Electric power system operation state estimation method based on quantum computing

Abstract

The invention provides an electric power system operation state estimation method based on quantum computing, relates to the field of electric power system operation analysis and state awareness, and aims to solve the electric power system operation state through constructing an electric power system operation state estimation model based on an unconstrained quadratic binary optimization model (QUBO model) and realizing solution through a light quantum computer. And obtaining an electric power system operation state estimation result. According to the method, the electric power system operation state estimation problem is converted into a QUBO model and solved, a traditional calculation algorithm of the state estimation problem is converted into a quantum calculation algorithm, and a model normal form and a feasible algorithm are provided for constructing quantum calculation of the electric power system operation state estimation problem. According to the algorithm, the quantum calculation is introduced into the state estimation of the power system, and the quantum calculation can meet the requirement of higher calculation power in the future and meet the requirement of shorter response time.

Description

Technical Field

[0001] The present invention relates to the field of power system operation analysis and state perception, and specifically to a power system operation state estimation method based on quantum computing. Background Art

[0002] The new power system will include a high proportion of new energy access and a high proportion of power electronic equipment, which will bring high uncertainty and bring challenges to the safe and stable operation and dispatch of the power system. Power system state estimation is the basis of power system operation control. The new power system requires a shorter response time for power system state estimation, so the demand for computing power has increased sharply. Quantum computing is a new form of computing that uses quantum mechanical laws to control quantum mechanical units to achieve calculations. Compared with classical computers, quantum computers use quantum bits as computing units. Based on the characteristics of quantum superposition, they can achieve exponential computing power of classical computers with the same number of bits. Existing research on the use of quantum computing for power system state estimation is based on quantum circuits, and the HHL algorithm is used to solve linear equations. The hardware basis of this algorithm is a general-purpose quantum computer. Today, the number of available quantum bits is growing slowly, and it is difficult to break through technical barriers in a short time. The coherent Ising machine is a special optical quantum computer for solving optimization problems. It is expected to achieve higher quantum bit growth in a relatively short period of time and achieve commercialization. Summary of the invention

[0003] In view of the fact that general-purpose quantum computers are difficult to realize in the short term, while special-purpose optical quantum computers can be commercialized in a relatively short period of time, the present invention proposes a method for estimating the operating state of a power system based on quantum computing, with a special-purpose optical quantum computer (coherent Ising machine) as the hardware support, so as to meet the requirements of computing power for state estimation of future new power systems.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] The present invention is a method for estimating the operating state of an electric power system based on quantum computing. The hardware-based coherent Ising machine of the present invention can realize the efficient solution of the Ising model. The Ising model and the QUBO model (unconstrained quadratic binary optimization model) are equivalent and can be converted to each other. The least squares algorithm (WLS) is the most widely used state estimation algorithm at present. The Newton iteration method is traditionally used to solve the WLS problem. The present invention converts the solution equation in the Newton iteration method into the solution equation of the QUBO model, realizes the conversion of the traditional algorithm into the quantum computing algorithm, and provides a construction paradigm for the quantum algorithm of the electric power system state estimation.

[0006] The coherent Ising machine is a hybrid quantum computing system consisting of two subsystems: an optical system and an electrical system. The optical system is responsible for the preparation and storage of optical quantum bits. First, a femtosecond fiber laser (Pulsed laser) with a frequency of 100MHz is used to generate laser pulses. Since the output power of the laser pulse is low (100mW), it is necessary to amplify the power through an erbium-doped fiber amplifier (EDFA). The amplified laser is then doubled in frequency to a 780nm laser through a periodically planned lithium niobate (PPLN1) crystal. Subsequently, the frequency-doubled light is used as a pump light source for a synchronously pumped phase-sensitive amplifier, and then converted into a 1560nm signal light through a PPLN2 crystal, and finally a degenerate optical parametric oscillator (DOPO) is formed to generate optical pulses with specific phases and amplitudes. These optical pulses are stored in a fiber cavity as optical quantum bits and used for subsequent calculations. The electrical system is responsible for the control and calculation of the optical quantum bits. First, the Ising matrix of the problem to be solved is downloaded to the programmable logic gate array (FPGA) through the host computer. After the FPGA obtains the phase and amplitude information of the optical particle bit through a balanced homodyne detector, it calculates the feedback signal and modulates the feedback light through the intensity modulator (IM) and the phase modulator (PM). The feedback light and the optical pulses in the optical fiber loop interfere with each other, causing the optical quantum bit to evolve in the direction of the lowest Hamiltonian of the Ising problem. After reaching the lowest Hamiltonian, the phase information of the optical quantum bit is the final solution to the Ising problem.

[0007] The Ising model is a random model that describes physical phase transitions. The mathematical form of the Ising model is as follows: Where H represents the Hamiltonian, σ is the spin variable to be determined, whose value is ±1, and J and h are the coefficients of the quadratic term and the linear term, respectively.

[0008] The QUBO model is an unconstrained quadratic binary optimization model. The mathematical modeling of the QUBO model is as follows: Where Γ={1,2,3,…,k} is the set of subscripts of variable x, representing a total of k 0-1 variables, β is the coefficient of the quadratic term, and α is the coefficient of the linear term.

[0009] The power system state estimation refers to: given the measurement vector z (such as node voltage, node injection power, line power flow, m×1) and the measurement function h (essentially a mapping of power flow equations, generally a nonlinear equation group), considering that the measurement itself has noise, the relationship between the distribution network state variable x (such as node voltage amplitude and phase angle, n×1) and the measurement vector z can be established: z=h(x)+e. The redundant measurement values ​​about the operating state are obtained as input through the measurement device, and the voltage amplitude and phase angle are output as the state variables to be determined, and the estimated value that is closest to the current operating state of the system is solved.

[0010] The weighted least squares algorithm (WLS) is: This method assumes that the measurement error obeys The normal distribution of the measured value and the estimated value of the measured value is minimized as the estimation criterion. Based on the state estimation model, given the measurement vector z and the measurement function h, the objective function constructed according to the weighted least squares criterion is as follows: min J(x) = [zh(x)] T W[zh(x)]

[0011] Where: W is the weighted coefficient matrix, which represents the confidence of the measurement. Its specific construction form is a diagonal matrix as shown below:

[0012] The Newton iteration method means that since the measurement h(x) in the model is a nonlinear function, x cannot be expressed analytically, h(x) needs to be Taylor expanded, the higher-order terms are omitted to make it linear, and the Newton method is used to iteratively solve the WLS problem. At this time, for the kth iteration, the objective function can be written as follows:

[0013] J(Δx)=[Δz k -H(x k )Δx k ] T W[Δz k -H(x k )Δx k ] (1)

[0014] in,

[0015]

[0016] Δz k =z k -h(x k ) (3)

[0017] By expanding formula (3), we can get its iteration direction

[0018] x k+1 =x k +Dxk (4)

[0019] G(x k )Dx k =H(x k ) T WΔz k (5)

[0020] in,

[0021] G(x k )=H(x k ) T WH(x k ) (6)

[0022] Therefore, the k-th iteration result is:

[0023] Δx k =G(k) -1 H(x k ) T W[zh(x k )] (7)

[0024] Where: H(x k ) is the value of h and x k The Jacobian matrix of .

[0025] The method for estimating the operating state of a power system based on quantum computing of the present invention comprises the following steps:

[0026] Step S1: Obtain the measured data of the voltage and power of each node, preset the measurement standard deviation, and construct the weighted coefficient matrix W. Initialize the node voltage and phase angle, and preset the step value for jumping out of iteration.

[0027] Step S2: Calculate the measurement function h(x) and the Jacobian matrix model H(x) according to the initial voltage phase angle.

[0028] Step S3: Determine the binary expression matrix by the preset model accuracy (number of bits) Find the Q matrix for the QUBO model.

[0029] Step S4: Input the Q matrix into CIM (simulated annealing / GUROBI is also used for verification later), set the solution parameters, obtain the solution vector, and the binary expression matrix The product of this vector is the state quantity correction vector, and the estimated values ​​of the node voltage and phase angle are corrected according to formula (25).

[0030] Step S5: Determine whether the maximum absolute value of the iteration step is less than a preset value. If it is greater than the preset value, return to step 2 and continue the next iteration. If it is less than the preset value, exit the loop to obtain the final state estimation node voltage value and phase angle value.

[0031] In step S1, the initialized voltage per unit value is generally 1, the initialized phase angle per unit value is generally 0, and W is a weighting coefficient matrix that represents the confidence of the measurement. Its construction method is consistent with that in WLS:

[0032]

[0033] Where σ is the standard deviation of a single measurement, representing the confidence level of that measurement.

[0034] In step S2, the measurement function of the distribution network is as follows:

[0035] 1) Node voltage measurement model

[0036]

[0037] 2) Node injection active power measurement model

[0038]

[0039] 3) Node injection reactive power measurement model

[0040]

[0041] 4) Line transmission active power measurement model

[0042]

[0043] 5) Line transmission reactive power measurement model

[0044]

[0045] The measurement function h(x) in step S2 is derived to have the corresponding Jacobian matrix in the form shown below:

[0046]

[0047] Among them, the partial derivative results of each measurement function are as follows:

[0048] 6) Node voltage measurement model

[0049]

[0050] 7) Node injection active power measurement model

[0051]

[0052] 8) Node injection reactive power measurement model

[0053]

[0054] 9) Line transmission active power measurement model

[0055]

[0056] 10) Line transmission reactive power measurement model

[0057]

[0058]

[0059] The symbols in the above formula are explained as follows:

[0060] 1) Subscript i: node number;

[0061] 2) Superscript z: indicates that the value is a measured value;

[0062] 3) V i : voltage amplitude of node i (pu);

[0063] 4)θ i : voltage phase angle value of node i (degree);

[0064] 5)θ ij : The phase angle difference between node i and node j (degree);

[0065] 6)P i : injected active power of node i (pu);

[0066] 7) Q i : injected reactive power of node i (pu);

[0067] 8)P ij : The transmitted active power of the line from node i to node j (pu);

[0068] 9) Q ij : The transmitted reactive power (pu) of the line from node i to node j;

[0069] 10) G ij : admittance matrix conductance;

[0070] 11) B ij : admittance matrix susceptance;

[0071] 12) Self-susceptance of node i.

[0072] In step S3, the binary expression matrix Convert continuous variables into 0-1 binary variables. For continuous variables w, express the continuous variable w through binary expression as follows:

[0073]

[0074] Where: is a matrix of n×nK; It is an nK×1 vector composed of 0-1 binary variables, where n is the number of independent variables and the value of K determines the accuracy of the binary expression.

[0075] In step S3, the transformation of the QUBO model is derived as follows:

[0076] The solution of each iteration in Newton iteration is in the form of:

[0077] Δx k =[H(x k ) T WH(x k )] -1 H(x k ) T W[zh(x k )] (26)

[0078] For the following least squares regression problem:

[0079]

[0080] Its analytical solution is:

[0081] w=(X T X) -1 X T Y (28)

[0082] Therefore, let have to,

[0083]

[0084] make:

[0085] X 1 =W 1 H(x k ),Y 1 =W 1 [zh(x k )] (30)

[0086] The solution in Newton iteration can be written in the form of the following least quadratic solution:

[0087] Δx k =(X 1 T X 1 ) -1 X 1 TY 1 (31)

[0088] Combining the binary expansion of equation (25) and equation (27), each iterative solution can be regarded as solving the following QUBO model:

[0089]

[0090] Then the corresponding Q matrix is:

[0091]

[0092] Step S4 uploads the Q matrix to the quantum cloud platform, and calls the optical quantum computer to solve the QUBO model, returns the solution vector, and obtains the iteration vector according to formula (25), and updates the state vector, including the state estimation values ​​of voltage and phase angle.

[0093] Step S5 determines whether the maximum iteration step of the iteration vector in S4 is less than a preset value. If it is less than the preset value, it is considered that the model has converged, otherwise, the iteration continues.

[0094] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention proposes a method for estimating the operating state of a power system based on quantum computing, and provides a construction paradigm for the evolution of future state estimation quantum algorithms. Compared with traditional algorithms, this algorithm introduces quantum computing into power system state estimation, and quantum computing will be able to meet higher computing power requirements and shorter response time requirements in the future. Compared with existing research based on quantum circuits, the present invention is based on a dedicated optical quantum computer and is expected to be applied and commercialized in a shorter time. BRIEF DESCRIPTION OF THE DRAWINGS

[0095] Figure 1 It is a flow chart of the present invention;

[0096] Figure 2 This is a comparison chart of voltage state estimation results;

[0097] Figure 3 This is a comparison chart of the phase angle state estimation results; DETAILED DESCRIPTION

[0098] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0099] Step S1: Obtain the measured data of the voltage and power of each node, preset the measurement standard deviation, and construct the weighted coefficient matrix W. Initialize the node voltage and phase angle, and preset the step value for jumping out of iteration.

[0100] Step S2: Calculate the measurement function h(x) and the Jacobian matrix model H(x) according to the initial voltage phase angle.

[0101] Step S3: Determine the binary expression matrix by the preset model accuracy (number of bits) Find the Q matrix for the QUBO model.

[0102] Step S4: Input the Q matrix into CIM (simulated annealing / GUROBI is also used for verification later), set the solution parameters, obtain the solution vector, and the binary expression matrix The product of this vector is the state quantity correction vector, and the estimated values ​​of the node voltage and phase angle are corrected according to formula (25).

[0103] Step S5: Determine whether the maximum absolute value of the iteration step is less than a preset value. If it is greater than the preset value, return to step 2 and continue the next iteration. If it is less than the preset value, exit the loop to obtain the final state estimation node voltage value and phase angle value.

[0104] In step S1, the initialized voltage per unit value is generally 1, the initialized phase angle per unit value is generally 0, and W is a weighting coefficient matrix that represents the confidence of the measurement. Its construction method is consistent with that in WLS:

[0105]

[0106] Where σ is the standard deviation of a single measurement, representing the confidence level of that measurement.

[0107] In step S2, the measurement function of the distribution network is as follows:

[0108] 1) Node voltage measurement model

[0109] V i z =V i (35)

[0110] 2) Node injection active power measurement model

[0111]

[0112] 3) Node injection reactive power measurement model

[0113]

[0114] 4) Line transmission active power measurement model

[0115]

[0116] 5) Line transmission reactive power measurement model

[0117]

[0118] The measurement function h(x) in step S2 is derived to have the corresponding Jacobian matrix in the form shown below:

[0119]

[0120] Among them, the partial derivative results of each measurement function are as follows:

[0121] 11) Node voltage measurement model

[0122]

[0123] 12) Node injection active power measurement model

[0124]

[0125] 13) Node injection reactive power measurement model

[0126]

[0127]

[0128] 14) Line transmission active power measurement model

[0129]

[0130] 15) Line transmission reactive power measurement model

[0131]

[0132] The symbols in the above formula are explained as follows:

[0133] 1) Subscript i: node number;

[0134] 2) Superscript z: indicates that the value is a measured value;

[0135] 3) V i : voltage amplitude of node i (pu);

[0136] 4)θ i : voltage phase angle value of node i (degree);

[0137] 5)θ ij : The phase angle difference between node i and node j (degree);

[0138] 6)P i : injected active power of node i (pu);

[0139] 7) Q i : injected reactive power of node i (pu);

[0140] 8)P ij : The transmitted active power of the line from node i to node j (pu);

[0141] 9) Q ij : The transmitted reactive power (pu) of the line from node i to node j;

[0142] 10) G ij : admittance matrix conductance;

[0143] 11) B ij : admittance matrix susceptance;

[0144] 12) Self-susceptance of node i.

[0145] In step S3, the binary expression matrix Convert continuous variables into 0-1 binary variables. For continuous variables w, express the continuous variable w through binary expression as follows:

[0146]

[0147] Where: is a matrix of n×nK; It is an nK×1 vector composed of 0-1 binary variables, where n is the number of independent variables and the value of K determines the accuracy of the binary expression.

[0148] In step S3, the transformation of the QUBO model is derived as follows:

[0149] The solution of each iteration in Newton iteration is in the form of:

[0150]

[0151] For the following least squares regression problem:

[0152]

[0153] Its analytical solution is:

[0154]

[0155] Therefore, let have to,

[0156]

[0157] make:

[0158] X 1 =W 1 H(x k ),Y 1 =W 1 [zh(x k )] (63)

[0159] The solution in Newton iteration can be written in the form of the following least quadratic solution:

[0160] Δx k =(X 1 T X 1 ) -1 X 1 T Y 1 (64)

[0161] Combining the binary expansion of equation (25) and equation (27), each iterative solution can be regarded as solving the following QUBO model:

[0162]

[0163] Then the corresponding Q matrix is:

[0164]

[0165] Step S4 uploads the Q matrix to the quantum cloud platform, and calls the optical quantum computer to solve the QUBO model, returns the solution vector, and obtains the iteration vector according to formula (25), and updates the state vector, including the state estimation values ​​of voltage and phase angle.

[0166] Step S5 determines whether the maximum iteration step of the iteration vector in S4 is less than a preset value. If it is less than the preset value, it is considered that the model has converged, otherwise, the iteration continues.

[0167] After specific actual experiments, the system used for simulation testing in the present invention is an adapted 4-node distribution network system, which is adapted from the case4_dist.m system provided in MATPOWER, removing the power supply at the original bus numbered 400 to make it conform to the power flow equation of the distribution network.

[0168] Measurement data description: The bus number indicates the measurement location of the variable; the true value is the theoretical value obtained through power flow calculation; the standard deviation is used to describe the error between the measured value and the true value due to factors such as sensor delay and accuracy error, and the standard deviation is used to construct the weighting coefficient matrix in the WLS model; the measured value is the simulated measurement data obtained by random sampling of the normal distribution based on the true value and the standard deviation; the error is the numerical difference between the measured value and the true value.

[0169] Calculation result data description: The residual is the absolute maximum value of the voltage and phase angle update during each iteration. When the residual is very small, it means that the iteration result has approached the optimal value. When the residual is lower than the preset value, the iteration can be exited to obtain the final state estimation value.

[0170] Table 1 gives the true value, standard deviation setting and measured value of the adapted 4-bus system.

[0171] Table 2 shows the final state estimation results. The true value of the system is obtained through power flow calculation, and three different methods are used for state estimation. One is the traditional least squares algorithm WLS, the second is the QUBO model for state estimation using the GUROBI solver, and the third is the QUBO model for state estimation using an optical quantum computer.

[0172] From Table 2 and Figure 2 and Figure 3 , the correctness of the power system state estimation method based on the QUBO model of the present invention has been verified, and the final result is close to the traditional algorithm and the system true value. However, due to the limitations of current experimental conditions, the maximum available bit of the optical quantum real machine (i.e., CIM real machine) is 100, and the Q matrix data accuracy is within 8 bits. Therefore, the binary expansion accuracy is low in the experiment, and the Q matrix has been scaled, so the error is slightly greater than the result of the traditional WLS algorithm.

[0173] Table 1 Adapted 4-node measurement data

[0174] Serial number name Standard Deviation Busbar number Measurements True value error 0 Voltage Magnitude 0.001 1 1.0006449 1 0.000645 1 Active Power Injection 0.001 1 -0.0003996 0.0004 0.0008 2 Active Power Injection 0.001 2 -0.0003989 0.0004 0.000799 3 Active Power Injection 0.001 3 -0.0004000 0.0004 0.0008 4 Reactive Power Injection 0.001 1 -0.0002003 0.0002 0.0004 5 Reactive Power Injection 0.001 2 -0.0002000 0.0002 0.0004 6 Reactive Power Injection 0.001 3 -0.0002001 0.0002 0.0004

[0175] Table 2 System true value and final state estimation results

[0176]

[0177] Compared with the prior art, the present invention introduces quantum computing into power system state estimation, which is a forward-looking exploration of quantum computing in the field of power system state estimation and can meet higher requirements for computing power in the future. Compared with the implementation route of quantum circuits, the present invention is based on a dedicated optical quantum computer and is more likely to achieve higher quantum bit growth in a short period of time, thereby achieving commercialization.

[0178] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

[0179] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A method for estimating the operating state of a power system based on quantum computing, characterized in that: The steps include: Step S1: Obtain the measurement data of the voltage and power of each node, preset the measurement standard deviation, construct the weighted coefficient matrix W, initialize the node voltage and phase angle, and preset the step value for jumping out of iteration; Step S2: Calculate the measurement function h(x) and the Jacobian matrix model H(x) according to the initial voltage phase angle; Step S3: Determine the binary expression matrix by the preset model accuracy Obtain the Q matrix of the QUBO model; Step S4: Input the Q matrix into CIM, set the solution parameters, obtain the solution vector, and the binary expression matrix The product of the vector is the state quantity correction vector, and the estimated values ​​of the node voltage and phase angle are corrected according to formula (25); Binary Expression Matrix Convert continuous variables into 0-1 binary variables. For continuous variables w, express the continuous variable w through binary expression as follows: Where: is a matrix of n×nK; It is an nK×1 vector composed of 0-1 binary variables, where n is the number of independent variables and the value of K determines the accuracy of the binary expression; Step S5: Determine whether the maximum absolute value of the iteration step is less than a preset value. If it is greater than the preset value, return to step 2 and continue the next iteration. If it is less than the preset value, exit the loop to obtain the final state estimation node voltage value and phase angle value.

2. The method for estimating the operating state of a power system based on quantum computing according to claim 1, characterized in that: In step S1, the initialized voltage per unit value is 1, the initialized phase angle per unit value is 0, and W is a weighting coefficient matrix that represents the confidence of the measurement. Its construction method is consistent with that in WLS: Where σ is the standard deviation of a single measurement, representing the confidence level of that measurement.

3. The method for estimating the operating state of a power system based on quantum computing according to claim 2, characterized in that: In step S2, the measurement function of the distribution network is as follows: 1) Node voltage measurement model V i z =V i (3) 2) Node injection active power measurement model 3) Node injection reactive power measurement model 4) Line transmission active power measurement model 5) Line transmission reactive power measurement model The measurement function h(x) in step S2 is derived to have the corresponding Jacobian matrix in the form shown below: Among them, the partial derivative results of each measurement function are as follows: 1) Node voltage measurement model 2) Node injection active power measurement model 3) Node injection reactive power measurement model 4) Line transmission active power measurement model 5) Line transmission reactive power measurement model In the above formula: 1) Subscript i: node number; 2) Superscript z: indicates that the value is a measured value; 3) V i : voltage amplitude of node i (pu); 4)θ i : voltage phase angle value of node i (degree); 5)θ ij : The phase angle difference between node i and node j (degree); 6)P i : injected active power of node i (pu); 7) Q i : injected reactive power of node i (pu); 8)P ij : The transmitted active power of the line from node i to node j (pu); 9) Q ij : The transmitted reactive power of the line from node i to node j (pu); 10) G ij : admittance matrix conductance; 11) B ij : admittance matrix susceptance; 12) Self-susceptance of node i.

4. The method for estimating the operating state of a power system based on quantum computing according to claim 3 is characterized in that: In step S3, the binary expression matrix Convert continuous variables into 0-1 binary variables. For continuous variables w, express the continuous variable w through binary expression as follows: Where: is a matrix of n×nK; It is an nK×1 vector composed of 0-1 binary variables, where n is the number of independent variables and the value of K determines the accuracy of the binary expression.

5. The method for estimating the operating state of a power system based on quantum computing according to claim 4 is characterized in that: In step S3, the transformation of the QUBO model is derived as follows: The solution of each iteration in Newton iteration is in the form of: For the following least squares regression problem: Its analytical solution is: w=(X T X) -1 X T Y (29) Therefore, let have to, make: X1=W1H(x k ),Y1=W1[z-h(x k )] (31) The solution in Newton iteration can be written in the form of the following least quadratic solution: Δx k =(X1 T X1) -1 X1 T Y1 (32) Combining the binary expansion of equation (25) and equation (27), each iterative solution can be regarded as solving the following QUBO model: Then the corresponding Q matrix is:

6. The method for estimating the operating state of a power system based on quantum computing according to claim 5, characterized in that: The step S4 uploads the Q matrix to the quantum cloud platform, calls the optical quantum computer to solve the QUBO model, returns the solution vector, and obtains the iteration vector according to formula (25), updates the state vector, including the state estimation values ​​of voltage and phase angle.

7. The method for estimating the operating state of a power system based on quantum computing according to claim 6, characterized in that: The step S5 determines whether the maximum iteration step of the iteration vector in S4 is less than a preset value. If it is less than the preset value, it is considered that the model has converged. Otherwise, the iteration continues. The final iteration value obtained by the algorithm is compared with the true value and the traditional WLS solution value to obtain the effectiveness of the method.

Images