A dual-layer robust state estimation method and system for an electric-thermal integrated energy system

Abstract

The present invention provides a double-layer robust state estimation method for an electric-thermal integrated energy system, comprising: obtaining power system quantity measurements and thermal system quantity measurements of the electric-thermal integrated energy system; the thermal system quantity measurements include hydraulic network quantity measurements and thermal network quantity measurements; inputting the power system quantity measurements and hydraulic network quantity measurements into a hydraulic network state estimation model of the power system to obtain power system state variable estimates and hydraulic network state variable estimates, as well as hydraulic network branch flow and node injection flow estimates; inputting the hydraulic network branch flow and node injection flow estimates and thermal network quantity measurements into a thermal network state estimation model to obtain thermal network state variable estimates. The thermal network state estimation model of the present invention is constructed by taking into account temperature constraints between nodes and thermal power measurement, which equivalently increases the measurement redundancy of the thermal network part and improves the accuracy of the state variable estimates of the thermal network part.

Description

Technical Field

[0001] The present invention belongs to the field of integrated energy state assessment, and in particular relates to a double-layer robust state estimation method and system for an electric-thermal integrated energy system. Background Art

[0002] Integrated energy systems are considered the primary form of energy application in future human society. With the application and development of various energy conversion devices (such as combined heat and power (CHP)—CHP refers to the simultaneous generation of electricity and usable heat from a variety of sources, including fossil fuels, surplus energy, renewable energy, and electricity), the coupling of various energy sources (such as electricity, heat, and natural gas) has gradually increased. Among them, the coupling of electricity and heat is one of the primary applications of integrated energy systems. To achieve comprehensive and accurate perception of the integrated electricity-heat system (IEHS), a state estimation (SE) method for the IEHS is required. This provides reliable, mature data for the corresponding energy management system (EMS), enabling unified management and scientific scheduling of the IEHS.

[0003] While state estimation and its application in power systems are relatively mature, their application in thermal systems remains an ongoing research area. A bilinear robust state estimation method for integrated electric and thermal energy systems (IEHS) is proposed in the paper "A Bilinear Robust State Estimation Method for Integrated Electric and Thermal Energy Systems." First, the measurement equations of the IEHS are linearized by introducing auxiliary state variables and measurements. Second, a weighted least absolute value (WLAV) SE model is established, and the auxiliary variables are estimated by solving this model. Finally, estimates of the original IEHS state variables are obtained through a single nonlinear transformation and a linear weighted least squares (WLS) method. This method requires no nonlinear iterations, is computationally efficient, and has strong robustness against multiple imperfect data. However, this method has the following issues: The thermal system modeling fails to consider the constraints between node temperatures and thermal power measurements, resulting in relatively large errors in the estimated thermal system state variables. Furthermore, the introduction of auxiliary variables effectively increases the number of power system state variables while keeping the number of measurements unchanged, resulting in a partial loss of measurement redundancy during the solution process. Therefore, how to improve the accuracy of state estimation of the current electric-thermal integrated energy system is a problem that needs to be solved by those skilled in the art. Summary of the Invention

[0004] To overcome the above-mentioned deficiencies of the prior art, the present invention provides a two-layer robust state estimation method for an electric-thermal integrated energy system, comprising:

[0005] Obtaining power system quantity measurements and thermal system quantity measurements of an electric-thermal integrated energy system; wherein the thermal system includes: a hydraulic network and a thermal network, and the thermal system quantity measurements include hydraulic network quantity measurements and thermal network quantity measurements;

[0006] Input the power system quantity measurements and the water network quantity measurements into a pre-built power system water network state estimation model to obtain power system state variable node voltage amplitude estimation values, node phase angle estimation values, water network state variable node pressure head estimation values, water network branch flow estimation values, and node injection flow estimation values;

[0007] Input the estimated values ​​of branch flow of the hydraulic network, the estimated values ​​of node injection flow, and the measured values ​​of thermal network quantities into a pre-built thermal network state estimation model to obtain the estimated values ​​of node heating temperature and node heat recovery temperature of the thermal network state variables;

[0008] The thermal network state estimation model is constructed by considering temperature constraints and thermal power measurements between thermal network nodes.

[0009] Preferably, the construction of the power system hydraulic network state estimation model includes:

[0010] determining a power system linear measurement equation based on a set first auxiliary state variable and a first auxiliary quantity measurement;

[0011] determining a hydraulic network linear measurement equation based on a set second auxiliary state variable and a second auxiliary quantity measurement;

[0012] Based on the linear measurement equations of the power system, the linear measurement equations of the hydraulic network, and the coupling mode of the coupling nodes of the power system and the thermal system, a unified linear measurement model of the power system and the hydraulic network is constructed;

[0013] Based on the unified linear measurement model of the power system and the water network, a unified linear weighted minimum absolute value state estimation model of the power system and the water network is constructed;

[0014] Constructing a power system and hydraulic network state estimation model based on the relationship between the auxiliary state variables of the power system and the unified linear weighted minimum absolute value state estimation model of the power system and the hydraulic network;

[0015] Among them, the first auxiliary state variable is set by the power system state variable; the first auxiliary quantity measurement is set by the power system quantity measurement; the second auxiliary state variable is set by the hydraulic network state variable; and the second auxiliary quantity measurement is set by the hydraulic network quantity measurement.

[0016] Preferably, the first auxiliary state variable and the first auxiliary quantity are measured as follows:

[0017]

[0018]

[0019] in, is the first auxiliary state variable, is the first auxiliary quantity introduced, is the first auxiliary quantity measurement, U i is the voltage amplitude of node i, P i is the injected active power of node i, Q i is the reactive power injected into node i, P ij is the active power of branch ij, Q ij is the reactive power of branch ij;

[0020] The calculation formula of the first auxiliary amount is as follows:

[0021]

[0022] Among them, U j The voltage amplitude at node j, θ ij is the phase angle difference between node i and node j.

[0023] Preferably, the linear measurement equation of the power system is as follows:

[0024]

[0025] Where U i The voltage amplitude at node i, P i is the injected active power of node i, Q i is the reactive power injected into node i, P ij is the active power of branch ij, Q ij is the reactive power of branch ij, Ni is the number of nodes in the power system, is the first auxiliary quantity, g si is the equivalent resistance to ground of node i, b si is the equivalent reactance to ground of node i, g ij is the equivalent resistance of branch ij, b ij is the equivalent reactance of branch ij, G ij Determined by the equivalent resistance of branch ij, B ij Determined by the equivalent reactance of branch ij.

[0026] Preferably, the second auxiliary state variable and the second auxiliary quantity are measured as follows:

[0027]

[0028]

[0029] in, is the second auxiliary state variable, α a is the second auxiliary quantity, Second auxiliary quantity measurement, is the water flow of branch ij, Inject water flow into the node;

[0030] The α a The calculation formula of the second auxiliary amount is as follows:

[0031]

[0032] in, is α a The elements in p ij is the pipeline pressure head loss, s ij Determined by the relationship between the pressure head at node i and the pressure head at node j.

[0033] Preferably, the hydraulic network linear measurement equation is as follows:

[0034]

[0035] Where, is the water flow of branch ij, Inject water flow into the node, K ij is the pipeline impedance coefficient of branch ij, p ij is the pipeline pressure head loss, s ij Determined by the relationship between the pressure head at node i and the pressure head at node j, is the second auxiliary amount α a Elements in .

[0036] Preferably, the unified linear measurement model of the power system and the water network is as follows:

[0037]

[0038] Where x a is the unified auxiliary state variable of the power system and hydraulic network, z a For unified auxiliary quantity measurement of power system and water network, is the first auxiliary state variable, is the second auxiliary state variable, is the first auxiliary quantity measurement, is the second auxiliary quantity measurement, is the heat energy generated by gas turbine or internal combustion engine coupling, N1 is the number of nodes coupled with gas turbine or internal combustion engine, is the heat energy generated by the steam turbine coupling mode, N2 is the number of nodes using steam turbine coupling, Ha is the unified constant coefficient matrix of the power system and the hydraulic network, r a It is the unified measurement error of the power system and water network.

[0039] Preferably, a unified linear weighted minimum absolute value state estimation model for the power system and the hydraulic network is as shown below:

[0040] minw(u+v)

[0041]

[0042] Where w is the unified measurement weight matrix of the power system and the water network, u and v are two types of non-negative variables introduced by the unified model of the power system and the water network, and z a For unified auxiliary quantity measurement of power system and water network, H a is the unified constant coefficient matrix of the power system and the hydraulic network, x a It is a unified auxiliary state variable for the power system and hydraulic network.

[0043] Preferably, based on the relationship between the auxiliary state variables of the power system and the unified linear weighted minimum absolute value state estimation model of the power system and the hydraulic network, a power system hydraulic network state estimation model is constructed, including:

[0044] Construct the relationship between auxiliary state variables of the power system;

[0045] Constructing a second-order cone inequality constraint based on a relationship between auxiliary state variables of the power system;

[0046] Based on the second-order cone inequality constraint and the unified linear weighted minimum absolute value state estimation model of the power system and the hydraulic network, a power system hydraulic network state estimation model is constructed.

[0047] Preferably, the second-order cone inequality constraint is as shown below:

[0048]

[0049] Where, is the first auxiliary amount.

[0050] Preferably, the power system hydraulic network state estimation model is as follows:

[0051] min∑λ ij R ij -w(u+v)

[0052]

[0053] Where w is the unified measurement weight matrix of the power system and the water network, u and v are two types of non-negative variables introduced by the unified model of the power system and the water network, and z a For unified auxiliary quantity measurement of power system and water network, H a is the unified constant coefficient matrix of the power system and the hydraulic network, x a is the unified auxiliary state variable of the power system and hydraulic network, is the first auxiliary quantity, λ ij It is an adjustment parameter whose size is equal to the weight value of the power measurement of the corresponding branch ij in the power system.

[0054] Preferably, the power system quantity measurements and the hydraulic network quantity measurements are input into a pre-built power system hydraulic network state estimation model to obtain power system state variable node voltage amplitude estimation values, node phase angle estimation values, hydraulic network state variable node pressure head estimation values, hydraulic network branch flow estimation values, and node injection flow estimation values, including:

[0055] Input the power system quantity measurement and the water network quantity measurement into a pre-built power system water network state estimation model to calculate the first auxiliary state variable estimation value and the second auxiliary state variable estimation value;

[0056] Based on the first auxiliary state variable estimate and the second auxiliary state variable estimate, after nonlinear transformation, calculate the power system branch phase angle difference estimate, the node voltage amplitude estimate and the hydraulic network pipeline pressure head loss estimate;

[0057] Based on the estimated values ​​of the phase angle differences of the power system branches and the estimated values ​​of the pressure head losses of the hydraulic network pipelines, the estimated values ​​of the phase angles of the power system nodes and the estimated values ​​of the pressure heads of the hydraulic network nodes are calculated through linear transformation;

[0058] Based on the estimated value of the hydraulic network node pressure head, the estimated value of the hydraulic network branch flow and the node injection flow is calculated.

[0059] Preferably, the construction of the thermal network state estimation model includes:

[0060] Based on the thermal network state variables and thermal network quantity measurements, the thermal network linear measurement equation is constructed;

[0061] Based on the thermal network linear measurement equation, a linear weighted minimum absolute value state estimation model of the thermal network is constructed;

[0062] The estimated flow rates of hydraulic network branches and node injection flow rates are set as pseudo-measurements, and the temperature constraints between nodes in the thermal network are determined.

[0063] A thermal network state estimation model is constructed based on the linear weighted minimum absolute value state estimation model of the thermal network and the constraints.

[0064] Preferably, the thermal network state variables and thermal network quantity measurements are as follows:

[0065] x t =[T s ;T r ]

[0066] z t =[φ;T s ;T r ]

[0067] Among them, x t is the state variable of the thermal network, T s is the node heating temperature, T r is the node reheating temperature; z t is the thermal network quantity measurement, and φ is the node thermal power.

[0068] Preferably, the thermal network linear measurement equation is as follows:

[0069]

[0070] Where, φ i is the thermal power of node i, is the water injection flow at node i, C p is the specific heat capacity of water, T si is the heating temperature of node i, T ri is the reheating temperature of node i.

[0071] Preferably, the linear weighted minimum absolute value state estimation model of the thermal network is as follows:

[0072] minw t (u t +v t )

[0073]

[0074] Where w t is the measurement weight matrix of the thermal network, u t and v t Two types of non-negative variables are introduced for the thermal network, z t is the thermal network quantity measurement, x t is the state variable of the thermal network, H t Constant coefficient matrix of the thermal network.

[0075] Preferably, the constant coefficient matrix H of the thermal network t It is expressed as follows:

[0076]

[0077] Among them, C p is the specific heat capacity of water, Inject water flow estimates into nodes.

[0078] Preferably, the temperature constraint between the nodes of the thermal network is as follows:

[0079]

[0080]

[0081] Where, T end is the pipe end temperature, T start is the node temperature at the beginning of the pipeline, T a is the ambient temperature, λ h is the heat transfer coefficient per unit length of the pipeline, L is the length of the pipeline, C p is the specific heat capacity of water, is the estimated value of the injected water flow at node i, is the estimated total flow of a pipe outflowing the node, is the estimated flow rate of a pipeline injected into the node, T out is the node mixing temperature, T in is the node temperature at the beginning of the pipeline.

[0082] Preferably, the thermal network state estimation model is as shown below:

[0083] minw t (u t +v t )

[0084]

[0085] Where w t is the measurement weight matrix of the thermal network, u t and v t Two types of non-negative variables are introduced for the thermal network, z t is the thermal network quantity measurement, x t is the state variable of the thermal network, H t The constant coefficient matrix of the thermal network, T end is the pipe end temperature, T start is the node temperature at the beginning of the pipeline, T a is the ambient temperature, λ h is the heat transfer coefficient per unit length of the pipeline, L is the length of the pipeline, C p is the specific heat capacity of water, is the estimated value of the injected water flow at node i, is the estimated total flow of a pipe outflowing the node, is the estimated flow rate of a pipeline injected into the node, T out is the node mixing temperature, T in is the node temperature at the beginning of the pipeline.

[0086] Based on the same concept, the present invention also provides a two-layer robust state estimation system for an electric-thermal integrated energy system, comprising:

[0087] A data acquisition module is used to obtain power system quantity measurements and thermal system quantity measurements of the electric and thermal integrated energy system; wherein the thermal system includes: a hydraulic network and a thermal network, and the thermal system quantity measurements include hydraulic network quantity measurements and thermal network quantity measurements;

[0088] A power system hydraulic network state estimation module is used to input the power system quantity measurements and hydraulic network quantity measurements into a pre-built power system hydraulic network state estimation model to obtain power system state variable node voltage amplitude estimation values, node phase angle estimation values, hydraulic network state variable node pressure head estimation values, hydraulic network branch flow estimation values, and node injection flow estimation values;

[0089] a thermal network state estimation module, configured to input the hydraulic network branch flow estimate and node injection flow estimate, as well as thermal network quantity measurements, into a pre-built thermal network state estimation model to obtain thermal network state variables, namely, node heating temperature estimate and node heat return temperature estimate;

[0090] The thermal network state estimation model is constructed by considering temperature constraints and thermal power measurements between thermal network nodes.

[0091] Compared with the closest prior art, the present invention has the following beneficial effects:

[0092] The present invention provides a double-layer robust state estimation method for an electric-thermal integrated energy system, comprising: obtaining power system quantity measurements and thermal system quantity measurements of the electric-thermal integrated energy system; wherein the thermal system comprises a hydraulic network and a thermal network, and the thermal system quantity measurements comprise hydraulic network quantity measurements and thermal network quantity measurements; inputting the power system quantity measurements and hydraulic network quantity measurements into a pre-built power system hydraulic network state estimation model, obtaining power system state variable node voltage amplitude estimation values, node phase angle estimation values, hydraulic network state variable node pressure head estimation values, hydraulic network branch flow estimation values, and Node injection flow estimation value; the hydraulic network branch flow estimation value and the node injection flow estimation value and the thermal network quantity measurement are input into a pre-constructed thermal network state estimation model to obtain the thermal network state variable node heating temperature estimation value and the node heat recovery temperature estimation value, the thermal network state estimation model takes into account the temperature constraints and thermal power measurement construction between the thermal network nodes, the thermal network state estimation model of the present invention takes into account the linear temperature constraints and thermal power measurement construction between the thermal network nodes, equivalently increases the measurement redundancy of the thermal network part, and improves the accuracy of the state variable estimation value of the thermal network part.

[0093] At the same time, the present invention introduces second-order cone constraints in the process of establishing the power system hydraulic network state estimation model, which makes up for the loss of measurement redundancy caused by the introduction of auxiliary variables in the power system part and improves the accuracy of the state variable estimation of the power system part. BRIEF DESCRIPTION OF THE DRAWINGS

[0094] Figure 1 A schematic diagram of a double-layer robust state estimation method for an electric-thermal integrated energy system provided by the present invention;

[0095] Figure 2 A schematic diagram of a double-layer robust state estimation system for an electric-thermal integrated energy system provided by the present invention;

[0096] Figure 3 A schematic diagram of a solution process for a dual-layer robust state estimation model provided in an embodiment of the present invention;

[0097] Figure 4 Provides an average value of the estimated error of the state variables of the power system in an embodiment of the present invention;

[0098] Figure 5 The embodiment of the present invention provides an average value of the estimation error of the state variables of the thermal system. DETAILED DESCRIPTION

[0099] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0100] Example 1:

[0101] The present invention provides a two-layer robust state estimation method for an electric-thermal integrated energy system, as shown in the schematic diagram. Figure 1 As shown, it includes: obtaining power system quantity measurements and thermal system quantity measurements of the electric-thermal integrated energy system; wherein the thermal system includes: a hydraulic network and a thermal network, and the thermal system quantity measurements include hydraulic network quantity measurements and thermal network quantity measurements; inputting the power system quantity measurements and hydraulic network quantity measurements into a pre-built power system hydraulic network state estimation model to obtain power system state variable node voltage amplitude estimation values, node phase angle estimation values ​​and hydraulic network state variable node pressure head estimation values, as well as hydraulic network branch flow estimation values ​​and node injection flow estimation values; inputting the hydraulic network branch flow estimation values ​​and node injection flow estimation values ​​and thermal network quantity measurements into a pre-built thermal network state estimation model to obtain thermal network state variable node heating temperature estimation values ​​and node heat recovery temperature estimation values.

[0102] S1 obtains the power system quantity measurement and thermal system quantity measurement of the electric-thermal integrated energy system; wherein the thermal system includes: a hydraulic network and a thermal network, and the thermal system quantity measurement includes a hydraulic network quantity measurement and a thermal network quantity measurement.

[0103] S2 inputs the power system quantity measurements and hydraulic network quantity measurements into a pre-built power system hydraulic network state estimation model to obtain the power system state variable node voltage amplitude estimation value, node phase angle estimation value and hydraulic network state variable node pressure head estimation value, hydraulic network branch flow estimation value and node injection flow estimation value.

[0104] Construction of S2-1 Power System Hydraulic Network State Estimation Model

[0105] S2-1-1 IEHS measurement model

[0106] S2-1-1-1 Measurement equations of power systems

[0107] In the power system, the state variable x e Including node phase angle θ i and node voltage amplitude U i . Quantity measurement z e Generally includes: node voltage amplitude U i , node injection power P i , Q i And the branch power P ij , Q ij The measurement equation is shown in formula (1):

[0108]

[0109] For simplicity, the measurement noise is ignored here. At the same time, Equation (1) uses a π-type equivalent circuit, and the parameters of the equivalent circuit are: g ij =g s / k,b ij =b s / k,g si =(1-k)g s / k 2 , b si =(1-k)b s / k 2 +b c / 2, G ij =-g ij , B ij =-b ij , g s +jb s is the series susceptance, b c is the charging susceptance, k represents the branch ratio (for branches without transformers, k = 1, b c =0).

[0110] Where U i 、U j are the voltage amplitudes of nodes i and j respectively, P i , Q i is the active power and reactive power injected into node i, P ij , Q ij are the active power and reactive power of branch ij respectively, N i is the number of nodes in the power system, θ ij is the phase angle difference between node i and node j, g s 、b s are the actual resistance and actual reactance of branch ij, g ij 、b ij are the equivalent resistance and equivalent reactance of branch ij, g si 、b si are the equivalent resistance and reactance to ground of node i, respectively.

[0111] S2-1-1-2 Measurement equations of thermal systems

[0112] A thermal system consists of a hydraulic network and a thermal network. Heat, in the form of water or steam, is transferred between the heat source and the user through heat supply and heat return pipes. When analyzing a thermal system, it is typically modeled as a hydraulic network model and a thermal network model, respectively.

[0113] S2-1-1-2-1 Measurement equations for hydraulic networks

[0114] Since the variables in the hydraulic network generally include pressure and flow, the state variable x in the hydraulic network of the present invention ish is the node pressure head p i . Quantity measurement z h Including the node pressure head p i , branch ij flow Inject traffic with node i The expression of the measurement equation is (for simplicity, measurement noise is ignored here):

[0115]

[0116] Where, is the water flow of branch ij, K ij is the pipe impedance coefficient of branch ij, when p i ≥p j When, s ij =1; otherwise, s ij =-1. K is the pipeline impedance coefficient, usually given by the Colebrook-White equation, expressed as:

[0117]

[0118] Where L represents the length of the pipe (m), D represents the diameter of the pipe (m), and ρ is the density of water (kg / m 3 ), g is the acceleration due to gravity (kg·m / s 2 ). f represents the friction factor, expressed as:

[0119]

[0120]

[0121] Where, ε is the roughness of the pipe (m), Re is the Reynolds number, v is the flow velocity in the pipe (m / s), and μ is the kinematic viscosity of water (m 2 / s). Formula (4) is an implicit equation. The present invention adopts the Haaland formula

[19] Approximate solution f:

[0122]

[0123] S2-1-1-2-2 Measurement equations of thermal network

[0124] State variable x in the thermal network t Including node heating temperature T si and node reheating temperature T ri . Quantity measurement z t Generally includes the node heating temperature T si , node reheating temperature T ri , node injection traffic and node thermal power φi The measurement equation is expressed as (for simplicity, measurement noise is ignored here):

[0125]

[0126] Among them, C p Indicates specific heat capacity (J / (kg·K)). At the same time, the temperature T at the beginning and end of the pipeline start With T end The following relationship exists:

[0127]

[0128] Among them, T a Represents the ambient temperature, which is a constant; λ h is the heat transfer coefficient per unit length of the pipeline (W / (m·K)). When a node is connected to multiple injection pipelines at the same time, the mixed temperature of the node is expressed as:

[0129]

[0130] Among them, T out represents the mixed temperature of the node, is the total flow out of the node. T in The flow rate of a certain pipeline injected into the node and the temperature of the corresponding head end node.

[0131] Measurement model of S2-1-1-3 coupling unit

[0132] The coupling unit of the present invention mainly considers combined heat and power (CHP) equipment, whose main types include gas turbines, internal combustion engines and steam turbines. For gas turbines and internal combustion engines, the heat energy φ CHP With electric energy P CHP The relationship is:

[0133]

[0134] For steam turbine, φ CHP With P CHP The relationship is expressed as:

[0135]

[0136] Among them, c m , Z represents the heat-electricity generation ratio, P con is the maximum output power of the steam turbine, F in is the input fuel power, η e The conversion efficiency of electrical energy.

[0137] IEHS operation modes include island operation and grid-connected operation. 1) Island operation: The slack nodes of the heating network are connected to a node on the grid (excluding slack nodes) via a CHP, while the slack nodes on the grid are connected to a node on the heating network (excluding slack nodes) via another CHP. 2) Grid-connected operation: The slack nodes of the heating network are connected to a node on the grid (excluding slack nodes) via a CHP, while the slack nodes on the grid are connected to a larger grid.

[0138] Therefore, the coupling relationship between the power grid and the heat network under island operation is closer. This paper mainly studies the IEHS under island operation. In addition, in addition to the relaxed nodes, other types of nodes (such as PQ, PV nodes, heat load nodes, etc.) can also be connected through CHP. Therefore, the coupling between the power grid and the heat network can be regarded as a zero measurement constraint, expressed as:

[0139]

[0140] in, They represent the heat energy and electric energy generated by the CHP that satisfy the coupling relations (10) and (11), respectively. N1 and N2 represent the number of coupling nodes that satisfy the coupling relations (10) and (11), respectively. N1 , 0 N2 are column vectors of N1 and N2 dimensions respectively.

[0141] In summary, the IEHS measurement model is expressed as:

[0142]

[0143] Among them, h e (x e ),h h (x h ) and h t (x t ) are shown in formulas (1), (2) and (7) respectively. e , r h With r t represent the measurement errors of the power system, water network, and thermal network respectively.

[0144] S2-1-2 Establishment of the first-level robust state estimation (RSE) model based on second-order cone programming

[0145] The first-level state estimation modeling targets power systems and hydropower networks. The main purpose of this modeling is to transform the traditional nonlinear, non-convex SE model into a linear, convex optimization RSE model. This involves linearizing the measurement equations, establishing the WLAV model, and introducing second-order cone inequality constraints.

[0146] S2-1-2-1 Linearization of measurement equations

[0147] Traditional nonlinear, nonconvex SE methods may yield locally optimal solutions or fail to converge. The fundamental issue is the nonlinearity of the measurement equations in the SE model. This invention linearizes the measurement equations by introducing appropriate auxiliary state variables and auxiliary measurements.

[0148] S2-1-2-1-1 Linearization of Power System Measurement Equations

[0149] Let the square of the node voltage amplitude be Alternative node voltage amplitude U i As a new quantity measurement, the rest remain unchanged. At the same time, select the auxiliary variable in, Auxiliary quantity measurement The linearized measurement equation is expressed as (for simplicity, measurement noise is ignored here):

[0150]

[0151]

[0152] Linearization of S2-1-2-1-2 Hydraulic Network Measurement Equations

[0153] Let the square root of the pressure head loss of pipe ij be (p ij =p i -p j ) instead of the pressure head p at nodes i and j i With p j .choose As auxiliary variables, Auxiliary quantity measurement The linearized measurement equation is expressed as:

[0154]

[0155]

[0156] S2-1-2-1-3 Establish a unified linear measurement model for the power system and hydraulic network

[0157] Considering the thermal power measurement of the heating network node corresponding to the active power of the power system coupling node Then the unified linear measurement model of the power system and the hydraulic network is expressed as:

[0158]

[0159] Among them, r a=[r e ; r h ; r cp1 ; r cp2 ],r cp1 、r cp2 They are The corresponding measurement error. a is a constant coefficient matrix, specifically composed of the following parts:

[0160]

[0161] Establishment of S2-1-2-2WLAV model

[0162] The above linearized measurement model (18) can be constructed as a WLAV-based RSE model:

[0163] minw|r a |

[0164] stz a -H a x a =r a (20)

[0165] Where w is the measurement weight matrix. By introducing two types of non-negative variables u and v, we can obtain an equivalent linear WLAV (Linear-WLAV, L-WLAV) model, which is expressed as:

[0166]

[0167] Establishment of S2-1-2-3 linear second-order cone programming model

[0168] Model (21) already belongs to the linear convex optimization RSE model. However, for the power system, the introduction of auxiliary state variables increases the number of state variables by B. e , where B e is the number of branches in the power system. This reduces the measurement redundancy of the power system and affects the estimation accuracy.

[0169] Furthermore, the auxiliary state variables in the power system and They have the following relationship:

[0170]

[0171] Relax (22) to transform the quadratic equality into a second-order cone inequality constraint, and we get:

[0172]

[0173] Adding (23) to model (21) yields the RSE model based on second-order cone programming:

[0174] min∑λ ij R ij -w(u+v)

[0175]

[0176] Among them, λ ij is the adjustment parameter, which is equal to the weight value of the power measurement of the corresponding branch ij in the power system. It can be seen that the number of inequality constraints (23) is B e , so it is equivalent to increasing the number of quantity measurements in the power system part of model (24) by B e , thus making up for the loss of some measurement redundancy in model (21).

[0177] Solution of S2-2 Power System Hydraulic Network State Estimation Model

[0178] The first-layer power system hydraulic network state estimation model (24) is a linear SOCP model, and the present invention uses the optimization software MOSEK to solve it. It includes the following steps:

[0179] S2-2-1 solves model (24) and obtains auxiliary variables in the power system and auxiliary variables in hydraulic networks estimated value.

[0180] S2-2-2 nonlinear transformation. For the hydraulic network, the pipe pressure head loss p ij With auxiliary variables The relationship is:

[0181]

[0182] For the power system, the phase angle difference θ of its branches ij , state variable U i With auxiliary variables The following relationship exists:

[0183]

[0184] Therefore, through nonlinear changes (28) and (29), we can get p ij ,θ ij and U i Estimated value.

[0185] S2-2-3 linear transformation. For power systems and hydropower networks, the branch phase angle difference θ b , Pipeline pressure head loss p b The relationship with state variables θ and p is:

[0186]

[0187] Among them, θ and p are the column vectors of the node phase angle and the node pressure head respectively; θ b 、p b are the column vectors of branch phase angle difference and pipeline pressure head loss respectively; A e 、A h are the reduced-order node-branch correlation matrices (excluding slack nodes) in the power system and the hydraulic network, respectively.

[0188] Therefore, in the first level state estimation, by solving model (24), performing a nonlinear transformation and a linear transformation, the estimated values ​​of the state variables of the power system and the hydraulic network can be obtained. In addition, based on the estimated value of p, the pipeline flow m and the node injection flow m can be obtained. q Estimated value of and Both will be used as pseudo-quantity measurements during the second-layer state estimation.

[0189] S3 inputs the estimated values ​​of the hydraulic network branch flow, the estimated values ​​of the node injection flow, and the thermal network quantity measurements into a pre-built thermal network state estimation model to obtain the estimated values ​​of the thermal network state variables node heating temperature and node heat recovery temperature.

[0190] Construction of S3-1 thermal network state estimation model

[0191] The modeling object of the second-level state estimation is the thermal network, and the modeling idea is the same as the first-level. Among them, the estimated value of branch flow is obtained through the first-level state estimation Estimated value of node injection flow Both are considered as pseudo measurements for state estimation in the second layer.

[0192] Construction of S3-1-1 Thermal Network L-WLAV Model

[0193] The measurement equation of the thermal network is a linear equation, where x t =[T s ;T r ],z t =[φ;T s ;T r ]. Therefore, the RSE model based on L-WLAV can be directly constructed:

[0194] minw t (u t +v t )

[0195]

[0196] Among them, w t is the measurement weight matrix of the thermal network, u t and v t Two types of non-negative variables are introduced for the thermal network, z t is the thermal network quantity measurement, x t is the state variable of the thermal network, H t The constant coefficient matrix of the thermal network has the following structure:

[0197]

[0198] S3-1-1 Construction of thermal network state estimation model based on thermal network L-WLAV model and node temperature constraints

[0199] The measurement redundancy of the thermal system itself is low (about 1.5). Therefore, when bad data is generated in the measurement, the state estimation result of the thermal system is easily affected. From the analysis in the first section, it can be seen that there are equality constraints (8) and (9) between the temperatures of the nodes. The measurement estimate obtained by the first layer state estimation is and Considered as a pseudo-measurement, the corresponding linear equality constraint is obtained and added to the model (25), resulting in:

[0200] minw t (u t +v t )

[0201]

[0202] At this point, the RSE model of the thermal system based on L-WLAV is completed.

[0203] Solution of the S3-1-2 second-layer thermal network state estimation model

[0204] The second-level RSE model (27) belongs to a linear programming model. The present invention uses CPLEX to solve. The state variable T in the thermal network can be directly obtained. s With T r estimated value.

[0205] The dual-layer robust state estimation model for IEHS essentially separates the hydraulic system from the thermal system in the heating network. First, the state estimation of the power system and the hydraulic system is performed to obtain the measured estimated values ​​of the pipeline flow and node injection flow, which are used as pseudo-measurements for the state estimation of the thermal system in the second layer. The specific solution process is as follows: Figure 3 shown.

[0206] Advantages of the two-layer robust state estimation model

[0207] 1) Make up for the loss of power system measurement redundancy

[0208] Assume that in the regional IEHS, the number of nodes in the power system is N e , the number of branches is B e ; The number of nodes in the heat network (including hydraulic and thermal networks) is N h , the number of pipelines is B h . and N e =B e -1, N h =B h -1.

[0209] The present invention first realizes the linearization of the power system measurement equation by introducing auxiliary variables. Compared with the nonlinear power system measurement model (13), the number of measurements of the linearized measurement model remains unchanged, but the number of state variables is increased from 2N e -1 increases to N e +2B e , that is, N is increased e -1, resulting in loss of measurement redundancy. By relaxing equation (22), we can obtain the second-order cone inequality constraint (23), the number of which is N e -1. Therefore, the introduction of the second-order cone constraint (23) can be regarded as an equivalent increase in the number of quantity measurements by N. e -1, thus compensating for the loss of measurement redundancy in some power systems.

[0210] 2) Increased measurement redundancy of the thermal system

[0211] When establishing a thermal network measurement model, the prior art only considers node thermal power measurement, node heating temperature measurement, and node heat recovery temperature measurement, which results in low measurement redundancy in the thermal network. The present invention considers the constraint relationship between node temperatures, and effectively increases the measurement redundancy of the thermal network. At the same time, when constructing the constraint between node temperatures, the present invention does not directly use the pipeline flow m and the node injection flow m. q Instead of using raw measurement data, a more reliable measurement estimate is obtained through the first-level state estimation. and Make the constraints more accurate.

[0212] 3) No initial value is required and it has good anti-error performance

[0213] Reference

[12] uses the SE method based on WLS to solve IEHS. Generally, WLS uses the Newton method to iteratively solve the nonlinear measurement equation. Therefore, WLS requires the initial value of the state variable to be given, and the initial value and the true value must be close enough. In the hydraulic network, it can be seen from formula (2) that due to the square root term The existence of makes it difficult to select the initial value of the state variable p. Since model (24) is linear, the two-layer state estimation of the present invention does not need to consider the initial value of the state variable. In addition, both layers of state estimation models are based on WLAV, so they have good recognition ability for bad data.

[0214] Example 2

[0215] The electric-thermal integrated energy system adopted by the present invention adopts an island operation mode. Among them, the relaxed node 1 in the modified IEEE 14-node power system is connected to the heat source node 31 in the Bali 32-node thermal system through a gas turbine, and the proportional coefficient c m1 =1.3; PV node 6 in the power system is connected to slack node 1 in the thermal system through a steam turbine, where Z = 8.1, P con =0.2pu; PV node 2 in the power system is connected to heat source node 32 in the thermal system through an internal combustion engine, and the proportional coefficient c m2 =1.266.

[0216] Test analysis under normal measurement

[0217] The measured value of IEHS in the present invention under normal conditions is composed of Gaussian noise added to the true value of the power flow. The standard deviation of the measurement noise of the power system and the thermal system is set to 10 -3 , where the true value of the IEHS power flow is obtained by the decoupled electric-thermal coupled system power flow calculation method in the literature

[15] . In this section, the proposed double-layer robust state estimation method is compared with the traditional nonlinear WLS

[12] , and the BRSE method proposed in the literature [9] were compared and analyzed.

[0218] The present invention selects the estimated error average value of the state variable The maximum estimation error δ of the state variables max , measurement error statistics S M , estimated error statistics S H and λ are used as indicators for state estimation performance analysis

[21] . They are expressed as:

[0219]

[0220]

[0221]

[0222]

[0223] λ=S H / S M(34)

[0224] Among them, x true 、 are the true value and estimated value of the state variable respectively; x i,true 、 x true 、 The true value and estimated value of the power flow calculation of the i-th state variable in z i,j 、h i (x true )and are the measured value, true value of the power flow calculation and the estimated value of the i-th quantity in the j-th experiment respectively; σ i is the standard deviation of the measurement error of the i-th quantity.

[0225] λ is used to evaluate the filtering effect of the state estimation method. The smaller λ is, the better the filtering effect is. m is the total number of measurements, T is the number of Monte Carlo simulation experiments, and the present invention conducts T=1000 Monte Carlo simulation experiments. The average values ​​of the estimated errors of the state variables of the power system and the thermal system obtained by WLS, BRSE and TL-RSE are as follows: Figure 4 、 5 As shown:

[0226] Depend on Figure 4 、 5 It can be concluded that: 1) For the power system, the average value of the estimation error of each state variable obtained by WLS is the smallest, while the estimation result of TL-RSE is better than BRSE; 2) For the node pressure head in the thermal system, the average value of the estimation error obtained by WLS is smaller, and for the node heating temperature T s and node reheating temperature T r , the average value of the estimation error obtained by TL-RSE is significantly smaller than the estimation results obtained by WLS and BRSE.

[0227] The performance analysis indicators (31) to (34) obtained by the three state estimation methods are shown in Tables 1 and 2 respectively. As can be seen from Table 1, in the power system, the maximum estimation error of the state variables U and θ obtained by WLS is the smallest among the three methods, while the estimation accuracy of TL-RSE is better than BRSE; in the thermal system, the maximum estimation error of the state variable p obtained by WLS is the smallest, and the state variable T obtained by TL-RSE is the smallest. s With T r The maximum error is the smallest, where T s The maximum error has been significantly reduced.

[0228] Table 1 Maximum estimation errors of state variables obtained by three state estimation methods

[0229]

[0230] It can be seen from Table 2 that the estimated error statistics S of the TL-RSE part H It is significantly smaller than WLS and BRSE, so the obtained λ is significantly reduced, which means that the filtering effect of TL-RSE is better than WLS and BRSE.

[0231] Table 2 Results of performance analysis indicators (31) to (34) obtained by three state estimation methods

[0232]

[0233] In summary, under normal measurement conditions: 1) For power systems, TL-RSE has higher estimation accuracy than BRSE, which is consistent with the theoretical analysis above; 2) For thermal systems, TL-RSE has significantly better estimation accuracy for node heating temperature and node heat recovery temperature than WLS and BRSE due to the consideration of physical constraints between node temperatures.

[0234] Robustness test analysis

[0235] The robustness tests in this section include tests for general bad data and tests for strongly correlated bad data. Since WLS itself is not robust, the analysis in this section adds the largest normalized residual (LNR) identification step after WLS, i.e., WLS+LNR. A comparative analysis is conducted between WLS+LNR, BRSE, and the dual-layer robust state estimation method proposed in this paper.

[0236] General bad data robustness test

[0237] General bad data include: 1) bad data without interaction, that is, the corresponding elements S(i,j) of bad data i and j in the residual sensitivity matrix S are ≈ 0; 2) bad data with interaction (S(i,j) is large), but the changes are inconsistent with each other.

[0238] The present invention sets general bad data in the power system and the thermal system for testing and analysis.

[0239] For the power system, the test results of the 10 bad data settings and the three state estimation methods are shown in Table 3, where (P 10 ,P 10-11 ) and (P 14 ,P 13-14 ) belongs to the second type of general bad data. The specific identification process of WLS+LNR is shown in Table 4.

[0240] Table 3 Robustness test results of three state estimation methods in power systems

[0241]

[0242]

[0243] Table 4. Identification process of general bad data in power system by WLS+LNR

[0244]

[0245] In Table 4, z ide Represents the standardized residual r obtained for each identification N,max The largest quantity is measured and removed before the next WLS. As can be seen from Tables 3 and 4, WLS+LNR successfully identified the 10 general bad data through the first 10 identifications. At the same time, BRSE and TL-RSE were not affected by the bad data and obtained measurement estimates close to the true values.

[0246] For the thermal system, the test results of 15 bad data settings and three state estimation methods are shown in Table 5, where (m q3 ,Φ3) and (m q18 ,Φ 18 ) belongs to the second category of general bad data. Table 6 describes the specific identification process of WLS+LNR. The shaded area in Table 5 represents the estimated value that is significantly different from the true value, and the shaded area in Table 6 represents the incorrectly identified ground quantity measurement.

[0247] It can be seen that: 1) For WLS+LNR, in the first 14 identifications, WLS+LNR mistakenly measured Φ1, T r3 and Φ2 were treated as bad data and eliminated, so in Table 5, the bad data Φ3, T r1 With T r1 The estimated value deviates greatly from the true value; 2) For BRSE, when bad data occurs in the node temperature measurement, a large error will occur between the corresponding estimated result and the true value; 3) For TL-RSE, it can effectively resist the general bad data that appears in the heating network.

[0248] Table 5 Robustness test results of three state estimation methods in thermal systems

[0249]

[0250]

[0251]

[0252] Table 6 WLS+LNR identification process for general bad data in thermal system

[0253]

[0254]

[0255] Robustness test for strongly correlated bad data

[0256] Strongly correlated bad data refers to bad data that interact closely with each other and change consistently. 12 With P 15 Set as bad data, and measure P 21 With P 51 The bad data setting and the estimation results of BRSE and TL-RSE are shown in Table 7. The specific identification process of WLS+LNR is shown in Table 8.

[0257] Table 7 Various statistical values ​​obtained by three state estimation methods

[0258]

[0259] Table 8 Various statistical values ​​obtained by three state estimation methods

[0260]

[0261] As shown in Tables 7 and 8, BRSE and TL-RSE can identify the strongly correlated bad data in IEHS, while WLS+LNR can only identify the bad data P in the second time in the first five identifications. 15 Therefore, WLS+LNR cannot identify the strongly correlated bad data in IEHS.

[0262] Example 3:

[0263] Based on the same inventive concept, the present invention also provides a two-layer robust state estimation system for an electric-thermal integrated energy system, such as Figure 2 As shown, including:

[0264] A data acquisition module is used to obtain power system quantity measurements and thermal system quantity measurements of the electric and thermal integrated energy system; wherein the thermal system includes: a hydraulic network and a thermal network, and the thermal system quantity measurements include hydraulic network quantity measurements and thermal network quantity measurements;

[0265] A power system hydraulic network state estimation module is used to input the power system quantity measurements and hydraulic network quantity measurements into a pre-built power system hydraulic network state estimation model to obtain power system state variable node voltage amplitude estimation values, node phase angle estimation values, hydraulic network state variable node pressure head estimation values, hydraulic network branch flow estimation values, and node injection flow estimation values;

[0266] a thermal network state estimation module, configured to input the hydraulic network branch flow estimate and node injection flow estimate, as well as thermal network quantity measurements, into a pre-built thermal network state estimation model to obtain thermal network state variables, namely, node heating temperature estimate and node heat return temperature estimate;

[0267] The thermal network state estimation model is constructed by considering temperature constraints and thermal power measurements between thermal network nodes.

[0268] Preferably, the building blocks of the power system hydraulic network state estimation model include:

[0269] A power system linear measurement equation building module, configured to determine a power system linear measurement equation based on a set first auxiliary state variable and a first auxiliary quantity measurement;

[0270] A hydraulic network linear measurement equation construction module, configured to determine a hydraulic network linear measurement equation based on a set second auxiliary state variable and a second auxiliary quantity measurement;

[0271] A unified linear measurement model construction module for the power system and the hydraulic network, which is used to construct a unified linear measurement model for the power system and the hydraulic network based on the linear measurement equations of the power system, the linear measurement equations of the hydraulic network, and the coupling mode of the coupling nodes of the power system and the thermal system;

[0272] A module for constructing a unified linear weighted minimum absolute value state estimation model for the power system and the water network, which is used to construct a unified linear weighted minimum absolute value state estimation model for the power system and the water network based on the unified linear measurement model of the power system and the water network;

[0273] The power system hydraulic network state estimation model construction module is used to construct the power system hydraulic network state estimation model based on the relationship between the auxiliary state variables of the power system and the unified linear weighted minimum absolute value state estimation model of the power system and the hydraulic network.

[0274] Among them, the first auxiliary state variable is set by the power system state variable; the first auxiliary quantity measurement is set by the power system quantity measurement; the second auxiliary state variable is set by the hydraulic network state variable; and the second auxiliary quantity measurement is set by the hydraulic network quantity measurement.

[0275] Preferably, the power system hydraulic network state estimation module includes:

[0276] Calculation module 1, used for inputting the power system quantity measurement and the water network quantity measurement into a pre-built power system water network state estimation model to calculate the first auxiliary state variable estimation value and the second auxiliary state variable estimation value;

[0277] Calculation module 2, configured to calculate an estimated value of a phase angle difference of a power system branch, an estimated value of a node voltage amplitude, and an estimated value of a pressure head loss of a hydraulic network pipeline based on the first and second estimated values ​​of the auxiliary state variables through nonlinear transformation;

[0278] Calculation module 3, for calculating the estimated phase angle value of the power system node and the estimated pressure head value of the hydraulic network node based on the estimated phase angle difference value of the power system branch and the estimated pressure head loss value of the hydraulic network pipeline through linear transformation;

[0279] The calculation module 4 is used to calculate the estimated values ​​of the hydraulic network branch flow and the node injection flow based on the estimated values ​​of the hydraulic network node pressure head.

[0280] Preferably, the building blocks of the thermal network state estimation model include:

[0281] Thermal network linear measurement equation construction module, used to construct thermal network linear measurement equations based on thermal network state variables and thermal network quantity measurements;

[0282] A state estimation model construction module of a thermal network with a linear weighted minimum absolute value, configured to construct a state estimation model of a thermal network with a linear weighted minimum absolute value based on the thermal network linear measurement equation;

[0283] A constraint construction module is used to set the estimated flow rates of hydraulic network branches and node injection flow rates as pseudo-quantity measurements and determine the temperature constraints between nodes in the thermal network;

[0284] The thermal network state estimation model construction module is used to construct a thermal network state estimation model based on the linear weighted minimum absolute value state estimation model of the thermal network and the constraints.

[0285] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0286] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0287] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0288] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0289] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit its scope of protection. Although the present application has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that after reading this application, those skilled in the art may still make various changes, modifications or equivalent substitutions to the specific implementation methods of the application, but these changes, modifications or equivalent substitutions are all within the scope of protection of the pending claims of the application.

Claims

1. A two-layer robust state estimation method for an electric-thermal integrated energy system, characterized in that: include: Obtaining power system quantity measurements and thermal system quantity measurements of the electric and thermal integrated energy system; The thermal system includes: a hydraulic network and a thermal network, and the thermal system quantity measurement includes hydraulic network quantity measurement and thermal network quantity measurement; Input the power system quantity measurements and the water network quantity measurements into a pre-built power system water network state estimation model to obtain power system state variable node voltage amplitude estimation values, node phase angle estimation values, water network state variable node pressure head estimation values, water network branch flow estimation values, and node injection flow estimation values; Input the estimated values ​​of branch flow of the hydraulic network, the estimated values ​​of node injection flow, and the measured values ​​of thermal network quantities into a pre-built thermal network state estimation model to obtain the estimated values ​​of node heating temperature and node heat recovery temperature of the thermal network state variables; The thermal network state estimation model is constructed by considering the temperature constraints and thermal power measurements between the thermal network nodes; The construction of the power system hydraulic network state estimation model includes: Determining a linear measurement equation for the power system based on a set first auxiliary state variable and a first auxiliary quantity measurement; determining a linear measurement equation for the hydraulic network based on a set second auxiliary state variable and a second auxiliary quantity measurement; Based on the linear measurement equations of the power system, the linear measurement equations of the hydraulic network, and the coupling mode of the coupling nodes of the power system and the thermal system, a unified linear measurement model of the power system and the hydraulic network is constructed; Based on the unified linear measurement model of the power system and the water network, a unified linear weighted minimum absolute value state estimation model of the power system and the water network is constructed; Constructing a power system and hydraulic network state estimation model based on the relationship between the auxiliary state variables of the power system and the unified linear weighted minimum absolute value state estimation model of the power system and the hydraulic network; Among them, the first auxiliary state variable is set by the power system state variable; the first auxiliary quantity measurement is set by the power system quantity measurement; the second auxiliary state variable is set by the hydraulic network state variable; and the second auxiliary quantity measurement is set by the hydraulic network quantity measurement.

2. The method according to claim 1, wherein The first auxiliary state variable and the first auxiliary quantity are measured as follows: in is the first auxiliary state variable, Vi a 、 The first auxiliary quantity introduced is the first auxiliary quantity measurement, U i is the voltage amplitude of node i, P i is the injected active power of node i, Q i is the reactive power injected into node i, P ij is the active power of branch ij, Qij is the reactive power of branch ij; The calculation formula of the first auxiliary amount is as follows: Among them, U j The voltage amplitude at node j, θ ij is the phase angle difference between node i and node j.

3. The method according to claim 1, wherein The linear measurement equation of the power system is as follows: Where U i The voltage amplitude at node i, P i is the injected active power of node i, Q i is the reactive power injected into node i, P ij is the active power of branch ij, Q ij is the reactive power of branch ij, Ni is the number of nodes in the power system, V i a 、 is the first auxiliary quantity, g si is the equivalent resistance to ground of node i, b si is the equivalent reactance to ground of node i, g ij is the equivalent resistance of branch ij, b ij is the equivalent reactance of branch ij, G ij Determined by the equivalent resistance of branch ij, B ij Determined by the equivalent reactance of branch ij.

4. The method according to claim 1, wherein The second auxiliary state variable and the second auxiliary quantity are measured as follows: in, is the second auxiliary state variable, α a is the second auxiliary quantity, Second auxiliary quantity measurement, is the water flow of branch ij, Inject water flow into the node; The α a The calculation formula of the second auxiliary amount is as follows: in, is α a The elements in p ij is the pipeline pressure head loss, s ij Determined by the relationship between the pressure head at node i and the pressure head at node j.

5. The method according to claim 1, wherein The hydraulic network linear measurement equation is as follows: Where, is the water flow of branch ij, Inject water flow into the node, K ij is the pipeline impedance coefficient of branch ij, p ij is the pipeline pressure head loss, s ij Determined by the relationship between the pressure head at node i and the pressure head at node j, is the second auxiliary amount α a Elements in .

6. The method according to claim 1, wherein The unified linear measurement model of the power system and water network is shown in the following formula: Where x a is the unified auxiliary state variable of the power system and hydraulic network, z a For unified auxiliary quantity measurement of power system and water network, is the first auxiliary state variable, is the second auxiliary state variable, is the first auxiliary quantity measurement, is the second auxiliary quantity measurement, is the heat energy generated by gas turbine or internal combustion engine coupling, N1 is the number of nodes coupled with gas turbine or internal combustion engine, is the heat energy generated by the steam turbine coupling mode, N2 is the number of nodes using the steam turbine coupling mode, H a is the unified constant coefficient matrix of the power system and the hydraulic network, r a It is the unified measurement error of the power system and water network.

7. The method according to claim 1, wherein The unified linear weighted minimum absolute value state estimation model of the power system and the hydraulic network is shown in the following formula: minw(u+v) Where w is the unified measurement weight matrix of the power system and the hydropower network, u and v are two types of non-negative variables introduced by the unified model of the power system and the hydropower network, za is the unified auxiliary quantity measurement of the power system and the hydropower network, Ha is the unified constant coefficient matrix of the power system and the hydropower network, and xa is the unified auxiliary state variable of the power system and the hydropower network.

8. The method according to claim 1, wherein The state estimation model of the power system and the hydraulic network is constructed based on the relationship between the auxiliary state variables of the power system and the unified linear weighted minimum absolute value of the power system and the hydraulic network, including: Construct the relationship between auxiliary state variables of the power system; Constructing a second-order cone inequality constraint based on a relationship between auxiliary state variables of the power system; Based on the second-order cone inequality constraints and the state of the unified linear weighted minimum absolute value of the power system and the hydraulic network Estimation model, build the power system hydraulic network state estimation model.

9. The method according to claim 8, wherein The second-order cone inequality constraint is shown as follows: Where, Vi a 、Vj a 、 is the first auxiliary amount.

10. The method according to claim 1, wherein The power system hydraulic network state estimation model is as follows: minΣλ ij R ij -w(u+v) Where w is the unified measurement weight matrix of the power system and the water network, u and v are two types of non-negative variables introduced by the unified model of the power system and the water network, and z a For unified auxiliary quantity measurement of power system and water network, H a is the unified constant coefficient matrix of the power system and the hydraulic network, x a is the unified auxiliary state variable of the power system and hydraulic network, V i a 、V j a 、 is the first auxiliary quantity, λ ij It is an adjustment parameter whose size is equal to the weight value of the power measurement of the corresponding branch ij in the power system.

11. The method according to claim 1, wherein The power system quantity measurement and the hydraulic network quantity measurement are input into a pre-built power system hydraulic network state estimation model to obtain power system state variable node voltage amplitude estimation value, node phase angle estimation value and hydraulic network state variable node pressure head estimation value, hydraulic network branch flow estimation value and node injection flow estimation value, including: Input the power system quantity measurement and the water network quantity measurement into a pre-built power system water network state estimation model to calculate the first auxiliary state variable estimation value and the second auxiliary state variable estimation value; Based on the first auxiliary state variable estimate and the second auxiliary state variable estimate, after nonlinear transformation, calculate the power system branch phase angle difference estimate, the node voltage amplitude estimate and the hydraulic network pipeline pressure head loss estimate; Based on the estimated values ​​of the phase angle differences of the power system branches and the estimated values ​​of the pressure head losses of the hydraulic network pipelines, the estimated values ​​of the phase angles of the power system nodes and the estimated values ​​of the pressure heads of the hydraulic network nodes are calculated through linear transformation; Based on the estimated value of the hydraulic network node pressure head, the estimated value of the hydraulic network branch flow and the node injection flow is calculated.

12. The method according to claim 1, wherein The construction of the thermal network state estimation model includes: Based on the thermal network state variables and thermal network quantity measurements, the thermal network linear measurement equation is constructed; Based on the thermal network linear measurement equation, a linear weighted minimum absolute value state estimation model of the thermal network is constructed; The estimated flow rates of hydraulic network branches and node injection flow rates are set as pseudo-measurements, and the temperature constraints between nodes in the thermal network are determined. A thermal network state estimation model is constructed based on the linear weighted minimum absolute value state estimation model of the thermal network and the constraints.

13. The method according to claim 12, wherein: The thermal network state variables and thermal network quantity measurements are as follows: xt=[Ts;Tr] zt=[φ;Ts;Tr] Among them, x t is the state variable of the thermal network, T s is the node heating temperature, T r is the node reheating temperature; z t is the thermal network quantity measurement, and φ is the node thermal power.

14. The method according to claim 12, wherein: The thermal network linear measurement equation is as follows: Where, φ i is the thermal power of node i, is the injection water flow rate of node i, Cp is the specific heat capacity of water, T si is the heating temperature of node i, T ri is the reheating temperature of node i.

15. The method according to claim 12, wherein The linear weighted minimum absolute value state estimation model of the thermal network is as follows: minw t (u t +v t ) Where w t is the measurement weight matrix of the thermal network, u t and v t Two types of non-negative variables are introduced for the thermal network, z t is the thermal network quantity measurement, x t is the state variable of the thermal network, H t Constant coefficient matrix of the thermal network.

16. The method according to claim 12, wherein The constant coefficient matrix H of the thermal network t It is expressed as follows: Among them, C p is the specific heat capacity of water, Inject water flow estimates into nodes.

17. The method according to claim 12, wherein The temperature constraints between the nodes of the thermal network are as follows: Where, T end is the pipe end temperature, T start is the temperature of the node at the beginning of the pipeline, T a is the ambient temperature, λ h is the heat transfer coefficient per unit length of the pipeline, L is the length of the pipeline, C p is the specific heat capacity of water, is the estimated value of the injected water flow at node i, is the estimated total flow of a pipe outflowing the node, is the estimated flow rate of a pipeline injected into the node, T out is the node mixing temperature, T in is the node temperature at the beginning of the pipeline.

18. The method according to claim 12, wherein The thermal network state estimation model is shown in the following formula: minwt(ut+vt) Where w t is the measurement weight matrix of the thermal network, u t and v t Two types of non-negative variables are introduced for the thermal network, z t is the thermal network quantity measurement, x t is the state variable of the thermal network, H t The constant coefficient matrix of the thermal network, T end is the pipe end temperature, T start is the temperature of the node at the beginning of the pipeline, T a is the ambient temperature, λ h is the heat transfer coefficient per unit length of the pipeline, L is the length of the pipeline, C p is the specific heat capacity of water, is the estimated value of the injected water flow at node i, is the estimated total flow of a pipe outflowing the node, is the estimated flow rate of a pipeline injected into the node, T out is the node mixing temperature, T in is the node temperature at the beginning of the pipeline.

19. A double-layer robust state estimation system for an electric and thermal integrated energy system, characterized in that: include: Data acquisition module, used to obtain power system quantity measurements and thermal system quantity measurements of the electric and thermal integrated energy system; The thermal system includes: a hydraulic network and a thermal network, and the thermal system quantity measurement includes hydraulic network quantity measurement and thermal network quantity measurement; A power system hydraulic network state estimation module is used to input the power system quantity measurements and hydraulic network quantity measurements into a pre-built power system hydraulic network state estimation model to obtain power system state variable node voltage amplitude estimation values, node phase angle estimation values, hydraulic network state variable node pressure head estimation values, hydraulic network branch flow estimation values, and node injection flow estimation values; a thermal network state estimation module, configured to input the hydraulic network branch flow estimate and node injection flow estimate, as well as thermal network quantity measurements, into a pre-built thermal network state estimation model to obtain thermal network state variables, namely, node heating temperature estimate and node heat return temperature estimate; The thermal network state estimation model is constructed by considering the temperature constraints and thermal power measurements between the thermal network nodes; The construction of the power system hydraulic network state estimation model includes: Determining a linear measurement equation for the power system based on a set first auxiliary state variable and a first auxiliary quantity measurement; determining a linear measurement equation for the hydraulic network based on a set second auxiliary state variable and a second auxiliary quantity measurement; Based on the linear measurement equations of the power system, the linear measurement equations of the hydraulic network, and the coupling mode of the coupling nodes of the power system and the thermal system, a unified linear measurement model of the power system and the hydraulic network is constructed; Based on the unified linear measurement model of the power system and the water network, a unified linear weighted minimum absolute value state estimation model of the power system and the water network is constructed; Constructing a power system and hydraulic network state estimation model based on the relationship between the auxiliary state variables of the power system and the unified linear weighted minimum absolute value state estimation model of the power system and the hydraulic network; Among them, the first auxiliary state variable is set by the power system state variable; the first auxiliary quantity measurement is set by the power system quantity measurement; the second auxiliary state variable is set by the hydraulic network state variable; and the second auxiliary quantity measurement is set by the hydraulic network quantity measurement.

Images