CN111626003A - Heating system heat load layered prediction method, storage medium and prediction equipment - Google Patents

Heating system heat load layered prediction method, storage medium and prediction equipment Download PDF

Info

Publication number
CN111626003A
CN111626003A CN202010427163.3A CN202010427163A CN111626003A CN 111626003 A CN111626003 A CN 111626003A CN 202010427163 A CN202010427163 A CN 202010427163A CN 111626003 A CN111626003 A CN 111626003A
Authority
CN
China
Prior art keywords
pipe network
prediction
equation
load
temperature
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
Application number
CN202010427163.3A
Other languages
Chinese (zh)
Other versions
CN111626003B (en
Inventor
赵安军
席江涛
于军琪
任延欢
冉彤
张万虎
周昕玮
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xian University of Architecture and Technology
Original Assignee
Xian University of Architecture and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Xian University of Architecture and Technology filed Critical Xian University of Architecture and Technology
Priority to CN202010427163.3A priority Critical patent/CN111626003B/en
Publication of CN111626003A publication Critical patent/CN111626003A/en
Application granted granted Critical
Publication of CN111626003B publication Critical patent/CN111626003B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/28—Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24D—DOMESTIC- OR SPACE-HEATING SYSTEMS, e.g. CENTRAL HEATING SYSTEMS; DOMESTIC HOT-WATER SUPPLY SYSTEMS; ELEMENTS OR COMPONENTS THEREFOR
    • F24D19/00—Details
    • F24D19/10—Arrangement or mounting of control or safety devices
    • F24D19/1006—Arrangement or mounting of control or safety devices for water heating systems
    • F24D19/1009—Arrangement or mounting of control or safety devices for water heating systems for central heating
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/08—Thermal analysis or thermal optimisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Economics (AREA)
  • General Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • General Engineering & Computer Science (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Health & Medical Sciences (AREA)
  • Combustion & Propulsion (AREA)
  • Mathematical Analysis (AREA)
  • Operations Research (AREA)
  • Thermal Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Mechanical Engineering (AREA)
  • Game Theory and Decision Science (AREA)
  • Algebra (AREA)
  • Computing Systems (AREA)
  • Fluid Mechanics (AREA)
  • Quality & Reliability (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Physics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • Development Economics (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a heat load hierarchical prediction method, a storage medium and prediction equipment for a heating system, wherein a high-rise building group using a regional heating system is divided into a building layer, a distribution pipe network layer and a main pipe network layer; then, predicting the thermal load of the building layer by using a linear model based on historical data and meteorological data of the heat exchange station, providing an error measurement method, and respectively establishing corresponding physical models of a distribution pipe network and a main pipe network based on a graph theory and a hydrodynamics three-large equation; then, the prediction result of the building layer is used as the Dirichlet boundary condition of the three major equations of the distribution pipe network layer, the mass equation and the momentum equation of the distribution pipe network are solved by using a SIMPLE algorithm with the added relaxation factors to obtain flow, and the temperature is obtained through solving; and finally, in the same way, the prediction result of the distribution pipe network layer is used as the Dirichlet boundary condition of the three major equations of the main pipe network layer, the flow and the temperature of the main pipe network layer are obtained by solving, and then the heat loads of the three layers are accurately predicted.

Description

Heating system heat load layered prediction method, storage medium and prediction equipment
Technical Field
The invention belongs to the technical field of building heat load prediction, and particularly relates to a heat load hierarchical prediction method, a storage medium and prediction equipment for a heating system.
Background
In a district heating system of a high-rise building group, municipal hot water is generally supplied and then primary heat exchange is performed, and since a large number of high-rise buildings are arranged in an area in charge of the primary heat exchange station, secondary heat exchange is generally performed under each high-rise building.
One of the problems commonly existing in the existing regional heating system is that the heating temperature is too high or too low, which causes energy waste and high dissatisfaction of residents on the indoor environment, so that a method for carrying out layered prediction on heat load in the regional heating system of a high-rise building group is provided, thereby realizing the purpose of accurately predicting the future hourly heat load, laying scientific basis and solid foundation for the control and optimization of the regional heating system, and having positive effect on the maintenance and expansion of a later-period system.
Disclosure of Invention
The technical problem to be solved by the present invention is to provide a method, a storage medium, and a prediction device for predicting the thermal load of a heating system in a high-rise building area hierarchically, which solve the problem in the prior art that the thermal load of the system cannot be extracted accurately due to high-dimensional, nonlinear, and dynamic reasons of load data.
The invention adopts the following technical scheme:
a thermal load hierarchical prediction method for a heating system of a high-rise building area comprises the following steps:
s1, performing regional division on the high-rise building group by taking each heat exchange station as a mark, and dividing the system into a building layer, a distribution pipe network layer and a main pipe network layer by taking the primary heat exchange station, the heat supply pipe network nodes in the region and the secondary heat exchange station as boundary points;
s2, determining model input variables having large influence on heat load, and determining output variables of the model as temperature and flow according to requirements;
s3, defining the prediction model required in the step S2 as a linear model in a matrix form, predicting an output matrix Y according to an input matrix X, completing prediction of the thermal load of the building layer, and verifying the prediction result;
s4, building a physical model of the distribution pipe network and the main pipe network by using three large equations of graph theory knowledge and fluid dynamics;
s5, taking the building layer thermal load predicted in the step S3 as Dirichlet boundary conditions of three major equations corresponding to the distribution pipe network layer in the step S4, solving a mass equation and a momentum equation by using a semi-implicit method of a pressure coupling equation set to obtain flow, and then solving the energy equation to obtain temperature to complete thermal load prediction of the distribution pipe network layer;
s6, taking the building layer heat load predicted in the step S5 as Dirichlet boundary conditions of three major equations corresponding to the main pipe network layer in the step S4, solving a mass equation and a momentum equation by using a semi-implicit method of a pressure coupling equation set to obtain flow, and then solving the energy equation to obtain temperature to complete heat load prediction of the distribution pipe network layer.
Specifically, in step S2, a correlation analysis method is used to discuss the relationship between the influence factors and the heat supply load, the influence factors related to the heat supply load are used as variables x, the heat supply load is used as a variable y, the correlation coefficient r value of each influence factor and the heat supply load is calculated, the correlation coefficient is used to determine the correlation size, and by calculating the significance level value, when the significance level value is greater than or equal to the corresponding number in the table, the relationship is significant, otherwise, the relationship is not significant; the determined model input variables includeAverage temperature T of previous daym,d-1(ii) a Lowest temperature T of the previous daymin,d-1(ii) a Maximum temperature T of previous daymax,d-1And the average temperature T of the daym,d(ii) a The measured quantity comprises a flow G of the primary side of the heat exchange station; water temperature T at the primary side inlet1(ii) a Water temperature T at the primary side outlet2(ii) a Water temperature T at secondary side inlet4(ii) a Water temperature T at secondary side outlet3And the ambient temperature Tenv。
Further, the correlation coefficient r is specifically:
Figure BDA0002499127760000031
wherein the content of the first and second substances,
Figure BDA0002499127760000032
is the covariance of x and y,xis the standard deviation of x and is,yis the standard deviation of y, x is the first variable, y is the second variable,
Figure BDA0002499127760000033
is the average number of x, and is,
Figure BDA0002499127760000034
is the average of y.
Further, the significance level values are specifically:
Figure BDA0002499127760000035
wherein n-2 is a degree of freedom and r is a correlation coefficient.
Specifically, in step S3, the accuracy of the prediction model, the degree of fitting γ, and the average absolute percentage MAPE error of the evaluation index of the prediction model are used to measure the model accuracy2The fitting degree of the prediction result and the actual value is represented as follows:
Figure BDA0002499127760000036
Figure BDA0002499127760000037
wherein, yiI time actual thermal load, yiThe' predicted load when it is i, and n is the number of data.
Specifically, in step S3, the training data test data set is preprocessed by combining the standard score and mean interpolation method, and the input variable data set is set to have a size N, and the processing procedure is as follows:
s301, calculating a mean e (i) and a variance v (i) of the data in the data set N for each input variable respectively as follows:
Figure BDA0002499127760000038
Figure BDA0002499127760000041
s302, defining standard score deviation rate ξ (i, n), and setting maximum allowable deviation rate ξmax；
Figure BDA0002499127760000042
S303, when | ξ (i, n) | is equal to or more than ξmaxAnd eliminating abnormal data, and correcting by adopting a mean interpolation method as follows:
Figure BDA0002499127760000043
the magnitude of each parameter of the input data is different, and the input data is normalized by dispersion normalization as follows:
Figure BDA0002499127760000044
performing inverse normalization processing on the output value of the model to obtain an actual predicted value, specifically:
ysi＝ymin+oi(ymax-ymin)
wherein x isiIs the original value of the sample; x is the number ofminIs the original sample minimum; x is the number ofmaxIs the original sample maximum; x is the number ofi' is a normalized processed value; y ismaxThe maximum output value of the original sample; y isminIs the minimum output value of the original sample; oiA predicted value for the model output; y issiAnd outputting a reduction value for the model.
Specifically, in step S4, according to the one-dimensional space method, each pipe of the pipe network is regarded as a branch starting from one node and ending at another node; the incidence matrix A describes the topology of the pipe network by representing the connection between nodes and branches, the number of rows of the matrix is the same as the number of nodes, the number of columns is the same as the number of branches, and if a branch j enters or exits nodes i and o, a general element AijAnd when the temperature is equal to 1 or-1, establishing a mass conservation equation at a node and a momentum conservation equation at a branch, solving a coupling equation through a SIMPLE algorithm to obtain the flow G, establishing an energy conservation equation at the node and solving the differential equation to obtain the temperature T.
Specifically, in step S5 and step S6, the pressure P and the flow rate G are calculated as follows:
P＝P′+α′Pcorr
G＝G′+α′Gcorr
wherein P 'and G' are respectively an initial pressure field and a velocity field of the algorithm, Pcorr、GcorrRespectively, the correction amount of the current calculation process, α' is a relaxation factor.
A computer readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by a computing device, cause the computing device to perform any of the methods of claims 1-8.
A prediction device comprising:
one or more processors, memory, and one or more programs stored in the memory and configured for execution by the one or more processors, the one or more programs including instructions for performing any of the methods of claims 1-8.
Compared with the prior art, the invention has at least the following beneficial effects:
the invention relates to a heat load layered prediction method for a heating system of a high-rise building area, which can realize accurate prediction of heat load by combining a primary heat exchange station, a transmission pipe network, a secondary heat exchange station and a building system in a heating area, and can be used for modeling a linear model for a building layer, thereby laying a good mathematical foundation for later heat load prediction work for accurate modeling of knowledge of graph theory for the transmission pipe network and heat flow mechanics.
Further, in step S2, the heat load of the district heating system is affected by various factors such as outdoor weather parameters, building types, and heat consumption characteristics, and in the prediction, a common way is to select an independent variable having a large influence on a dependent variable by using a qualitative analysis method, and if more influencing factors are adopted, the calculation amount and complexity are increased, and the calculation accuracy and efficiency are affected. In the step, correlation coefficient is adopted to carry out correlation analysis on variables possibly influencing heat load, and dependent variables with large influence on heat load are selected as input of the model.
Furthermore, step S3 defines the prediction model as a linear model, which can reduce the computational complexity of the prediction of the thermal load on the building floor, and at the same time, the input and the output are both in a matrix form, so as to achieve the purpose of multiple inputs and multiple outputs to comprehensively and accurately predict the thermal load.
Further, the data collected at the heat exchange station and the data obtained by measuring the outdoor meteorological conditions in step S3 may have problems such as equipment damage or various human factors causing vacancy and inaccurate measurement, so that the data preprocessing is performed on the data, thereby facilitating the smooth operation of the subsequent calculation work.
Further, in step S4, the graph theory knowledge and the thermal flow mechanics are used to accurately describe the pipe network model, so that the problems of pressure change, heat loss and the like generated during transmission in the pipe network can be accurately calculated.
Further, the results obtained by the previous layer prediction are used as the dirichlet boundary condition of the present layer in steps S5 and S6, firstly to ensure that the equation has a solution in the calculation region, and then the thermal load systems of the respective layers of the region can be combined to realize accurate thermal load prediction.
In summary, the invention can accurately predict the heat load from the combination of the primary heat exchange station, the transmission pipe network, the secondary heat exchange station and the building system in a heating area, and solve the problem that the heat load of the system cannot be accurately extracted due to high-dimensional, nonlinear and dynamic reasons of load data in the prior art.
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
Drawings
FIG. 1 is a schematic diagram of a hierarchical prediction method;
FIG. 2 is a schematic diagram of the division of a high-rise building group into hierarchical regions;
FIG. 3 is a flow chart of solving a thermodynamic equation of heat flow;
FIG. 4 is a flow chart of the SIMPLE algorithm;
FIG. 5 is a diagram of the results of a prediction of the thermal load at a heating system building level for a region;
fig. 6 is a diagram showing the result of prediction of the thermal load of each floor of a heating system for a certain area.
Detailed Description
The invention provides a heat load layered prediction method of a high-rise building regional heating system, which comprehensively and accurately predicts the heat load from a building layer and a pipe network layer so as to improve the prediction precision, firstly, the high-rise building group using the regional heating system is divided into regions, and the whole regional heating system is divided into three layers by mainly taking a heat exchange station and a pipe network node as a boundary point, wherein the three layers are respectively the building layer, a distribution pipe network layer and a master pipe network layer; the thermal load on the building level is then modeled using a linear model Y ═ γ based on historical data and meteorological data for the heat exchange stations0X+γ1Predicting the distribution network and the main network, providing a method for measuring errors, and respectively establishing the correspondence of the distribution network and the main network based on three major equations of graph theory and fluid dynamicsThe method comprises the steps of obtaining a prediction result of a building layer, using the prediction result of the building layer as a Dirichlet boundary condition of three major equations of a distribution pipe network layer, solving a mass equation and a momentum equation of the distribution pipe network by using a SIMPLE algorithm with added relaxation factors α' to obtain flow G, solving the energy equation to obtain temperature T, and finally using the prediction result of the distribution pipe network layer as the Dirichlet boundary condition of the three major equations of a main pipe network layer to obtain the flow G and the temperature T of the main pipe network layer by solving, so that the heat loads of the three layers are accurately predicted.
Referring to fig. 1, the present invention provides a method for predicting thermal load of a heating system in a high-rise building area hierarchically, which comprises the following steps:
s1, performing regional division on a high-rise building group in a certain region by taking each heat exchange station as a mark, and mainly highlighting a first-stage heat exchange station connected with municipal hot water, a main pipe network and a distribution pipe network between the first-stage heat exchange station and a second-stage heat exchange station, connecting the distribution pipe network and the second-stage heat exchange stations of each high-rise building, thereby dividing the whole system into three levels;
the method comprises the steps that firstly, regional division is carried out on high-rise building groups in a certain region by taking each heat exchange station as a node, a pipe network part for municipal hot water supply is not considered, most of the high-rise building groups adopt a regional heating method, after the municipal hot water reaches the region, primary heat exchange is carried out firstly, then secondary heat exchange is carried out under each high-rise building, so that the whole system is divided into three layers by taking the primary heat exchange stations, the heat supply pipe network nodes in the region and the secondary heat exchange stations as demarcation points, namely a building layer (comprising a secondary heat exchange station), a distribution pipe network layer and a main pipe network layer (comprising a primary heat exchange station), and the final region division result schematic diagram is shown in figure 2.
The concrete simplification and abstraction of the pipe network are subject to the following principles:
in order to facilitate the expression and analysis by using graphs and data, the actual pipe network system must be simplified and abstracted firstly, and the simplified and abstracted pipe network mathematical model should be capable of fully embodying the topological structure and hydraulic characteristics among the pipe sections of the pipe network.
The simplification of the pipe network mainly omits a part of pipe network accessory equipment which is not important relative to the main pipeline, and the basic principle of the simplification is a macroscopic equivalence principle and a deviation minimization principle. The main measures are as follows:
(1) deleting the secondary pipeline and keeping the primary pipeline;
(2) the fully-opened valve can be deleted, and the fully-closed valve is disconnected;
(3) the pipes are different or the pipelines are connected in parallel, and the same pipe or a single pipeline can be equivalent by adopting a hydraulic equivalent principle;
(4) splitting a large system into a plurality of small systems as much as possible;
(5) removing ancillary facilities with little impact on global hydraulic characteristics;
(6) multiple identical fixtures at the same location are merged.
The simplified pipe network needs further abstraction, so that the pipe network becomes a pipe network model only comprising two types of elements of pipe sections and nodes. In the abstracted pipe network model, pipe sections and nodes are correlated, two ends of one pipe section are two nodes of the pipe section, and the two nodes are communicated through the pipe sections. The attribute characteristics of the nodes and the pipe sections comprise three aspects of construction attributes, topological attributes and hydraulic attributes. The structural attributes are used as the basis of both the topological structure attributes and the hydraulic attributes, the topological attributes represent the mutual correlation of the pipe sections and the nodes, and the hydraulic attributes represent the hydraulic characteristics of the pipe sections and the nodes in the pipe network.
S2, determining a model input variable which has a large influence on the heat load through correlation analysis, and determining the output variables of the model as temperature T and flow G according to the requirement;
the relationship between the influence factors and the heating load is discussed by using a correlation analysis method, and a correlation coefficient calculation formula is as follows:
Figure BDA0002499127760000091
wherein the content of the first and second substances,
Figure BDA0002499127760000092
is the covariance of x and y,xis the standard deviation of x，yIs the standard deviation of y, x is the first variable, y is the second variable,
Figure BDA0002499127760000093
is the average number of x, and is,
Figure BDA0002499127760000094
is the average of y.
The correlation coefficient properties are as follows:
when r 1, x and y are fully correlated;
when the | r | is more than 1, the x and the y are positively correlated, and when r <0, the x and the y are negatively correlated;
when | r | ═ 0, it means that there is no correlation between x and y;
when r < 0.3, it is called weak correlation;
when the absolute r is less than 0.5 and less than 0.3, the correlation is called low correlation;
when 0.5 ≦ r | < 0.8, it is said to be significantly correlated;
when 0.8 ≦ r | < 1, it is said to be highly correlated;
the influence factors related to the heat supply amount are respectively used as variables x, the heat supply amount is used as a variable y, the correlation coefficient r value of each influence factor and the heat supply amount is calculated, and the correlation coefficient is used for determining the magnitude of the correlation between the influence factors and the heat supply amount.
By calculating the significance level value, when the significance level value is greater than or equal to the corresponding number in the table, the relationship is significant, otherwise, the relationship is not significant, and the common confidence coefficient is 95%.
The significance level was calculated as follows:
Figure BDA0002499127760000095
wherein, n-2 is a degree of freedom, which is a significance level.
The determined model input variables include the average temperature T of the previous daym,d-1(ii) a Lowest temperature T of the previous daymin,d-1(ii) a Maximum temperature T of previous daymax,d-1And the average temperature T of the daym,d；
In the fitting process to the curve, the quantities measured include the following:
flow G at the primary side of the heat exchange station; water temperature T at the primary side inlet1(ii) a Water temperature T at the primary side outlet2(ii) a Water temperature T at secondary side inlet4(ii) a Water temperature T at secondary side outlet3And the ambient temperature Tenv。
The vector form of the input parameters is:
X＝[1 Tm,dTmin,d-1Tmax,d-1Tm,d-1]
the vector form of the output parameters is:
Y＝[G T1T2T3T4]。
s3, defining a linear model in matrix form for the prediction model required in step S2: y ═ gamma0X+γ1For representing the mapping of input and output, wherein X, Y represents the input matrix and the output matrix, respectively, and γ0、γ1And respectively representing a coefficient matrix and a constant matrix, determining the model through a large amount of historical data, and after the model is determined, under the condition that X is known, completing the prediction of Y, thereby completing the prediction of the thermal load of the building layer and verifying the prediction result.
Constructing a linear model in a matrix form, training the linear model by using a large amount of historical data to obtain a constant term, calculating corresponding heat load under the condition of known input variables after the model is successfully trained, predicting the accuracy degree of an evaluation index average absolute percentage MAPE error measurement model of the model, and predicting the fitting degree gamma of the model2And (3) representing the degree of fitting between the prediction result and the actual value, and specifically calculating as follows:
Figure BDA0002499127760000101
Figure BDA0002499127760000102
wherein, yiI time actual thermal load, yi' represents the predicted load at i, and n is the number of data.
In addition, the accuracy of the training data is the basic guarantee of the model prediction accuracy, and if abnormal values or loss of the data occur, the training result of the model can be directly influenced; the method is characterized in that a standard fraction and mean interpolation method is combined to preprocess test data excitation of training data, the size of an input variable data set is set to be N, and the processing process is as follows:
s301, respectively calculating a mean value E (i) and a variance V (i) of the data in the data set N when i is input into each input variable;
Figure BDA0002499127760000111
Figure BDA0002499127760000112
s302, defining standard score deviation rate ξ (i, n), and setting maximum allowable deviation rate ξmax；
Figure BDA0002499127760000113
S303, when | ξ (i, n) | is equal to or more than ξmaxAnd eliminating abnormal data, and correcting by adopting a mean interpolation method, wherein the mean interpolation method comprises the following steps:
Figure BDA0002499127760000114
the magnitude of each parameter of the input data is different, and the input data is normalized by dispersion standardization.
Figure BDA0002499127760000115
Performing inverse normalization processing on the output value of the model to obtain an actual predicted value, specifically:
ysi＝ymin+oi(ymax-ymin)
wherein x isiIs the original value of the sample; x is the number ofminIs the original sample minimum; x is the number ofmaxIs the original sample maximum; x is the number ofi' is a normalized processed value; y ismaxThe maximum output value of the original sample; y isminIs the minimum output value of the original sample; oiA predicted value for the model output; y issiAnd outputting a reduction value for the model.
S4, building a physical model for the distribution pipe network and the main pipe network by using three equations (mass conservation, momentum conservation and energy conservation) of graph theory knowledge and fluid dynamics, wherein the mass conservation equation is A.G + G ext0, and the conservation of momentum equation G Y AT·P+Y·ΔPpumpsThe energy conservation equation is
Figure BDA0002499127760000122
The establishment of a physical model of the pipe network is accomplished by using the knowledge of graph theory and the three general equations of fluid dynamics to accomplish a mathematical representation of the pipe network structure, according to the one-dimensional spatial approach, each pipe of the pipe network is considered as a branch starting from one node (the entry node) and ending at another node (the exit node).
The incidence matrix A describes the topology of the pipe network by representing the connection between nodes and branches, the number of rows of the matrix is the same as the number of nodes, the number of columns is the same as the number of branches, if a branch j enters or exits nodes i and o, the general element AijEqual to 1 or-1, the establishment of the mass conservation equation at the node and the establishment of the momentum conservation equation at the branch are to solve the coupling equation by the SIMPLE algorithm to obtain the flow G, the establishment of the energy conservation equation at the node and the solution of the differential equation to obtain the temperature T, and the specific solving process is shown in fig. 3.
The specific schematic of the correlation matrix a is as follows:
Figure BDA0002499127760000121
in simplifying the pipe network and establishing an equation in an abstract way, the following two assumptions are required:
the unsteady terms are not considered, because the hydrodynamic disturbance only needs several seconds of time when passing through the whole pipe network, and is much smaller than a calculation time step.
② the density of the fluid in the pipe network is regarded as constant, so the speed change of the inlet and outlet on one way can be ignored.
The three general equations of thermal flow are as follows:
the conservation of mass equation at node ① is A G + Gext＝0；
Wherein G is a vector comprising the mass flow rate in the branch, GextVector of mass flow rate into or out of the node, G, when fluid flows into the nodeextIs positive, when fluid is flowing out of the node, GextIs negative.
The conservation of momentum equation of the ② branch is G ═ Y.AT·P+Y·ΔPpumps；
Where P is the vector formed by the pressures of the nodes, Δ PpumpsThe vector formed by the pressure difference created for pumping.
Diagonal matrix Y represents the hydrodynamic conductance of the branches as:
Figure BDA0002499127760000131
the energy conservation equation at node ③ is:
Figure BDA0002499127760000132
wherein M is a quality matrix and is a diagonal matrix; k is a stiffness matrix containing coefficients linearly related to temperature; γ is a known constant term.
S5, taking the building layer heat load predicted in the step S3 as Dirichlet boundary conditions of three major equations corresponding to the distribution pipe network layer in the step S4, solving a mass equation and a momentum equation by using a semi-implicit method (SIMPLE algorithm) of a pressure coupling equation set to obtain flow G, and then solving the energy equation to obtain temperature T, so that the heat load prediction of the distribution pipe network layer is completed;
s6, taking the building layer heat load predicted in the step S5 as Dirichlet boundary conditions of three major equations corresponding to the main pipe network layer in the step S4, solving a mass equation and a momentum equation by using a semi-implicit method (SIMPLE algorithm) of a pressure coupling equation set to obtain flow G, and solving the energy equation to obtain temperature T, so that the heat load prediction of the distribution pipe network layer is completed.
In order to solve the equation to be solved in the solving area, a dirichlet boundary condition exists, the boundary condition is derived from the predicted values G and T of the thermal load at the upper level, the coupling equation set in claim 5 is solved by using a SIMPLE algorithm, the conventional SIMPLE algorithm is improved in the using process, and a relaxation factor α' is added in an iterative relation between the pressure P and the flow G in the iterative calculation process, which is specifically shown as follows:
P＝P′+α′Pcorr
G＝G′+α′Gcorr
wherein P 'and G' are respectively an initial pressure field and a velocity field of the algorithm, Pcorr、GcorrRespectively, the correction amounts of the current calculation process. The specific flow of the SIMPLE algorithm is shown in fig. 4.
Assuming that Y is Y', the inlet and outlet flow of the system should be set at G in terms of setting the boundary conditionextIn particular, the following principle should be followed in the setting of the boundary conditions:
for the water supply network, the simulation result of the main pipe network is used as the boundary condition of the physical model of the distribution pipe network.
And for the water return pipe network, the simulation result of the distribution pipe network is used as the boundary condition of the physical model of the main pipe network.
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 5, the method of the present invention is used to predict the building level thermal load of a heating system in a certain area, wherein the calculation formula of the predicted thermal load is:
Φ＝Gc(TSUP-TRET)
wherein G is the flow rate of hot water, c is the specific heat capacity of hot water, and TSUPFor temperature of the water supply, TRETThe temperature of the return water is; as can be seen, in the early morning to 9 am: 00 and 16: 00 to night 22: the thermal load value after 00 is high, about 380KW, whereas at 9: 00 and 16: the thermal load value between 00 is lower, because the outdoor environment temperature is higher at this stage, the corresponding thermal load value is lower; the predicted heat load value is compared with the actual heat load value of the load, the error percentage of the predicted heat load value is less than 15%, and the actual working condition of the heating system in the area is loaded, so that the method provided by the invention can reasonably and accurately predict the heat load of the floor of the building.
Referring to fig. 6, the method of the present invention is used to predict the thermal load of each layer of a heating system in a certain area, and the time-by-time thermal load curves of the three layers are compared in a unified manner, and the calculation formula of the predicted thermal load is as follows:
Φ＝Gc(TSUP-TRET)
wherein G is the flow rate of hot water, c is the specific heat capacity of hot water, and TSUPFor temperature of the water supply, TRETThe temperature of the return water is; as can be seen, in the early morning to 9 am: 00 and 16: 00 to night 22: the thermal load value after 00 is high, around 950MW, and is higher at 9: 00 and 16: the thermal load value between 00 is relatively low because the outdoor ambient temperature is relatively high at this stage, and the corresponding thermal load value ratioThe lower the cost;
it can be known from the figure that the predicted values of the heat load of the distribution pipe network layer are all higher than the predicted values of the heat load of the building layer, because the heat loss is partially formed by the heat exchange between the heat medium and the outside of the environment in the transmission process of the heating hot water, and the partial heat loss is generated at the junction of the return water pipelines of each building heating system, and the phenomenon is consistent with the working condition of the actual heating system, and the heat load of the pipe network layer can be predicted more accurately, which shows that the method can accurately predict the heat load of each layer of the regional heating system.
The above-mentioned contents are only for illustrating the technical idea of the present invention, and the protection scope of the present invention is not limited thereby, and any modification made on the basis of the technical idea of the present invention falls within the protection scope of the claims of the present invention.

Claims (10)

1. A thermal load hierarchical prediction method for a heating system of a high-rise building area is characterized by comprising the following steps:
s1, performing regional division on the high-rise building group by taking each heat exchange station as a mark, and dividing the system into a building layer, a distribution pipe network layer and a main pipe network layer by taking the primary heat exchange station, the heat supply pipe network nodes in the region and the secondary heat exchange station as boundary points;
s2, determining model input variables having large influence on heat load, and determining output variables of the model as temperature and flow according to requirements;
s3, defining the prediction model required in the step S2 as a linear model in a matrix form, predicting an output matrix Y according to an input matrix X, completing prediction of the thermal load of the building layer, and verifying the prediction result;
s4, building a physical model of the distribution pipe network and the main pipe network by using three large equations of graph theory knowledge and fluid dynamics;
s5, taking the building layer thermal load predicted in the step S3 as Dirichlet boundary conditions of three major equations corresponding to the distribution pipe network layer in the step S4, solving a mass equation and a momentum equation by using a semi-implicit method of a pressure coupling equation set to obtain flow, and then solving the energy equation to obtain temperature to complete thermal load prediction of the distribution pipe network layer;
s6, taking the building layer heat load predicted in the step S5 as Dirichlet boundary conditions of three major equations corresponding to the main pipe network layer in the step S4, solving a mass equation and a momentum equation by using a semi-implicit method of a pressure coupling equation set to obtain flow, and then solving the energy equation to obtain temperature to complete heat load prediction of the distribution pipe network layer.
2. The hierarchical prediction method for the thermal load of the heating system for high-rise building areas according to claim 1, wherein in step S2, the relationship between the influence factors and the heating load is discussed using a correlation analysis method, the influence factors related to the heating load are used as a variable x, the heating load is used as a variable y, the correlation coefficient r value of each influence factor and the heating load is calculated, the correlation coefficient is used to determine the magnitude of the correlation, and by calculating the level of the significance value, when the level of the significance is greater than or equal to the corresponding number in the table, the relationship is significant, otherwise, the relationship is not significant; the determined model input variables include the average temperature T of the previous daym,d-1(ii) a Lowest temperature T of the previous daymin,d-1(ii) a Maximum temperature T of previous daymax,d-1And the average temperature T of the daym,d(ii) a The measured quantity comprises a flow G of the primary side of the heat exchange station; water temperature T at the primary side inlet1(ii) a Water temperature T at the primary side outlet2(ii) a Water temperature T at secondary side inlet4(ii) a Water temperature T at secondary side outlet3And the ambient temperature Tenv。
3. The method for hierarchically predicting the heat load of a heating system for a high-rise building area according to claim 2, wherein the correlation coefficient r is specifically:
Figure FDA0002499127750000021
wherein the content of the first and second substances,
Figure FDA0002499127750000022
is the covariance of x and y,xis the standard deviation of x and is,yis the standard deviation of y, x is the first variable, y is the second variable,
Figure FDA0002499127750000023
is the average number of x, and is,
Figure FDA0002499127750000024
is the average of y.
4. The high-rise building district heating system heat load stratification prediction method according to claim 2, characterized in that the significance level values are specifically:
Figure FDA0002499127750000025
wherein n-2 is a degree of freedom and r is a correlation coefficient.
5. The method according to claim 1, wherein in step S3, the accuracy of the model, the degree of fit γ, and the average absolute percentage MAPE error measure are used as the evaluation index of the prediction model2The fitting degree of the prediction result and the actual value is represented as follows:
Figure FDA0002499127750000026
Figure FDA0002499127750000031
wherein, yiI time actual thermal load, yiThe' predicted load when it is i, and n is the number of data.
6. The method for predicting the thermal load stratification of a high-rise building district heating system according to claim 1, wherein in step S3, the training data test data is preprocessed by combining a standard fraction and mean interpolation method, and the input variable data set is set to have a size N, and the processing procedure is as follows:
s301, calculating a mean e (i) and a variance v (i) of the data in the data set N for each input variable respectively as follows:
Figure FDA0002499127750000032
Figure FDA0002499127750000033
s302, defining standard score deviation rate ξ (i, n), and setting maximum allowable deviation rate ξmax；
Figure FDA0002499127750000034
S303, when | ξ (i, n) | is equal to or more than ξmaxAnd eliminating abnormal data, and correcting by adopting a mean interpolation method as follows:
Figure FDA0002499127750000035
the magnitude of each parameter of the input data is different, and the input data is normalized by dispersion normalization as follows:
Figure FDA0002499127750000036
performing inverse normalization processing on the output value of the model to obtain an actual predicted value, specifically:
ysi＝ymin+oi(ymax-ymin)
wherein x isiIs the original value of the sample; x is the number ofminIs the original sample minimum; x is the number ofmaxIs the original sample maximum; x is the number ofi' is a normalized processed value;ymaxthe maximum output value of the original sample; y isminIs the minimum output value of the original sample; oiA predicted value for the model output; y issiAnd outputting a reduction value for the model.
7. The hierarchical prediction method for the thermal load of a heating system for a high-rise building area according to claim 1, wherein in step S4, each pipe of the pipe network is regarded as a branch starting from one node and ending at another node according to a one-dimensional space method; the incidence matrix A describes the topology of the pipe network by representing the connection between nodes and branches, the number of rows of the matrix is the same as the number of nodes, the number of columns is the same as the number of branches, and if a branch j enters or exits nodes i and o, a general element AijAnd when the temperature is equal to 1 or-1, establishing a mass conservation equation at a node and a momentum conservation equation at a branch, solving a coupling equation through a SIMPLE algorithm to obtain the flow G, establishing an energy conservation equation at the node and solving the differential equation to obtain the temperature T.
8. The method for predicting the thermal load stratification of a high-rise building district heating system according to claim 1, wherein the pressure P and the flow rate G are calculated as follows in step S5 and step S6:
P＝P′+α′Pcorr
G＝G′+α′Gcorr
wherein P 'and G' are respectively an initial pressure field and a velocity field of the algorithm, Pcorr、GcorrRespectively, the correction amount of the current calculation process, α' is a relaxation factor.
9. A computer readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by a computing device, cause the computing device to perform any of the methods of claims 1-8.
10. A prediction apparatus, characterized by comprising:
one or more processors, memory, and one or more programs stored in the memory and configured for execution by the one or more processors, the one or more programs including instructions for performing any of the methods of claims 1-8.
CN202010427163.3A 2020-05-19 2020-05-19 Heating system heat load layered prediction method, storage medium and prediction equipment Active CN111626003B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010427163.3A CN111626003B (en) 2020-05-19 2020-05-19 Heating system heat load layered prediction method, storage medium and prediction equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010427163.3A CN111626003B (en) 2020-05-19 2020-05-19 Heating system heat load layered prediction method, storage medium and prediction equipment

Publications (2)

Publication Number Publication Date
CN111626003A true CN111626003A (en) 2020-09-04
CN111626003B CN111626003B (en) 2023-04-18

Family

ID=72258956

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010427163.3A Active CN111626003B (en) 2020-05-19 2020-05-19 Heating system heat load layered prediction method, storage medium and prediction equipment

Country Status (1)

Country Link
CN (1) CN111626003B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112084674A (en) * 2020-09-17 2020-12-15 华北电力大学 Electromagnetic transient parallel simulation method and system based on semi-implicit relaxation
CN112100871A (en) * 2020-11-19 2020-12-18 清华四川能源互联网研究院 Decoupling method and device of multi-energy coupling system, electronic device and storage medium
CN113803776A (en) * 2021-10-14 2021-12-17 北京嘉洁能科技股份有限公司 Residential building heating prediction method and heating system

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109253494A (en) * 2018-09-05 2019-01-22 东北大学 A kind of electric heat-storage device control system and method based on heat load prediction
CN110110388A (en) * 2019-04-12 2019-08-09 天津大学 Building air conditioning load forecasting method based on data pre-processing and wavelet decomposition
CN110793094A (en) * 2019-10-28 2020-02-14 浙江大学 Comprehensive flexible scheduling method and system for urban centralized heating system
WO2020093296A1 (en) * 2018-11-06 2020-05-14 国电南瑞科技股份有限公司 Interval power flow calculation method for power-heat integrated energy system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109253494A (en) * 2018-09-05 2019-01-22 东北大学 A kind of electric heat-storage device control system and method based on heat load prediction
WO2020093296A1 (en) * 2018-11-06 2020-05-14 国电南瑞科技股份有限公司 Interval power flow calculation method for power-heat integrated energy system
CN110110388A (en) * 2019-04-12 2019-08-09 天津大学 Building air conditioning load forecasting method based on data pre-processing and wavelet decomposition
CN110793094A (en) * 2019-10-28 2020-02-14 浙江大学 Comprehensive flexible scheduling method and system for urban centralized heating system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
李祥立等: "基于BP神经网络的地板辐射供暖系统逐时负荷预测", 《暖通空调》 *
王文标等: "基于气象因素的集中供热系统热负荷预测研究", 《计算机测量与控制》 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112084674A (en) * 2020-09-17 2020-12-15 华北电力大学 Electromagnetic transient parallel simulation method and system based on semi-implicit relaxation
CN112084674B (en) * 2020-09-17 2024-02-27 华北电力大学 Electromagnetic transient parallel simulation method and system based on semi-implicit relaxation
CN112100871A (en) * 2020-11-19 2020-12-18 清华四川能源互联网研究院 Decoupling method and device of multi-energy coupling system, electronic device and storage medium
CN112100871B (en) * 2020-11-19 2021-02-19 清华四川能源互联网研究院 Decoupling method and device of multi-energy coupling system, electronic device and storage medium
CN113803776A (en) * 2021-10-14 2021-12-17 北京嘉洁能科技股份有限公司 Residential building heating prediction method and heating system

Also Published As

Publication number Publication date
CN111626003B (en) 2023-04-18

Similar Documents

Publication Publication Date Title
CN111626003B (en) Heating system heat load layered prediction method, storage medium and prediction equipment
Cox et al. Real time optimal control of district cooling system with thermal energy storage using neural networks
del Hoyo Arce et al. Models for fast modelling of district heating and cooling networks
WO2019200662A1 (en) Stability evaluation and static control method for electricity-heat-gas integrated energy system
Baeten et al. A validated model for mixing and buoyancy in stratified hot water storage tanks for use in building energy simulations
CN108256672A (en) A kind of central heating system the whole network balance adjusting method
CN107025334B (en) Method and device for establishing overall dynamic model of heat users of central heating system
Zhou et al. Dynamic simulation of natural gas pipeline network based on interpretable machine learning model
CN112069692B (en) Optimization solving method for natural gas pipe network transmission difference calculation
Hwang et al. Using deep learning approaches with variable selection process to predict the energy performance of a heating and cooling system
Cembrano et al. PLIO: a generic tool for real-time operational predictive optimal control of water networks
CN113657660A (en) Heat source load prediction method based on substation load and heat supply network hysteresis model
CN112113146B (en) Synchronous self-adaptive check method for roughness coefficient and node water demand of water supply pipe network pipeline
CN112084631A (en) Heat supply pipe network steam back supply scheduling method and system based on simulation model
CN115344019A (en) Natural gas metering flow adjusting process based on composite intelligent algorithm
CN115374527B (en) Construction method of regional electricity-heat-cold comprehensive energy dynamic simulation system
CN112182905A (en) Heat supply pipe network simulation method and device for comprehensive energy system
CN116611706A (en) Dynamic carbon emission factor measuring and calculating method based on multi-energy main body
CN111563332B (en) Urban water supply pipe network leakage positioning method based on night pressure regulation actual measurement
Liu et al. Wiener model of pressure management for water distribution network
Duzinkiewicz et al. Drinking water distribution system modelling-an approach to skeletonization
RU2820047C1 (en) Method for optimal control of large-scale pipeline systems of power engineering using artificial intelligence technology
Sorocins et al. District heating simulation model development to solve optimization problems in the market conditions
Koor et al. Database driven updatable hydraulic model for decision making
EP4322076A1 (en) A method and system for characterizing thermal energy distribution in thermal energy exchange systems

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