem in the system comprises the following steps:
(1) determining equipment energy efficiency parameters and factors influencing equipment energy efficiency;
(2) selecting a self-adaptive learning sample based on clustering;
(3) and constructing an equipment self-adaptive energy efficiency model and an energy flow model.
3. The method according to claim 2, wherein the comprehensive intelligent energy optimization scheduling method based on dynamic adaptive modeling comprises: the determination and description of the energy conversion equipment, the equipment energy efficiency parameters and the factors influencing the equipment energy efficiency specifically comprise the following steps: the fluence model is represented as:
Qout(t)＝ηeq(t)·Qin(t)，
in the formula Qout(t)，Qin(t) represents the output and input energy of the device during the time period t, ηeq(t) represents the energy conversion efficiency of the equipment, i.e. the energy efficiency parameter, and the energy efficiency value of the equipment changes along with the internal and external conditions and the operating condition, so the energy efficiency model of the equipment is represented as:
ηeq(t)＝feq[θ1(t)，θ2(t)，…，θK(t)]，
wherein, theta1，θ2，…，θKTaking K factor values influencing the energy efficiency of the equipment as model input; f. ofeqAn adaptive model structure is constructed based on data mining.
4. The method according to claim 3, wherein the comprehensive intelligent energy optimization scheduling method based on dynamic adaptive modeling comprises: for air source heat pumps, ground source heat pumps and chiller plants, Qout(t)，Qin(t)，ηeq(t) respectively representing the output heat/cold quantity, the consumed electricity quantity and the COP heating/cold coefficient in the time period t;
the energy efficiency model of the air source heat pump is expressed as:
COP(t)＝f[PLR(t)，Ti(t)，To(t)，Te(t)，F(t)]，
wherein PLR (T), Ti(t)，To(t)，Te(t), F (t) respectively representing the load rate, the water inlet temperature, the water outlet temperature, the ambient air temperature and the wind speed of the heat pump at the moment t, and respectively modeling a heating mode and a cooling mode for an air source heat pump capable of heating and cooling;
or, the ground source heat pump energy efficiency model is expressed as:
COP(t)＝f[PLR(t)，Teo(t)，Tei(t)，Tgi(t)，Tgo(t)，Te(t)]，
wherein PLR (T), Teo(t)，Tei(t)，Tgi(t)，Tgo(t)，Te(t) respectively showing the load factor, the tail end side outlet water temperature, the tail end side inlet water temperature, the ground source side outlet water temperature and the ambient temperature of the ground source heat pump at the moment t; respectively modeling a heating mode and a cooling mode;
or, the energy efficiency model of the water chilling unit is expressed as:
COP(t)＝f[PLR(t)，Teo(t)，Tci(t)，Me(t)，Mc(t)]，
wherein PLR (T), Teo(t)，Tci(t)，Me(t)，McAnd (t) respectively representing the cooling load rate, the chilled water supply temperature, the cooling water inlet temperature, the chilled water flow and the cooling water flow of the unit at the time t, and respectively modeling different working conditions of the double-working-condition water chilling unit.
5. The method according to claim 2, wherein the comprehensive intelligent energy optimization scheduling method based on dynamic adaptive modeling comprises: for the energy storage device, the determination and description of the device energy efficiency parameters and the factors influencing the device energy efficiency specifically include: the fluence model is represented as:
Qst(t)＝Qst(t-1)·(1-μloss(t))+Qst_in(t)·ηin(t)-Qst_out(t)/ηout(t)
in the formula, Qst(t) represents time tEnergy storage of energy storage devices, Qst_in(t)，Qst_out(t) represents the energy storage input and the energy release output energy of the energy storage device in the time period from the t-1 moment to the t moment respectively, and mulossIndicating the energy loss rate, eta, of the energy storage devicein，ηoutRespectively representing the energy storage efficiency and the energy release efficiency of the energy storage device;
the energy efficiency model of the energy storage device is expressed as:
[μloss(t)，ηin(t)，ηout(t)]＝f[Sc(t-1)，Qst_in(t)，Qst_out(t)，Te(t)]，
can also be expressed in segments as
Energy storage working condition: [ mu ] ofloss(t)，ηin(t)]＝fx[Sc(t-1)，Qst_in(t)，Te(t)]
Energy release working conditions are as follows: [ mu ] ofloss(t)，ηout(t)]＝fs[Sc(t-1)，Qst_out(t)，Te(t)]
And (3) idle working condition: [ mu ] ofloss(t)]＝f0[Sc(t-1)，Te(t)]
In the formula (I), the compound is shown in the specification,
representing the state of energy storage at time T-1, Te(t) is the ambient air temperature.
6. The method according to claim 2, wherein the comprehensive intelligent energy optimization scheduling method based on dynamic adaptive modeling comprises: for energy conversion equipment, the specific process of selecting the self-adaptive learning sample based on load gradient clustering comprises the following steps:
(1) clustering is carried out on historical data in a certain time range of a period to be scheduled according to load gradient, namely, a load interval is equally divided into R subintervals according to the upper limit and the lower limit of the load of equipment operation, and samples with equipment load rates in the same subinterval form a cluster;
(2) randomly sampling in each clustering subsample, performing secondary filtering on the clustering subsample based on the corresponding air temperature value to form n/R samples, and finally forming a training sample set containing the n samples;
(3) the sample data is normalized.
7. The method according to claim 2, wherein the comprehensive intelligent energy optimization scheduling method based on dynamic adaptive modeling comprises: for the energy storage device, the specific process of selecting the self-adaptive learning sample based on the gradient clustering of the energy storage state comprises the following steps:
(1) clustering is carried out on historical data in a certain time range of a period to be scheduled according to the energy storage state gradient, namely, the energy storage state interval is equally divided into R subintervals according to the upper and lower limits of the energy storage state of equipment operation, and samples of the energy storage state of the equipment in the same subinterval form a cluster;
(2) randomly sampling in each clustering subsample, performing secondary filtering on the clustering subsample based on the corresponding air temperature value to form n/R samples, and finally forming a training sample set containing the n samples;
(3) the sample data is normalized.
8. The method according to claim 1, wherein the comprehensive intelligent energy optimization scheduling method based on dynamic adaptive modeling comprises: the specific process for constructing the self-adaptive energy efficiency model of the equipment based on the PSO-LSSVM comprises the following steps:
(1) setting the particle swarm scale, the maximum iteration times, the learning factor and the inertia coefficient range, and initializing the primary particle swarm;
(2) constructing an LSSVM model for each particle, calculating each body fitness, and recording an individual extreme value and a global extreme value;
(3) updating the speed and position of each particle;
(4) updating individual extreme values and global extreme values according to the fitness of each particle of the current generation calculated in the step (2);
(5) and (4) judging whether a termination condition is met, if so, outputting a corresponding LSSVM energy efficiency model according to the optimal value, otherwise, turning to the step (3).
9. The method according to claim 1, wherein the comprehensive intelligent energy optimization scheduling method based on dynamic adaptive modeling comprises: the comprehensive energy system optimization scheduling model comprises an objective function and a constraint condition of system optimization, wherein the objective function is that the total running cost of the comprehensive energy system is lowest or the running energy efficiency of the system is highest;
the constraint conditions of the comprehensive energy system optimization scheduling model comprise energy balance constraint, equipment output climbing constraint, energy storage constraint, gateway constraint and system flow network constraint;
or, the specific process of solving the comprehensive energy system optimization scheduling model to obtain the system optimal scheduling scheme includes:
(1) determining scheduling decision variables, including the starting and stopping states of each device, the load rate, the energy storage and release states of each energy storage device and the energy storage and release energy at each moment in a scheduling interval, and coding a scheduling scheme consisting of all the decision variables;
(2) initializing evolution algorithm parameters and generating an initial population, wherein the algorithm parameters comprise selection probability, variation probability and maximum iteration times;
(3) inputting the load rate and the energy storage quantity parameters of each individual in the population corresponding to the scheduling scheme at each moment and the predicted values of the external environment energy efficiency influence factors at the corresponding moment into the constructed equipment self-adaptive energy efficiency model to obtain the equipment energy efficiency at each moment, calculating the fitness of each individual in the population based on the optimized scheduling model, wherein the fitness function adopts the sum of a target function and a constraint unsatisfied penalty term;
(4) storing the current optimal individual, and executing selection, crossing or variation operation to generate a new generation of population;
(5) judging whether a termination condition is reached, and if so, outputting an optimal solution; if not, returning to the step (3).
10. A comprehensive intelligent energy optimization scheduling system based on dynamic adaptive modeling is characterized in that: the method comprises the following steps:
the forecasting module is configured to forecast cold, heat and electric loads based on weather, air temperature, load historical data and weather forecast information to obtain a dispatching time period load forecasting curve, and forecast distributed photovoltaic, distributed wind power and solar hot water hourly output in the system based on the weather, air temperature, output historical data and the weather forecast information to obtain a dispatching time period wind and light output forecasting curve;
the energy flow model building module is configured to build an adaptive energy efficiency model and an energy flow model of each device or subsystem in the system based on current and historical operating data of the device or system;
the scheduling model building module is configured to build an optimized scheduling model of the comprehensive energy system according to the structure of the comprehensive energy system and the operation parameters of each device;
and the solving module is configured to solve the comprehensive energy system optimization scheduling model by adopting an optimization solving algorithm to obtain a system optimal scheduling scheme.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110038209.7A CN112686571B (en) | 2021-01-12 | 2021-01-12 | Comprehensive intelligent energy optimization scheduling method and system based on dynamic adaptive modeling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110038209.7A CN112686571B (en) | 2021-01-12 | 2021-01-12 | Comprehensive intelligent energy optimization scheduling method and system based on dynamic adaptive modeling |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112686571A true CN112686571A (en) | 2021-04-20 |
| CN112686571B CN112686571B (en) | 2023-01-20 |
Family
ID=75457522
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110038209.7A Active CN112686571B (en) | 2021-01-12 | 2021-01-12 | Comprehensive intelligent energy optimization scheduling method and system based on dynamic adaptive modeling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112686571B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113469412A (en) * | 2021-06-02 | 2021-10-01 | 国核电力规划设计研究院有限公司 | Real-time operation strategy optimization method and system for comprehensive energy system |
| CN113726006A (en) * | 2021-07-29 | 2021-11-30 | 湖南五凌电力科技有限公司 | Hydroelectric production analysis decision support system |
| CN115581056A (en) * | 2022-11-09 | 2023-01-06 | 宁波亮控信息科技有限公司 | Energy-saving prediction control method and system suitable for water cooling system of data center |
| CN117824224A (en) * | 2024-03-04 | 2024-04-05 | 无锡冠亚恒温制冷技术有限公司 | Control method of low-temperature system for enhancing stability |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106485089A (en) * | 2016-10-21 | 2017-03-08 | 福州大学 | The interval parameter acquisition methods of harmonic wave user's typical condition |
| CN106952183A (en) * | 2017-04-24 | 2017-07-14 | 武汉大学 | A kind of short-term load forecasting method based on particle group optimizing least square method supporting vector machine |
| CN107506865A (en) * | 2017-08-30 | 2017-12-22 | 华中科技大学 | A kind of load forecasting method and system based on LSSVM optimizations |
| CN110619110A (en) * | 2019-09-18 | 2019-12-27 | 南京师范大学 | Coordinated operation optimization method for comprehensive energy system with heat pump |
| CN112001523A (en) * | 2020-07-09 | 2020-11-27 | 山东电力工程咨询院有限公司 | Comprehensive energy multi-scale optimization scheduling control method and system considering multiple energy storages |
-
2021
- 2021-01-12 CN CN202110038209.7A patent/CN112686571B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106485089A (en) * | 2016-10-21 | 2017-03-08 | 福州大学 | The interval parameter acquisition methods of harmonic wave user's typical condition |
| CN106952183A (en) * | 2017-04-24 | 2017-07-14 | 武汉大学 | A kind of short-term load forecasting method based on particle group optimizing least square method supporting vector machine |
| CN107506865A (en) * | 2017-08-30 | 2017-12-22 | 华中科技大学 | A kind of load forecasting method and system based on LSSVM optimizations |
| CN110619110A (en) * | 2019-09-18 | 2019-12-27 | 南京师范大学 | Coordinated operation optimization method for comprehensive energy system with heat pump |
| CN112001523A (en) * | 2020-07-09 | 2020-11-27 | 山东电力工程咨询院有限公司 | Comprehensive energy multi-scale optimization scheduling control method and system considering multiple energy storages |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113469412A (en) * | 2021-06-02 | 2021-10-01 | 国核电力规划设计研究院有限公司 | Real-time operation strategy optimization method and system for comprehensive energy system |
| CN113469412B (en) * | 2021-06-02 | 2024-04-09 | 国核电力规划设计研究院有限公司 | Real-time operation strategy optimization method and system for comprehensive energy system |
| CN113726006A (en) * | 2021-07-29 | 2021-11-30 | 湖南五凌电力科技有限公司 | Hydroelectric production analysis decision support system |
| CN113726006B (en) * | 2021-07-29 | 2022-08-19 | 湖南五凌电力科技有限公司 | Hydroelectric production analysis decision support system |
| CN115581056A (en) * | 2022-11-09 | 2023-01-06 | 宁波亮控信息科技有限公司 | Energy-saving prediction control method and system suitable for water cooling system of data center |
| CN115581056B (en) * | 2022-11-09 | 2023-03-21 | 宁波亮控信息科技有限公司 | Energy-saving prediction control method and system suitable for water cooling system of data center |
| CN117824224A (en) * | 2024-03-04 | 2024-04-05 | 无锡冠亚恒温制冷技术有限公司 | Control method of low-temperature system for enhancing stability |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112686571B (en) | 2023-01-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112686571B (en) | Comprehensive intelligent energy optimization scheduling method and system based on dynamic adaptive modeling | |
| Ebrahimi et al. | Unit commitment problem solution using shuffled frog leaping algorithm | |
| CN107134810B (en) | Independent micro-energy-grid energy storage system optimal configuration solving method | |
| CN108875992B (en) | Virtual power plant day-ahead optimization scheduling method considering demand response | |
| CN111555355B (en) | Scheduling strategy and optimization method for water-light-storage combined power generation | |
| CN107039975B (en) | Energy management method for distributed energy system | |
| CN111555281B (en) | Method and device for simulating flexible resource allocation of power system | |
| CN105870976B (en) | A kind of low-carbon dispatching method and device based on energy environment efficiency | |
| CN109711614A (en) | A kind of the dynamic optimization progress control method and system of distributed busbar protection | |
| CN111753431B (en) | Computing method and computing equipment for optimal configuration in comprehensive energy system | |
| CN112598195A (en) | Building type comprehensive energy system operation optimization method and device and terminal equipment | |
| CN115796393A (en) | Energy network management optimization method, system and storage medium based on multi-energy interaction | |
| CN115147245B (en) | Virtual power plant optimal scheduling method for industrial load participating in peak shaving auxiliary service | |
| CN116247648A (en) | Deep reinforcement learning method for micro-grid energy scheduling under consideration of source load uncertainty | |
| CN112001523A (en) | Comprehensive energy multi-scale optimization scheduling control method and system considering multiple energy storages | |
| CN113052389A (en) | Distributed photovoltaic power station ultra-short-term power prediction method and system based on multiple tasks | |
| CN115115193A (en) | Low-carbon analysis and optimization planning method for industrial park | |
| CN115062904A (en) | Digital water supply pipe network scheduling method and system | |
| CN111126675A (en) | Multi-energy complementary microgrid system optimization method | |
| CN112966444B (en) | Intelligent energy optimization method and device for building multi-energy system | |
| CN112308411B (en) | Comprehensive energy station random planning method and system based on dynamic carbon trading model | |
| CN112132379B (en) | Economic-considered new energy cross-region consumption evaluation method and storage medium | |
| CN113887809A (en) | Power distribution network supply and demand balance method, system, medium and computing equipment under double-carbon target | |
| CN117207803A (en) | Electric automobile intelligent charging strategy selection method based on economic dispatch | |
| CN116205458A (en) | Method for establishing comprehensive energy system optimization scheduling model of hydroelectric hydrogen production by considering uncertainty |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |