Steam return dispatching method and system for heating network based on simulation model

Abstract

The present invention discloses a steam return scheduling method for a heating network based on a simulation model, comprising: step S1, constructing a simulation mechanism model of a heating network with return users, and calibrating the simulation model according to historical operation data; step S2, establishing a return steam access decision evaluation model; step S3, determining a return steam scheduling scheme through real-time online simulation results of the heating network simulation mechanism model, and realizing real-time optimization of the entire network. The present invention provides a steam return scheduling method and system for a heating network based on a simulation model, so as to solve the safety problem of the heating network caused by unreasonable steam access strategies of return users.

Description

Technical Field

[0001] The invention relates to a steam return dispatching method and system for a heating pipe network based on a simulation model, and belongs to the field of intelligent control of industrial steam pipe networks. Background Art

[0002] At present, with the development of energy Internet in the field of heating, the traditional heating industry can use advanced control technology, information technology and intelligent management technology to interconnect the heat network energy nodes represented by heat sources and heat users, so as to effectively integrate all parties upstream and downstream of the industrial chain, form interactive transactions between supply and demand, realize two-way energy exchange, and effectively transform the roles of producers and consumers.

[0003] In the industrial steam heating network, there is a special type of heat user. When the production capacity decreases or the equipment fails, this type of heat user uses steam as a consumer. When the production capacity is sufficient, it can produce steam through the waste heat and supply part of the steam to the heating network as a producer. This type of heat user is called a return supply user, and the steam it supplies to the heating network is called return supply steam.

[0004] When there are heat users in an industrial park that can supply steam back, the heat network usually chooses to absorb low-grade industrial waste heat as a heat source supplement to improve the overall energy efficiency of the park. However, the supply steam is intermittent and volatile. The input of the supply steam will interfere with the flow rate and direction of the pipeline steam. In addition, the complex operation of multiple heat sources can easily cause safety hazards such as steam stagnation and backflow in some sections of the heat network.

[0005] Therefore, an unreasonable steam return access strategy will not only fail to achieve efficient energy utilization, but will also cause a decline in steam quality and even endanger the safe operation of the heating network. It can be seen that the optimal scheduling of steam return is crucial to the safe and efficient operation of the heating network. Therefore, there is an urgent need for an online real-time evaluation of the impact of steam return access on the entire network and a method to provide the best decision for steam return. Summary of the invention

[0006] The technical problem to be solved by the present invention is to overcome the deficiencies of the prior art and provide a steam return scheduling method and system for a heating network based on a simulation model to solve the safety problem of the heating network caused by unreasonable steam access strategy for return users.

[0007] In order to solve the above technical problems, the technical solution of the present invention is:

[0008] A steam return scheduling method for a heating network based on a simulation model, comprising:

[0009] Step S1, constructing a simulation mechanism model of a heating network with back-supply users, and calibrating the simulation model according to historical operation data;

[0010] Step S2, establishing a steam return access decision evaluation model;

[0011] Step S3, determining the scheduling plan for the return steam through the real-time online simulation results of the heating network simulation mechanism model, and realizing real-time optimization of the entire network.

[0012] Furthermore, the step of constructing a simulation mechanism model of a heating network with back-supply users in step S1 and calibrating the model according to historical operation data includes:

[0013] Step S11, establishing a basic model for thermal hydraulic calculation;

[0014] Step S12, supplementing and correcting the thermal hydraulic calculation basic model according to the characteristics of the supply back user;

[0015] Step S13, statically calibrating the thermal hydraulic calculation basic model according to the historical operation data of the heating network.

[0016] Furthermore, the step of establishing a thermal hydraulic calculation correction model in step S11 includes:

[0017] Step S111, abstracting the actual heating network structure into a directed flow chart model consisting of nodes and pipe segments;

[0018] Step S112, constructing an association matrix representing the connection relationship between the nodes and the pipe segments;

[0019] Step S113, writing the fluid physical balance equation into the directed flow chart model, and establishing a computational topology network model that conforms to the actual heating network characteristics, wherein the energy conservation constraint is in the form of:

[0020] The pressure at the node should meet the following requirements:

[0021] A T P=-diag 2 (Q)R P ;

[0022] In the above formula,

[0023] P=[P 1 ,P 2 ,...,P i ,...,P I ] T is the node steam pressure column vector;

[0024] diag() is a diagonal matrix function;

[0025] R P =[R P1 ,RP2 ,...,R Pj ,...,R PJ ] T is the column vector of the fluid resistance coefficient of the pipe section;

[0026] For the temperature of the node, it should satisfy:

[0027] A T T=-K T Β T [T-(T sur ) I×1 ];

[0028] In the above formula,

[0029] T=[T 1 ,T 2 ,...,T i ,...,T I ] T is the column vector of node steam temperature;

[0030] K T =[K T1 ,K T2 ,...,K Tj ,...,K TJ ] is the row vector of heat transfer coefficient between the working fluid and the environment at the inlet of the pipe section;

[0031] B is the node outflow matrix, which represents the corresponding connection relationship between the node and the node outflow pipe segment, and its mathematical expression is as follows:

[0032]

[0033] Furthermore, the step of supplementing and correcting the model according to the characteristics of the return user in step S12 includes:

[0034] Step S121, based on the thermal hydraulic calculation basic model, for heat users with return supply capability, adding heat user node net mass flow constraints that reflect their actual heat generation and use characteristics;

[0035] When the recycled user uses steam as a normal heat user, q re ≥0; when the return supply user transmits steam to the heating network, q re The following constraints should be met:

[0036]

[0037] In the above formula,

[0038] The maximum steam quantity that can be resupplied to the current resupplied user;

[0039] Step S122, distinguish the heat user node from other nodes, further supplement the basic model of thermal hydraulic calculation, and define the column vector P of the heat user node pressure and temperature u =[P 1 ,P 2 ,...,P u ,...,P U ] T and T u =[T 1 ,T 2 ,...,T u ,...,T U ] T , its mathematical expression is as follows:

[0040] P u =SP,

[0041] T u =ST;

[0042] In the above formula,

[0043] S=(s ui ) U×I Select the matrix for the node;

[0044] U represents the number of heat users currently consuming steam.

[0045] Furthermore, in step S13, historical operation data of some typical days of the target heating network are selected to statically calibrate the thermal hydraulic calculation correction model, calculate the prediction residual, and establish a second-order dynamic correction model:

[0046] F a,t =F f,t +a 2 dF f,t-2 +a 1 dF f,t-1 +a 0 +e(t);

[0047] In the above formula,

[0048] F is the steam temperature, pressure and flow rate;

[0049] F f,t is the current model prediction value;

[0050] F a,t is the predicted value of the model after correction;

[0051] dF f,t is the forecast residual for period t;

[0052] a i(i=0,1,2) is the residual autoregression coefficient;

[0053] e(t) is the inherent white noise of the model.

[0054] Furthermore, the step of establishing a steam return access decision evaluation model in step S2 includes:

[0055] Step S21, establishing a steam return scheduling objective function;

[0056] Step S22, constraining the objective function to ensure that reverse flow and stagnation do not occur during the resupply process, and that the steam parameters always reach or exceed the contract value.

[0057] Furthermore, in step S21, two factors, namely, absorbing as much return steam as possible and the change of steam parameters of the whole network after absorbing the return steam, are comprehensively considered to establish the following return steam scheduling objective function:

[0058]

[0059] In the above formula,

[0060] w 1 、w 2 They are the return steam quantity coefficient and the heating network steam parameter variation coefficient respectively;

[0061] P u,t , T u,t are respectively the steam pressure and temperature monitoring values ​​of heat user u at the current moment;

[0062] P u,t+1 , T u,t+1 They are the predicted values ​​of steam pressure and temperature of heat user u within one simulation time step respectively.

[0063] Furthermore, the objective function of step S21 in step S22 defines a solution domain that reflects the feasibility of the actual operation of the heating network, including constraints that the pipe section does not experience stagnation, reverse flow, and that the steam parameters reach the contract value:

[0064] (1) Security constraints:

[0065]

[0066] In the above formula,

[0067] Q j,t , Q j,t+1 are respectively the mass flow monitoring value of pipe segment j at the current moment and the mass flow prediction value within a simulation time step;

[0068] D j , jare the pipe diameter and working fluid density of pipe section j respectively;

[0069] u min The minimum steam flow rate to ensure that no stagnation occurs in the pipe section;

[0070] (2) Steam parameter constraints:

[0071]

[0072] In the above formula, are the contract values ​​of steam pressure and temperature of heat user u respectively.

[0073] Furthermore, in step S3, the scheduling scheme for returning steam is determined by the online simulation result of the heating network simulation mechanism model, and the step of achieving real-time optimization of the whole network includes:

[0074] Step S31, real-time monitoring of the status of the heat supply users, and modifying the parameters of the heat supply network simulation mechanism model according to the user status;

[0075] Step S32, the parameter input is solved with the heating network simulation mechanism model, and a preliminary scheduling plan for returning steam is generated according to the online simulation result;

[0076] Step S33, repeating step S32 according to the set optimization time step to achieve real-time optimization of steam return scheduling until the return user no longer has the need to transport steam to the heating network.

[0077] A steam return dispatching system for a heating network based on a simulation model, comprising:

[0078] A sensing unit, which is used to monitor steam parameters at each heat user and heat source in the heating network in real time, and provide input for model solving and target optimization;

[0079] A model calibration unit, which is used to perform static calibration on the model to provide a more accurate decision-making basis for the real-time optimization unit;

[0080] A real-time optimization unit, which is used to determine whether there is a demand for steam return in the heating network, and to perform real-time optimization of the return steam access amount according to the predicted value within the unit prediction time step, and to provide a scheduling plan for the return steam;

[0081] An execution unit is used to calculate the valve opening of the return steam according to the decision value given by the real-time optimization unit, and control the return steam flow rate.

[0082] The above-mentioned technical scheme is adopted. The present invention monitors the steam parameters at each heat user and heat source in the heating network in real time, combines the heating network simulation mechanism model, the return steam access decision evaluation system and the real-time optimization technology, and performs an online quantitative evaluation on the impact of the return steam access to the heating network. On this basis, the real-time access amount of the return steam is optimized online. The automatic control system acts according to the optimization value and adjusts the return steam flow rate to achieve the purpose of absorbing as much return steam as possible under the premise of safe operation of the entire network, taking into account the efficient utilization of energy in the steam return process and the safe operation of the heating network. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 It is an implementation flow chart of the steam return scheduling method for a heating network based on a simulation model of the present invention;

[0084] Figure 2 for Figure 1 Modular flow chart of

[0085] Figure 3 It is a structural schematic diagram of the heating pipe network of the present invention;

[0086] Figure 4 It is a flow chart of the steam return scheduling optimization algorithm based on real-time correction technology and model predictive control of the present invention;

[0087] Figure 5 The algorithm flow chart of the model prediction and decision evaluation of the present invention;

[0088] Figure 6 The algorithm flow chart of calculating the predicted value of the whole network parameter of the present invention;

[0089] Figure 7 It is a principle block diagram of the steam return dispatching system of the heating pipe network based on the simulation model of the present invention. DETAILED DESCRIPTION

[0090] In order to make the contents of the present invention more clearly understood, the present invention is further described in detail below based on specific embodiments in conjunction with the accompanying drawings.

[0091] Example 1

[0092] like Figure 1 As shown, the present invention provides a steam return scheduling method for a heating network based on a simulation model, comprising the following steps:

[0093] Step S1, construct a simulation mechanism model of the heating network with return steam users, and calibrate the model according to historical operation data; Step S2, establish a return steam access decision evaluation model with the premise of safe operation of the whole network and the goal of absorbing as much return steam as possible; Step S3, determine the optimal scheduling plan for return steam through the online simulation results of the heating network simulation mechanism model, and realize real-time optimization of the whole network.

[0094] Combination Figure 2 The above scheduling method is described in the modular process. Specifically, in step S1, Figure 2 In module 1, a basic model of the heating network is constructed according to the principles of thermodynamics, fluid mechanics and system engineering. According to the characteristics of the return users that can both use steam and produce steam, the basic model is modified by adding corresponding constraints to distinguish between ordinary heat users and return users, and the basic model is calibrated based on historical operation data. In step S2, according to Figure 2 In module 2, a decision-making evaluation system for steam return access is established by introducing a dimensionless objective function to quantify the trade-off between the maximum absorption capacity of the heating network for steam return and the impact of steam return access on the parameters of the entire heating network, as well as safety and economic constraints. In step S3, based on Figure 2 In modules 3 and 4, the real-time operation data of the heating network is input into the established model to predict the steam parameters of the entire network in the next period after a certain mass flow rate of return steam is connected. The optimal return steam quantity is solved by combining the linear optimization solver, and the optimal scheduling plan is generated and issued to the actuator. The automatic control system adjusts the opening of the return steam valve in real time to ensure the safety and efficiency of the return steam connection.

[0095] In step S1, a simulation mechanism model of a heating network with re-supply users is constructed, and the steps of calibrating the model according to historical operation data include: step S11, establishing a basic model for thermal-hydraulic calculations; step S12, supplementing and correcting the basic model for thermal-hydraulic calculations according to the characteristics of the re-supply users; step S13, statically calibrating the basic model for thermal-hydraulic calculations based on the historical operation data of the heating network.

[0096] The steps for establishing the basic model for thermal hydraulic calculation in step S11 are as follows:

[0097] Step S111, abstract the actual heating network structure into a directed flow chart model consisting of nodes and pipe segments. 1 ,N 2 ,...,N i ,...,N I} represents the set of network nodes with inflow and outflow of heat sources, heat users, tees, etc. in the heat network, i represents the node number; pipe segment E = {E 1 ,E2 ,...,E j ,...,E J} represents the set of steam pipes connecting two adjacent nodes, and j represents the pipe segment number.

[0098] Step S112, construct an association matrix A representing the connection relationship between nodes and pipe segments:

[0099]

[0100] where a ij Represents the connection relationship between the i-th node and the j-th pipe segment, and satisfies the following rules:

[0101]

[0102] In actual operation, a ij The value of is determined during the establishment of the heat network physical model, which represents the topological relationship between the specific nodes of the heat network and the pipe sections connected to them. In principle, the a corresponding to the terminal nodes such as heat source and heat user is ij The values ​​of are -1 and 1 respectively, but a ij The positive or negative value of does not affect the subsequent calculation process, but the numerical value of the calculation result has a different sign from the commonly accepted one.

[0103] Step S113, write the fluid physical balance equation into the directed graph to establish a computational topology network model that conforms to the actual heat network characteristics. When steam is used as the heat transfer medium, the model should satisfy the following equation:

[0104] (1) Conservation of mass:

[0105] AQ=q

[0106] in,

[0107] Q=[Q 1 ,Q 2 ,...,Q j ,...,Q J ] T is the mass flow column vector of the pipe section; q=[q 1 ,q 2 ,...,q i ,...,q I ] T is the net mass flow column vector of the node, q i >0 means N i There is a net mass flow in, q i <0 means N i There is a net mass flow out, q i =0 means N iThere is no net mass flow in or out. In general, for nodes such as heat sources that have only net mass flow out, q i Always less than 0; for nodes with only net mass outflow, such as hot users, q i Always greater than 0; for nodes such as tees where there is no net mass flow in or out, q i Always equal to 0. In the mass conservation equation of the present invention, since the flow of terminal nodes such as heat sources and heat users is obtained by flow meters in real time, and the flow value of intermediate nodes such as tees is always equal to 0, the net mass flow column vector q of the node is used as the input of the model, combined with the association matrix A determined by the topological structure of the heat network, the mass flow of each pipe section that is difficult to fully monitor in practice can be solved.

[0108] (2) Conservation of energy:

[0109] For the node pressure, it should satisfy:

[0110] A T P=-diag 2 (Q)R P

[0111] in,

[0112] P=[P 1 ,P 2 ,...,P i ,...,P I ] T represents the node steam pressure column vector; diag() is the diagonal matrix function; R P =[R P1 ,R P2 ,...,R Pj ,...,R PJ ] T It represents the column vector of the fluid resistance coefficient of the pipe section. The value of this coefficient vector is related to parameters such as steam density, pipe section diameter, friction factor, pipe section length and resistance element (such as valve) characteristics. It can be regarded as a constant vector within the decision time interval of connecting the return steam. The calculation method of the specific pipe section fluid resistance coefficient is as follows:

[0113]

[0114] Where ρ is the working fluid density; D is the inner diameter of the pipe section; f is the Darcy friction factor; l is the length of the pipe section; K n is the loss coefficient of the flow-blocking element.

[0115] For the node temperature, it should satisfy:

[0116] A T T=-K T Β T [T-(Tsur ) I×1 ]

[0117] Where T = [T 1 ,T 2 ,...,T i ,...,T I ] T is the node steam temperature column vector; K T =[K T1 ,K T2 ,...,K Tj ,...,K TJ ] is the row vector of the heat transfer coefficient between the working fluid at the inlet of the pipe section and the environment. The value of this coefficient vector is related to the circumference and length of the inner pipeline, the inner and outer diameters and thermal conductivity of the pipeline and the insulation layer, the convection heat transfer coefficient inside and outside the pipeline, the pipe section flow rate and the constant pressure specific heat capacity of the working fluid. It can be regarded as a constant vector within the decision time interval of connecting the return steam; Β is the node outflow matrix, which represents the corresponding connection relationship between the network node and the node outflow pipe section. Its mathematical expression is as follows:

[0118]

[0119] The calculation method of the heat transfer coefficient between a specific pipe section and the environment is as follows:

[0120]

[0121] Among them, c p is the constant pressure specific heat capacity of the working fluid; k is the heat transfer coefficient of the pipe section, which is calculated as follows:

[0122]

[0123] where h j 、h sur are the convection heat transfer coefficients between the working fluid and the pipe wall of pipe section j and between the pipe section and the environment; D j is the inner diameter of pipe segment j; 1 , D j,1 are the thermal conductivity and outer diameter of the pipe respectively; 2 ,λ 3 ,...,λ m , D j,2 ,D j,3 ,...,D j,m are the thermal conductivity and outer diameter of each insulation layer respectively.

[0124] In step S12, considering that the return supply users have the demand for steam and can also produce steam to be transported to the heating network, the basic thermal hydraulic calculation model established in step S11 is supplemented and corrected to obtain a hydraulic calculation correction model for the heating network with return supply users. The specific steps are as follows:

[0125] Step S121, based on the original basic model of thermal hydraulic calculation, for heat users with recycling capacity, add node net mass flow constraints that reflect their actual heat production and use characteristics. When the recycling user uses steam as an ordinary heat user, q re ≥0; when the return supply user transmits steam to the heat network, q re The following constraints should be met:

[0126]

[0127] in It is the maximum amount of steam that can be resupplied to the current resupplied user.

[0128] Step S122: To meet the needs of heat network status assessment, the heat user node is distinguished from other nodes (including the supply user node that is supplying steam back), the original model is further supplemented, and the column vector P of the heat user node pressure and temperature is defined. u =[P 1 ,P 2 ,...,P u ,...,P U ] T and T u =[T 1 ,T 2 ,...,T u ,...,T U ] T , its mathematical expression is as follows:

[0129] P u =SP

[0130] T u =ST

[0131] Where S = (s ui ) U×I is the node selection matrix, U represents the number of heat users currently consuming steam. Under given working conditions, S is a constant matrix used to select the corresponding heat user node parameters from the node pressure column vector and the node temperature column vector. The implementation method of S is as follows:

[0132] like Figure 3 The steam network structure shown in the figure has a corresponding correlation matrix:

[0133]

[0134] At this time, for hot users, the node selection matrix is:

[0135]

[0136] The implementation method is to first define an empty matrix S, filter out the rows in matrix A that have one and only one element 1, and when the i-th row of matrix A has the above characteristics, insert a full zero row into the S matrix, the number of columns is the number of nodes, and set the i-th element of the row to 1. This process is easy to implement in computer language.

[0137] In step S13, historical operation data of some typical days of the target heat network are selected to perform static calibration on the model parameters such as the pipe resistance coefficient and pipe heat transfer coefficient of the hydraulic calculation correction model of the heating network, so as to provide a benchmark model that can reflect the operation characteristics of the target heat network for subsequent online model prediction control, reduce the online optimization prediction residual, and improve the reliability of heat network regulation. The calibration process uses the error autoregression method, combines the prediction results of the steam heat network model with the historical operation data, calculates the prediction residual, and establishes a second-order dynamic correction model:

[0138] F a,t =F f,t +a 2 dF f,t-2 +a 1 dF f,t-1 +a 0 +e(t)

[0139] Wherein F refers to a parameter of interest, which in the present invention is a model parameter such as a pipe resistance coefficient and a pipe heat transfer coefficient; F f,t is the current model prediction value; F a,t is the model prediction value after correction; dF f,t is the forecast residual for period t; a i (i=0,1,2) is the residual autoregressive coefficient, which is a fixed value related to the characteristics of the heating network and the prediction model and can be determined by least squares recursion; e(t) is the inherent white noise of the model.

[0140] In step S2, the steps of establishing a return steam access decision evaluation model with the premise of safe operation of the entire network and the goal of absorbing as much return steam as possible include: step S21, establishing a return steam scheduling objective function; step S22, constraining the objective function to ensure that there will be no reverse flow and stagnation in the return process, and that the steam parameters always reach the contract value or above.

[0141] In step S21, the following steam return scheduling objective function is established by comprehensively considering two factors: absorbing as much return steam as possible and the change of steam parameters of the entire network after absorbing the return steam:

[0142]

[0143] where w 1 、w 2are the importance coefficients reflecting the change of the return steam volume and the steam parameters of the whole network, respectively. Both coefficients are greater than 0. The coefficients can be flexibly selected by decision makers based on experience; P u,t , T u,t are respectively the steam pressure and temperature monitoring values ​​of heat user u at the current moment; P u,t+1 , T u,t+1 They are the predicted values ​​of steam pressure and temperature of heat user u within one simulation time step respectively.

[0144] In step S22, the objective function of step S21 defines the solution domain that reflects the feasibility of the actual operation of the heat network, including the constraints that there is no stagnation or reverse flow in the pipe section and that the steam parameters reach the contract value:

[0145] (1) Security constraints:

[0146]

[0147] Where Q j,t , Q j,t+1 are the mass flow monitoring value of pipe segment j at the current moment and the mass flow prediction value within a simulation time step; D j , j are the pipe diameter and working fluid density of pipe section j respectively; u min The minimum steam flow rate to ensure that there is no stagnation in the pipe section.

[0148] (2) Steam parameter constraints:

[0149]

[0150] in are the contract values ​​of steam pressure and temperature of heat user u respectively.

[0151] Step S3, determine the best scheduling plan for return steam through the online simulation results of the heating network simulation mechanism model, and the steps to achieve real-time optimization of the entire network include: Step S31, real-time monitoring of the return user status, and correcting the model parameters according to the user status; Step S32, parameter input and model solution, generate a preliminary scheduling plan for return steam according to the online simulation results; Step S33, repeat step S32 according to the set optimization time step to achieve real-time optimization of steam return scheduling until the return user no longer has the need to transport steam to the heating network. The specific algorithm flow of step S3 is as follows Figure 4 As shown, Δt in the figure is a simulation time step; It is the maximum amount of steam that can be resupplied to the resupplied user at time t. When its value is greater than 0, it means that the resupplied user has a demand for resupplied steam at the current moment.

[0152] In step S32, based on the revised heat network mechanism model, the real-time monitoring parameter values ​​of steam flow, temperature, pressure, etc. obtained by monitoring ordinary heat users and heat source sides are used as input, and the return steam flow of the return steam user is used as the decision variable. The sequential solution calculation strategy is adopted to simulate online the theoretical steam parameters of all parts of the entire network after the return steam with a specific mass flow is connected. Combined with the simulation prediction value obtained by calculation and the return steam access decision evaluation model established in step S2, the return steam access amount is optimized. The specific implementation method of the algorithm is as follows: Figure 5 As shown in the figure, δ∈(0,1) is the attenuation coefficient, which can be set manually or determined by using the gradient descent method on the objective function; ε is the set accuracy. Figure 5 The specific method for calculating the predicted value of the whole network parameter is as follows:

[0153] (1) According to the network topology, a correlation matrix that can reflect the current connection status of the heating network nodes is established, and an indicator i is set to indicate whether static calibration is required. If necessary, i=1; otherwise, i defaults to 0.

[0154] (2) Real-time reading of the flow, temperature, and pressure parameters of the heat source in the system, as well as the heat user flow parameters.

[0155] (3) Initialize the node net mass flow vector q, fill in the measured flow parameters for the heat source and heat user nodes, and set the net mass flow of the intermediate nodes to 0.

[0156] (4) According to the law of conservation of mass, the mass flow rate of each pipe section is calculated to obtain the value of vector Q.

[0157] (5) According to the law of conservation of energy, starting from the temperature and pressure parameters of the heat source node, the pressure and temperature of each node are calculated to obtain the values ​​of vectors P and T.

[0158] (6) Output the temperature and pressure parameters of the heat user node, as well as the flow parameters of each pipe section.

[0159] (7) Determine the static calibration indicator. If correction is required, proceed to step (8); otherwise, output the predicted calculation parameters and end the calculation.

[0160] (8) Calculate the network error based on the actual steam temperature and pressure parameters of the heat user.

[0161] (9) Determine whether the network error meets the accuracy requirements. If so, reset the static calibration indicator to 0 and jump to step (6); otherwise, back-propagate the error parameters to calibrate the heat transfer coefficient and resistance coefficient of the pipe section and jump to step (5).

[0162] The detailed flow of the calculation process is as follows Figure 2 As shown:

[0163] Based on the optimization results, the optimal scheduling plan is generated and sent to the actuator to adjust the return steam flow.

[0164] Example 2

[0165] Based on Example 1, the present invention also provides a steam return scheduling system for a heating network based on a simulation model, such as Figure 7 As shown, the system mainly includes the following parts:

[0166] Sensing unit: used to monitor steam parameters at each heat user and heat source in the heating network in real time, such as steam temperature, pressure and flow, to provide input for model solving and target optimization;

[0167] Model calibration unit: used to perform static calibration on the model, providing a more accurate decision-making basis for the real-time optimization unit;

[0168] Real-time optimization unit: It includes two parts: the heating network mechanism correction module with steam return users and the steam return access decision evaluation module. It is used to determine whether there is a steam return demand in the heating network, and to optimize the steam return access quantity in real time according to the predicted value within the unit prediction time step, and to provide a scheduling plan for steam return;

[0169] Execution unit: Calculates the return steam valve opening according to the decision value given by the real-time optimization unit and controls the return steam flow rate.

[0170] The specific embodiments described above further illustrate the technical problems, technical solutions and beneficial effects solved by the present invention. It should be understood that the above are only specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A steam return scheduling method for a heating network based on a simulation model, characterized in that: include: Step S1, constructing a simulation mechanism model of a heating network with back-supply users, and calibrating the simulation model according to historical operation data; Step S2, establishing a steam return access decision evaluation model; Step S3, determining the scheduling scheme for returning steam through the real-time online simulation results of the heating network simulation mechanism model, and realizing real-time optimization of the entire network; The step S1 of constructing a simulation mechanism model of a heating network with back-supply users and calibrating the model according to historical operation data includes: Step S11, establishing a basic model for thermal hydraulic calculation; Step S12, supplementing and correcting the thermal hydraulic calculation basic model according to the characteristics of the supply back user; Step S13, statically calibrating the thermal hydraulic calculation basic model according to the historical operation data of the heating network; The step of establishing a thermal hydraulic calculation correction model in step S11 includes: Step S111, abstracting the actual heating network structure into a directed flow chart model consisting of nodes and pipe segments; Step S112, constructing an association matrix representing the connection relationship between the nodes and the pipe segments; Step S113, writing the fluid physical balance equation into the directed flow chart model, and establishing a computational topology network model that conforms to the actual heating network characteristics, wherein the energy conservation constraint is in the form of: The pressure at the node should meet the following requirements: A T P=-diag 2 (Q)R P ; In the above formula, P=[P1,P2,...,P i ,...,P I ] T is the node steam pressure column vector; diag() is a diagonal matrix function; R P =[R P1 ,R P2 ,...,R Pj ,...,R PJ ] T is the column vector of the fluid resistance coefficient of the pipe section; For the temperature of the node, it should be satisfied: A T T=-K T Β T [T-(T sur ) I×1 ]; In the above formula, T=[T1,T2,...,T i ,...,T I ] T is the column vector of node steam temperature; K T =[K T1 ,K T2 ,...,K Tj ,...,K TJ ] is the row vector of heat transfer coefficient between the working fluid and the environment at the inlet of the pipe section; B is the node outflow matrix, which represents the corresponding connection relationship between the node and the node outflow pipe segment, and its mathematical expression is as follows: The step of supplementing and correcting the model according to the characteristics of the return user in step S12 includes: Step S121, based on the thermal hydraulic calculation basic model, for heat users with the ability to supply back, adding a heat user node net mass flow constraint that reflects their actual heat production and use characteristics; When the recycled user uses steam as a normal heat user, q re ≥0; when the return supply user transmits steam to the heating network, q re The following constraints should be met: In the above formula, The maximum steam quantity that can be resupplied to the current resupplied user; Step S122, distinguish the heat user node from other nodes, further supplement the basic model of thermal hydraulic calculation, and define the column vector P of the heat user node pressure and temperature u =[P1,P2,...,P u ,...,P U ] T and T u =[T1,T2,...,T u ,...,T U ] T , its mathematical expression is as follows: P u =SP, T u =ST; In the above formula, S=(s ui ) U×I Select the matrix for the node; U represents the number of heat users currently consuming steam; In step S13, historical operation data of some typical days of the target heating network are selected to statically calibrate the thermal hydraulic calculation correction model, calculate the prediction residual, and establish a second-order dynamic correction model: F a,t =F f,t +a2dF f,t-2 +a1dF f,t-1 +a0+e(t); In the above formula, F is the steam temperature, pressure and flow rate; F f,t is the current model prediction value; F a,t is the predicted value of the model after correction; dF f,t is the forecast residual for period t; a i (i=0,1,2) is the residual autoregression coefficient; e(t) is the inherent white noise of the model; The step of establishing the steam return access decision evaluation model in step S2 includes: Step S21, establishing a steam return scheduling objective function; Step S22, constraining the objective function to ensure that reverse flow and stagnation do not occur during the resupply process, and that the steam parameters always reach or exceed the contract value; In step S21, two factors are comprehensively considered: absorbing as much return steam as possible and the change of steam parameters of the whole network after absorbing the return steam, and the following return steam scheduling objective function is established: In the above formula, w1 and w2 are respectively the coefficient of return steam quantity and the coefficient of variation of steam parameters of the heating network; P u,t , T u,t are respectively the steam pressure and temperature monitoring values ​​of heat user u at the current moment; P u,t+1 , T u,t+1 are respectively the predicted values ​​of steam pressure and temperature of heat user u within one simulation time step; The objective function of step S21 in step S22 defines a solution domain that reflects the feasibility of the actual operation of the heating network, including constraints that the pipe section does not experience stagnation, reverse flow, and that the steam parameters reach the contract value: (1) Security constraints: In the above formula, Q j,t , Q j,t+1 are respectively the mass flow monitoring value of pipe segment j at the current moment and the mass flow prediction value within a simulation time step; D j , j are the pipe diameter and working fluid density of pipe section j respectively; u min The minimum steam flow rate to ensure that no stagnation occurs in the pipe section; (2) Steam parameter constraints: In the above formula, are the contract values ​​of steam pressure and temperature of heat user u respectively.

2. The method for scheduling steam return in a heating network based on a simulation model according to claim 1, characterized in that: The step S3 of determining the scheduling scheme for returning steam by the online simulation result of the heating network simulation mechanism model to achieve real-time optimization of the whole network includes: Step S31, real-time monitoring of the status of the heat supply users, and modifying the parameters of the heat supply network simulation mechanism model according to the user status; Step S32, the parameter input is solved with the heating network simulation mechanism model, and a preliminary scheduling plan for returning steam is generated according to the online simulation result; Step S33, repeating step S32 according to the set optimization time step to achieve real-time optimization of steam return scheduling until the return user no longer has the need to transport steam to the heating network.

3. A heating network steam return scheduling system using the heating network steam return scheduling method based on a simulation model as described in any one of claims 1 to 2, characterized in that: include: A sensing unit, which is used to monitor steam parameters at each heat user and heat source in the heating network in real time, and provide input for model solving and target optimization; A model calibration unit, which is used to perform static calibration on the model to provide a more accurate decision-making basis for the real-time optimization unit; A real-time optimization unit, which is used to determine whether there is a demand for steam return in the heating network, and to perform real-time optimization of the return steam access amount according to the predicted value within the unit prediction time step, and to provide a scheduling plan for the return steam; An execution unit is used to calculate the valve opening of the return steam according to the decision value given by the real-time optimization unit, and control the return steam flow rate.

Images