Wireless charging system and method for unmanned inspection equipment of power system

Abstract

The invention discloses a wireless charging system and method for unmanned inspection equipment of a power system. The wireless charging system comprises the unmanned inspection equipment, an ET and a reconfigurable intelligent surface, a radio frequency signal output by the ET is connected with the beam former, one part of the radio frequency signal processed by the beam former is received by a receiving antenna of the unmanned inspection equipment, and the other part of the radio frequency signal is received by the receiving antenna after being processed by the reconfigurable intelligent surface beam former. The multiple energy collection circuits are nonlinear and are connected in parallel, radio frequency power output by the receiving antenna is connected with input of all the energy collection circuits connected in parallel through the power divider, output of all the energy collection circuits connected in parallel is connected with the storage battery, and the energy collection circuits rectify the radio frequency power into direct current power to charge the storage battery. The unmanned inspection equipment can supplement energy in a fixed time slot and continuously receive wireless charging signals, so that long-time hovering and operation are realized, and the microwave WPT transmission efficiency can be greatly improved.

Description

Technical Field

[0001] The present invention relates to an improvement in charging technology for unmanned inspection equipment, and in particular to a wireless charging system and method for unmanned inspection equipment in a power system, belonging to the technical field of wireless power transmission. Background Art

[0002] With the continuous growth of technological innovation and electricity demand, the construction scale of wind power, photovoltaic power generation, overhead lines, etc. has continued to expand. It is very important to maintain the safe and stable operation of the power system and prevent serious system failures by inspecting the equipment in the above scenarios. Manual inspections alone will result in high labor intensity, long inspection time, and low inspection efficiency. Therefore, various unmanned inspection equipment (i.e., intelligent unmanned equipment UIV) such as drones, robots, and operation and maintenance inspection vehicles are often used in power systems to realize intelligent inspections in various environments and tasks and improve inspection efficiency. Although these intelligent devices are flexible and efficient in performing tasks such as information collection, processing, and delivery, they have short battery life and require frequent battery replacement or wired charging, which not only limits the operating time of the device, but also causes unnecessary manpower consumption. Therefore, it is necessary to find a solution to provide continuous power supply.

[0003] Wireless power transfer technology (WPT) can charge devices, and there are usually two types: non-radiative WPT and radiative WPT. Non-radiative WPT mainly includes electric field coupling and magnetic field coupling, while radiative WPT is mainly divided into microwave, laser, etc. Non-radiative WPT has a shorter transmission distance, and the UIV still needs to return from the work site to the wireless charging device. Therefore, radiative WPT is more suitable for power transmission occasions with longer distances and larger capacities, but lasers are more susceptible to weather than microwaves, so it is more common to use microwave WPT to charge UIV.

[0004] However, the transmission efficiency of microwave WPT is low. In addition, the continuous phase shift control, digital beamformer, and linear energy harvesting (EH) model of reconfigurable smart surface (RIS) are usually not suitable for practical WPT systems. Therefore, it is urgent to propose a charging method for unmanned inspection equipment based on microwave wireless power transmission to improve the transmission efficiency of microwave WPT. Summary of the invention

[0005] In view of the problems of short battery life and low charging efficiency of existing intelligent unmanned inspection equipment, the purpose of the present invention is to provide a wireless charging system and method for unmanned inspection equipment in power systems. The present invention introduces an energy transmitter (ET) into the power system and uses sufficient electric energy in the environment to power it. The unmanned inspection equipment can replenish energy in fixed time slots and continuously receive wireless charging signals, thereby achieving long-term hovering and operation in the air. The present invention can greatly improve the transmission efficiency of microwave WPT.

[0006] The technical solution of the present invention is achieved in this way:

[0007] A wireless charging system for unmanned inspection equipment of an electric power system includes an unmanned inspection equipment, which has a receiving antenna, an energy collection circuit and a battery. It also includes a microwave energy transmitter and a reconfigurable smart surface; the electric power system is connected to the energy transmitter to provide a DC power supply to the energy transmitter, the radio frequency signal output by the energy transmitter is connected to a beamformer, a part of the radio frequency signal processed by the beamformer is directly received by the receiving antenna on the unmanned inspection equipment, and the other part of the radio frequency signal is received by the receiving antenna on the unmanned inspection equipment after being processed by the beamforming of the reconfigurable smart surface; the energy collection circuit is a nonlinear energy collection circuit and is a plurality of parallel-connected circuits, the radio frequency power output by the receiving antenna is connected to the input end of all energy collection circuits connected in parallel through a power splitter, the radio frequency power is distributed to each energy collection circuit by the power splitter, the output end of all energy collection circuits connected in parallel is connected to the battery, and the radio frequency power is rectified into DC power by the energy collection circuit to charge the battery.

[0008] The present invention also provides a wireless charging method for unmanned inspection equipment of a power system, which is performed according to the following steps:

[0009] 1) Construct the aforementioned wireless charging system for unmanned inspection equipment of the power system; then construct the energy transmission model and energy collection model respectively;

[0010] 1.1) Constructing energy transfer model

[0011] The beamformer adopts an analog beamformer equipped with a finite discrete phase shifter to enhance the signal spatial transmission gain; the analog beamformer f has R low-resolution phase shifters, each of which has a unit modulus and is composed of b ET Discrete phase of control, b ET is the phase resolution, and the set of possible phases for each phase shifter is defined as

[0012]

[0013] In this structure, the signal transmitted by the energy transmitter is

[0014]

[0015] Where, the total number of time slots is N, and any time slot p t [n] represents the transmission power at the energy transmitter in the nth time slot, f[n] is the analog beamforming vector of the nth time slot, s is a dedicated wireless power transmission signal, and is a 1×1 order deterministic signal satisfying |s|=1;

[0016] The reconfigurable smart surface is used to compensate for the channel path loss experienced by the transmission signal; the phase shift matrix of the reconfigurable smart surface is defined as is a diagonal matrix representing the effective phase shift applied by the reflective elements of the reconfigurable smart surface; φ = [φ1, φ2, …, φ Q ] T Depend on Composition, of which b RIS bit resolution for representing the phase shift of the reconfigurable smart surface;

[0017] The signal received by the unmanned inspection device combines the direct transmission from the energy transmitter and the signal reflected from the reconfigurable smart surface, which is expressed as

[0018]

[0019] where h EU [n], h RU [n] and H ER represents the equivalent complex baseband channel coefficients between the unmanned inspection device and the energy transmitter, between the unmanned inspection device and the reconfigurable smart surface, and between the reconfigurable smart surface and the energy transmitter in each time slot n. H represents the conjugate transpose. is additive Gaussian white noise;

[0020] Therefore, the power carried by the RF signal received by the receiving antenna in the nth time slot is approximately

[0021]

[0022] 1.2) Building an energy harvesting model

[0023] Assume that the RF signal received by the receiving antenna is divided into L parts through a power splitter, and each part is sent to a different energy harvesting circuit; define α l [n] is the power allocation ratio of the lth energy harvesting circuit in the nth time slot, Then the output DC power of each energy harvesting circuit in each time slot is adjusted according to the respective input streams. The relationship between the output DC power and the input RF power is as follows:

[0024]

[0025] in M represents the saturation upper limit of the output DC power. Constants a and b represent the combined effects of resistance, capacitance and circuit sensitivity on the rectification process. α[n]=[α1[n],α2[n],…,α L [n]] T is the power allocation ratio vector of the nth time slot, satisfying the condition To ensure the total input power of all energy harvesting circuits is conserved;

[0026] 2) Taking the maximization of the power transmission efficiency of the wireless charging system as the optimization goal, a joint optimization model of beamforming and power allocation is established; the model is expressed as

[0027]

[0028] In the formula, f r [n] represents the rth element of the transmit analog beamforming vector f[n] in the nth time slot, and Φ(q, q)[n] represents the Qth diagonal element in the diagonal matrix Φ[n]. Constraints (12a) and (12b) respectively ensure that the elements of the transmit analog beamformer and the passive reflection beamformer are selected in a finite resolution set. Constraints (12c) and (12d) manage the power distributed in the energy harvesting circuit of the unmanned inspection device to ensure that no circuit operates in a saturated state. Constraint (12e) ensures that the energy harvested by the unmanned inspection device in each time slot exceeds a predefined threshold value P. th ; Constraint (12f) limits the transmission power at the energy transmitter to prevent exceeding the maximum allowed power P max .

[0029] Furthermore, the establishment of the joint optimization model of beamforming and power allocation in step 2) is decomposed into the following two steps:

[0030] 2.1) Focusing on the maximum channel power gain in each time slot, assuming a fixed transmit power to maximize the channel gain, the channel power gain maximization problem is expressed as

[0031]

[0032] 2.2) Dynamically adjust the transmission power of the energy transmitter and the power splitter of the unmanned inspection equipment to maximize the overall power transmission efficiency η during the mission; expressed as:

[0033]

[0034] Furthermore, in step 2.1), the beamforming is optimized using an alternating optimization method, and the beamforming is converted into two problems: transmit beamforming and reflective beamforming.

[0035] The transmit beamforming and the reflected beamforming are implemented by Algorithm 1. The steps of Algorithm 1 are as follows:

[0036] 1a: Use the formula Generate exponential coefficient set c;

[0037] 1b: Calculation The corner boundary is Definition of the real and imaginary parts of ;

[0038] 1c: Sort the calculated corner boundaries in ascending order;

[0039] 1d: Selecting angles from consecutive sorted boundaries

[0040]

[0041] 1e: Ensure Falling Within the effective range of , and calculate the beamforming vector f of all R antennas of the energy transmitter;

[0042] 1f: Evaluate the objective function for each calculated beamforming vector;

[0043] 1g: Select the beamforming vector f that maximizes the objective function;

[0044] Where g = (h H EU +h H RU ΦH ER ) H represents the cascade channel between ET and UIV, θ represents the candidate angle boundary, Indicates auxiliary angle;

[0045] definition * indicates conjugation, h d =f H h EU , t is an auxiliary variable introduced; the reflection beamforming optimization problem is expressed as

[0046]

[0047] Where Σ is the diagonal matrix containing the eigenvalues ​​after eigenvalue decomposition of R, and U is the matrix containing the eigenvectors after eigenvalue decomposition of R. Use Algorithm 1 to solve equation (27) and obtain the optimal solution xopt , then at this time x opt The first N terms of * opt , the last item is t opt , that is, x opt =[φ * opt t opt ] T , T means taking the transpose, then the objective function

[0048]

[0049] To ensure the consistency of the objective function, the modified optimal transmit beamforming vector is φ * opt t * opt ;

[0050] Finally, the channel gain between the energy transmitter and the energy harvesting circuit is optimized in each time slot using the following iterative algorithm 2:

[0051] 2a: Initialization: Set a feasible initial solution φ (0) , iteration index j = 0;

[0052] 2b: Update the transmit beamforming vector: Given the value of φ at the jth iteration φ (j) , use Algorithm 1 to find the optimal f obtained in the j+1th iteration, that is, f (j+1) ;

[0053] 2c: Update the reflection beamforming vector: Use the optimized f (j+1) , use Algorithm 1 to determine the current Then update φ (j+1) is φ * opt t * opt ;

[0054] 2d: Iteration: If the convergence condition is met, go to step 2e, otherwise increase j and return to step 2b;

[0055] 2e: Output: Returns the optimized f (j+1) and φ (j+1) .

[0056] Furthermore, the power allocation ratio α l [n] and the transmission power p t [n] merged into a single variable y l [n] = α l [n]p t [n](h opt [n])2 ,h opt [n] is the optimal channel gain for each time slot; the power transmission efficiency in step 2.2) is expressed as

[0057]

[0058] Where P c is the other energy loss at ET;

[0059] Using the Dinkelbach algorithm, equation (31) is transformed into the following equation (32):

[0060]

[0061] Where η opt It is an auxiliary parameter introduced in the Dinkelbach algorithm;

[0062] Further introduce auxiliary variable τ l [n] non-convex Convert to convex form and use the SCA algorithm to transform equation (32) into the following equation (33):

[0063]

[0064] in Constraint (33b) is transformed using a first-order Taylor expansion into

[0065]

[0066] where τ l [n] (j) is the current value of the corresponding variable at the jth iteration;

[0067] The power allocation problem established in equation (33) is solved by Algorithm 3. The detailed steps of Algorithm 3 are as follows:

[0068] 3a: Initialization: Set a feasible initial solution {y l [n] (0)}, η opt , iteration index j = 0, stop threshold ∈ η and ∈ obj ;

[0069] 3b: External loop: When η opt (j+1) -η opt (j) >∈ η , loop to perform the following operations;

[0070] 3c: Inner loop: When the difference between the objective function of the j+1th and jth equation (33) is greater than ∈obj , loop to perform the following operations;

[0071] 3d: Pass Update τ l [n] (j) ;

[0072] 3e: Solve the convex problem of equation (33) and express its optimal solution as τ l [n] * and l [n]*;

[0073] 3f: Update y l [n] j+1 =y l [n] * ;

[0074] 3g: When the difference between the objective function of the j+1th and jth equations (33) is less than ∈ obj , end the inner loop, otherwise return to step 3c;

[0075] 3h: Update η opt ;

[0076] 3i: when η opt (j+1) -η opt (j) <∈ η , end the outer loop, otherwise return to step 3b;

[0077] 3j: Output: Returns the optimized y and calculates the optimal power allocation ratio and transmit power for each time slot.

[0078] Compared with the prior art, the present invention has the following beneficial effects:

[0079] 1. To enhance the power transmission efficiency PTE, the power gain in the wireless transmission process is improved by increasing the size of the antenna array and RIS. In this process, the existing technology usually uses digital beamforming, but its cost is difficult to estimate when the antenna scale is large. The use of analog beamforming to transmit RF signals in a directional manner can reduce costs. At the same time, the phase shift design of the reconfigurable smart surface (RIS) is usually assumed to be continuous, which cannot be achieved in actual hardware design. The proposed algorithm can design a discrete phase reconfigurable smart surface (RIS) in a less complex manner to obtain the optimal phase selection and achieve power gain in the transmission process to alleviate the power attenuation caused by distance.

[0080] 2. The previous design model usually expresses the nonlinear energy harvesting process in a linear form, which is not practical. Therefore, the present invention adopts a more practical nonlinear energy harvesting circuit. However, the nonlinear energy harvesting circuit has a saturation threshold, so a power splitter is used for power distribution, thereby greatly improving the power transmission efficiency.

[0081] In summary, the present invention can enhance the working life of unmanned inspection equipment and improve inspection efficiency by improving the wireless power transmission efficiency PTE. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 -Flowchart of the wireless charging method for unmanned inspection equipment of power system used in the present invention.

[0083] Figure 2 -The wireless charging system model and energy transmission flow chart of the present invention; wherein (a) is the wireless charging system model; (b) is the energy transmission flow chart.

[0084] Figure 3 -Schematic diagram of the energy harvesting circuit architecture based on power splitting of the present invention.

[0085] Figure 4 - Schematic diagram of 8-PSK phase on the complex circle of the present invention.

[0086] Figure 5 - Schematic diagram of the relationship between received RF power and transmitted power under Algorithm 1 of the present invention.

[0087] Figure 6 - Schematic diagram of the relationship between received RF power and transmitted power under Algorithm 2 of the present invention.

[0088] Figure 7 - A schematic diagram showing the relationship between the received RF power and the y-axis coordinates of the number of RIS units and UIV of the present invention.

[0089] Figure 8 - Schematic diagram of the relationship between the power transmission efficiency and the maximum allowable power of the present invention.

[0090] Fig. 9 - Schematic diagram of the relationship between the power transmission efficiency and UIV speed of the present invention.

[0091] Fig.10 - Schematic diagram of the relationship between the total energy consumption at ET and UIV speed of the present invention. DETAILED DESCRIPTION

[0092] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0093] See also Figure 2 and Figure 3The present invention is a wireless charging system for unmanned inspection equipment of an electric power system, comprising an unmanned inspection equipment, the unmanned inspection equipment having a receiving antenna, an energy collection circuit and a battery; further comprising a microwave energy transmitter and a reconfigurable smart surface; the electric power system is connected to the energy transmitter to provide a DC power supply to the energy transmitter, the radio frequency signal output by the energy transmitter is connected to a beamformer, a part of the radio frequency signal processed by the beamformer is directly received by the receiving antenna on the unmanned inspection equipment, and another part of the radio frequency signal is received by the receiving antenna on the unmanned inspection equipment after being processed by the beamforming of the reconfigurable smart surface; the energy collection circuit is a nonlinear energy collection circuit and is a plurality of parallel-connected circuits, the radio frequency power output by the receiving antenna is connected to the input end of all energy collection circuits connected in parallel through a power splitter, the radio frequency power is distributed to each energy collection circuit by the power splitter, the output end of all energy collection circuits connected in parallel is connected to the battery, and the radio frequency power is rectified into DC power by the energy collection circuit to charge the battery.

[0094] The wireless charging method for unmanned inspection equipment of the power system proposed by the present invention is carried out according to the following steps. Figure 1 ,

[0095] 1. Construct the aforementioned intelligent unmanned equipment wireless charging system (see Figure 2 and Figure 3 ), beamforming technology is used to enhance the spatial transmission gain of the signal. After the RF signal passes through the receiving antenna, the power is distributed by the power splitter, and then a nonlinear energy harvesting circuit is used to rectify the RF power into DC power for charging unmanned equipment.

[0096] The UIV performing power line inspection tasks is wirelessly powered by a microwave energy transmitter (ET) connected to the power system. The ET is equipped with R antennas, Q elements on the RIS, and the UIV has only one receiving antenna.

[0097] The total duration of performing the inspection task is defined as T, which is subdivided into N segments of equal duration, each of length ΔT. In the nth time slot, The upper left vertex positions of ET, RIS and UIV are p ET =(x ET ,y ET ,z ET ), p RIS =(x RIS ,y RIS ,z RIS ) and p UIV [n]=(x UIV [n],y UIV [n],z UIV[n]). At each time slot n, the position of the UIV is quasi-static. Furthermore, if the duration of each time slot is short enough relative to the UIV's velocity and the distance to the other endpoints, the maximum displacement of the UIV during the duration is negligible. Therefore, it can be assumed that at each time slot, the propagation environment and the distance between the UIV and the ET remain constant. The UIV is charged by following a pre-set patrol trajectory to perform a pre-defined mission.

[0098] (a) Channel Model

[0099] In each time slot n, the equivalent complex baseband channel coefficients between UIV and ET, between UIV and RIS, and between RIS and ET are expressed as h EU [n], h UR [n] and H ER express.

[0100] With the use of high-frequency bands and large-scale antenna arrays, UIV may fall into the near-field region of energy transmitters and RIS. Therefore, the channel between devices can be generated by replacing different transmit and receive antenna pairs (t, p). Assume that both ET and RIS are equipped with uniform planar arrays (UPA) with sizes Q = Q x ×Q z and R = R x ×R z For the vth antenna of v∈{p,t},V∈{Q,R}, the corresponding position vector can be expressed as

[0101]

[0102] The present invention selects a line-of-sight (LoS) channel model that depends on the antenna or element distance. The channel coefficient between the pth receiving antenna and the tth transmitting antenna can be expressed as

[0103] H p,t =A p,t e -jk|p-t| (2)

[0104] Where p=(x p ,y p ,z p ), t=(x t ,y t ,z t ) represent the positions of the pth receiving antenna and the tth transmitting antenna, respectively, and their distance is represented by |pt|, e -jk|p-t| represents the phase change of the signal from p to t, λ c is the wavelength. p,t represents the path gain coefficient

[0105]

[0106] F(Θ p,t ) is the radiation profile of each UPA element

[0107]

[0108] Where b is the aperture gain, Θ p,t =(θ p,t ,φ p,t ). In the line-of-sight channel model, the elevation angle θ p,t and azimuth φ p,t :

[0109]

[0110] (b) Beamforming design

[0111] An analog beamformer with finite discrete phase shifters is used. The analog beamformer f has R low-resolution phase shifters, each with unit modulus and a phase shifter consisting of b ET Discrete phase of control, b ET is the phase resolution, and the set of possible phases for each phase shifter is defined as

[0112] In this structure, the signal transmitted by ET is

[0113]

[0114] Where p t [n] represents the transmission power at the nth time slot ET, f[n] is the analog beamforming vector of the nth time slot, and s is a dedicated wireless power transmission signal, which is a 1×1 order deterministic signal satisfying |s|=1.

[0115] In addition, the channel path loss experienced by the transmitted signal is compensated by using RIS. The phase shift matrix of RIS is defined as is a diagonal matrix representing the effective phase shift applied by the RIS reflector element. φ=[φ1,φ2,…,φ Q ] T Depend on Composition, of which b RIS represents the bit resolution of the RIS phase shift. The signal received by the UIV combines the direct transmission from the ET and the signal reflected from the RIS and is expressed as

[0116]

[0117] Among them, H means taking the conjugate transpose, is additive white Gaussian noise (AWGN).

[0118] Therefore, the power carried by the received RF signal in the nth time slot can be approximated as

[0119] P[n]=p t [n]||(h EU [n] H +h RU [n] H Φ[n]H ER )f[n]|| 2 (9)

[0120] (c) Energy harvesting model

[0121] The UIV combines the signals captured by its antenna into a single signal stream through an analog energy combiner. The combined signal is then fed into the EH circuit, which includes nonlinear elements such as diodes. The simple linear model in which the RF-DC conversion efficiency depends only on the input RF power cannot accurately simulate the nonlinearity in the actual EH circuit. The present invention uses a nonlinear EH model to represent the relationship between the output DC power and the input RF power. The model is defined as

[0122]

[0123] in M represents the saturation upper limit of the output DC power, and constants a and b represent the combined effects of resistance, capacitance, and circuit sensitivity on the rectification process.

[0124] However, nonlinear EH circuits often operate in the saturation region at high input power, resulting in reduced conversion efficiency. To alleviate this problem, a power-divided EH receiver architecture is used. This architecture utilizes L independent EH circuits. The received RF signal is divided into L parts, each of which is fed into a different EH circuit, thereby preventing a single EH circuit from entering the saturation region and ensuring that each EH circuit operates within the optimal power range, improving the overall energy conversion efficiency. Define α l [n] is the power allocation ratio of the lth EH circuit in the nth time slot. Then the output DC power of each circuit in each time slot is adjusted according to the respective input flow, and the correction model is as follows

[0125]

[0126] α[n]=[α1[n],α2[n],…,α L [n]] T is the power allocation ratio vector of the nth time slot, satisfying the condition To ensure that the total input power of all EH circuits is conserved.

[0127] Second, with the optimization goal of maximizing the system power transmission efficiency, a joint optimization problem of beamforming and power allocation is established.

[0128] A strategy is designed to optimize the transmit power, simulated transmit beamforming, passive reflective beamforming, and power splitter to build a green and efficient wireless power transmission system. The optimization problem is formulated as

[0129]

[0130] In the formula, f r [n] represents the rth element of the transmit analog beamforming vector f[n] in the nth time slot, and Φ(q, q)[n] represents the Qth diagonal element in the RIS phase shift diagonal matrix Φ[n] in the [n]th time slot. Constraints (12a) and (12b) ensure that the elements of the transmit analog beamformer and the passive reflective beamformer are selected in a finite resolution set, respectively. Constraints (12c) and (12d) manage the power distributed in the EH circuits of the UIV, ensuring that no circuit operates in saturation. Constraint (12e) ensures that the energy harvested by the UAV in each time slot exceeds a predefined threshold P th Finally, constraint (12f) limits the transmit power at ET to prevent exceeding the maximum allowed power P max . The original problem is decomposed into two steps.

[0131] The first step focuses on the maximum channel power gain in each time slot, and assumes a fixed transmit power to maximize the channel gain without considering the variation of RF-DC power conversion efficiency. Therefore, the channel power gain maximization problem can be expressed as

[0132]

[0133] The second step involves dynamically adjusting the transmit power of the ET and the power splitter of the UIV to maximize the overall PTE during the mission. It is expressed as:

[0134]

[0135] 3. According to the established optimization model, the alternating optimization method is used to optimize the beamforming, and the beamforming design is converted into transmit beamforming and reflection beamforming problems. These two problems are solved by the proposed algorithm.

[0136] (1) Active transmit beamforming design

[0137] When the RIS phase shift matrix Φ[n] of the nth time slot remains unchanged, the beamforming optimization problem (P1) can be simplified as follows:

[0138]

[0139] In the formula, g = (h EU H +h RU H ΦH ER ) H Represents the cascade channel between ET and UIV.

[0140] Specifically, the present invention only involves channels of rank one. This problem can be converted into an element-to-element maximization problem with respect to different θ values.

[0141]

[0142] In the formula, * indicates conjugation.

[0143] The available values ​​of phase shift can be expressed as The circle is represented by a series of phases. The line is divided into non-intersecting line segments, and each line is used as a decision boundary. Figure 4 As shown, set b ET = 3. These boundaries determine the specific phase shift f r , at a given phase rotation θ acting on g r Maximize system performance when r The projection of exp{-jθ} on the circle falls into Figure 4 In region 1, the optimum phase is f 1 r . Select f r The decision of remains valid as long as θ does not cause the projection to cross the decision boundary. The mathematical expressions of these decision boundaries are as follows

[0144]

[0145] in Equivalent to

[0146]

[0147] It shows that when θ changes, f r May change. By forming all boundary terms into a vector and apply it to each g r exp{-jθ}, the whole system The boundary can be expressed as

[0148]

[0149] This means

[0150]

[0151] This equation means that the rth element of g is represented by the decision boundary Each of the exponential terms is phase shifted. Therefore, Each element of determines a The angle is applied to the introduced θ, each associated with a different beamforming candidate f. The bounds on θ that may change the beamforming candidate are calculated according to the formula

[0152]

[0153] In summary, the steps of Algorithm 1 for finding the best analog beamforming are as follows:

[0154] 1a: Use the formula Generate exponential coefficient set c;

[0155] 1b: Calculate using formula (21) The corner boundary is Definition of the real and imaginary parts of ;

[0156] 1c: Sort the calculated corner boundaries in ascending order;

[0157] 1d: Selecting angles from consecutive sorted boundaries

[0158]

[0159] 1e: Ensure Falling Within the effective range of , and calculate the beamforming vector f of all R antennas of the energy transmitter;

[0160] 1f: Evaluate the objective function under each calculated beamforming vector, that is,

[0161] 1g: Select the beamforming vector f that maximizes the objective function.

[0162] Where g = (h H EU +h H RU ΦH ER ) H represents the cascade channel between ET and UIV, θ represents the candidate angle boundary, Indicates auxiliary angle;

[0163] (2) Reflection beamforming design

[0164] After optimizing the transmit beamforming, design the reflective beamforming

[0165]

[0166] The optimization problem is a non-convex quadratic constrained quadratic programming problem (QCQP). By introducing an auxiliary variable t, the problem is transformed into a homogeneous QCQP problem and defined

[0167]

[0168] where R and x are defined as

[0169]

[0170] in and h d =f H h EU .

[0171] Then, the optimization problem can be reformulated using eigenvalue decomposition as

[0172]

[0173] Where Σ is the diagonal matrix containing the eigenvalues ​​after eigenvalue decomposition of R, and U is the matrix containing the eigenvectors after eigenvalue decomposition of R. (P1-B.2) can be processed using the same algorithm 1 as for finding the optimal transmit beamforming. The optimal solution x is obtained opt , define x at this time opt The first N terms of * opt , the last item is t opt , that is, x opt =[φ * opt t opt ] T , T means taking the transpose. Due to the existence of t, the objective function is

[0174]

[0175] In this case, the optimal solution for t may no longer be 1, but fall into the set Therefore, to ensure the consistency of the objective function, the modified optimal reflection beamforming vector is φ * opt t * opt .

[0176] Finally, the channel gain between ET and energy harvesting UIV is optimized in each time slot using the following iterative algorithm (Algorithm 2):

[0177] 2a: Initialization: Set a feasible initial solution φ (0) , iteration index j = 0;

[0178] 2b: Update the transmit beamforming vector: Given the value of φ at the jth iteration φ (j) , use Algorithm 1 to find the optimal f obtained in the j+1th iteration, that is, f (j+1) ;

[0179] 2c: Update the reflection beamforming vector: Given the optimized f (j+1) , determined using Algorithm 1 Then calculate φ (j+1) is φ * opt t * opt ;

[0180] 2d: Iteration: If the convergence condition is met, go to step 2e, otherwise j=j+1 and return to step 2b;

[0181] 2e: Output: Returns the optimized f (j+1) and φ (j+1) .

[0182] 4. Based on the optimized beamforming design, the Dinkelbach algorithm and the SCA algorithm are used for power allocation design.

[0183] The optimal channel gain for each time slot is expressed as

[0184] h opt [n]=|(h EU [n] H +h RU [n] H Φ opt [n]H ER )f opt [n]| (29)

[0185] f opt [n] and Φ opt [n] is obtained through Algorithm 2 and is the optimal transmit beamforming vector and reflection beamforming vector obtained when the time slot n is fixed. At this time, the objective function in (P2) is

[0186]

[0187] To simplify the design, the power allocation ratio α l [n] and the transmission power p t [n] merged into a single variable y l [n] = α l [n]p t [n](h opt [n]) 2 . (P2) can be reformulated as

[0188]

[0189] Where P c is the other energy loss at ET.

[0190] Using the Dinkelbach algorithm, (P2.1) is transformed into:

[0191]

[0192] Where η opt It is an auxiliary parameter introduced in the Dinkelbach algorithm;

[0193] In order to solve the non-convexity of the objective function and constraints in (P2.2), an auxiliary variable τ is introduced l [n] non-convex Convert to convex form and use the SCA algorithm. (P2.2) is reformulated as

[0194]

[0195] in Constraint (33b) can be transformed using a first-order Taylor expansion into

[0196]

[0197] where τ l [n] (j) is the current value of the corresponding variable at the jth iteration.

[0198] Formula (33) is the power allocation model that the present invention seeks to solve after a series of transformations; this transformation ensures that (P2.3) is a convex optimization problem that can be solved using standard convex optimization techniques (such as the interior point method).

[0199] Algorithm 3 summarizes the detailed steps of power parameter optimization as follows:

[0200] 3a: Initialization: Set a feasible initial solution {y l [n] (0)}, η opt , iteration index j = 0, stop threshold ∈ η and ∈ obj ;

[0201] 3b: External loop: When η opt (j+1) -η opt (j) >∈ η , loop to perform the following operations;

[0202] 3c: Inner loop: When the difference between the objective function of the j+1th and jth (P2.3) is greater than ∈obj , loop to perform the following operations;

[0203] 3d: Pass Update τ l [n] (j) ;

[0204] 3e: Solve the convex problem (P2.3) and express its optimal solution as τ l [n] * and l [n]*;

[0205] 3f: Update y l [n] j+1 =y l [n] * ;

[0206] 3g: When the difference between the objective function of the j+1th and jth (P2.3) is less than ∈ obj , end the inner loop, otherwise return to step 3c;

[0207] 3h: Update η opt ;

[0208] 3i: when η opt (j+1) -η opt (j) <∈ η , end the outer loop, otherwise return to step 3b;

[0209] 3j: Output: Returns the optimized y and calculates the optimal power allocation ratio and transmit power for each time slot.

[0210] Algorithm 3 includes the process of solving formula (33), but the ultimate goal is to solve the optimal η. Because after a series of transformations, the objective function is not the final η required, so the optimal y obtained by solving formula (33) needs to be substituted into the expression of the objective function η in P2.1 (i.e., formula 31) to make η meet the optimal. The inner loop in Algorithm 3 is the process of solving formula (33) to obtain the optimal y at this time, and the outer loop is the process of solving the final η required. The optimality is determined by whether the difference between the two iterations is less than the stop threshold.

[0211] In order to more clearly describe the algorithm steps in the technical solution, the following examples are given to evaluate the performance of the algorithm used for beamforming design and power allocation in the proposed charging method. Unless otherwise specified, the simulation parameters are set as follows: d = λ c / 4,f c =3GHz, p t =500W, b RIS =b ET=5. For the nonlinear EH circuit, the parameters used are: a = 0.082, b = 30, M = 40. The total average energy consumption of the transmitting circuit P C =30W. Set the endpoint position to (x ET ,y ET , z ET )=(0,0,15),(x RIS ,y RIS , z RIS )=(-1,5,15),(x UIV ,y UIV , z UIV )=(0,10,15).

[0212] In order to verify the performance of Algorithm 1 (referred to as "Algorithm 1" in the figure) and evaluate the impact of different beamforming techniques on the received RF power, experiments were conducted by optimizing one beamforming vector each time and randomly generating another beamforming vector. It also includes comparing the results obtained by exhaustive search (represented as "ES" in the figure) of the two beamforming vectors to prove the effectiveness of the proposed algorithm. Figure 5 It can be seen that the results obtained by Algorithm 1 are comparable to those obtained by the exhaustive search method, thus confirming the optimality of Algorithm 1 when selecting discrete phase shifts.

[0213] Then, the received RF power is evaluated as a function of the transmitted power for different bit resolution designs, setting b RIS =b T = b0. The benchmark includes the scenario of transmit beamforming and RIS continuous phase shift (denoted as “Continuous” in the figure), which is the upper limit of the best beamforming performance. In addition, a quantization method (in Figure 6 The continuous solution is quantized into discrete phase shifts (denoted as "Quantization" in ). It can be concluded that Algorithm 2 is suitable for scenarios where power gain is prioritized, especially in the case of lower 1-bit or 2-bit phase resolution that is common in practical implementations.

[0214] Considering the mobility of UIV and the use of large antenna arrays, wireless charging may occur in the near field. In this case, by setting (x RIS ,y RIS , z RIS )=(-1,80,15), Figure 7 The received RF power is shown as a function of the number of RIS elements and the position of the UIV.

[0215] Next, a practical inspection scenario is explored to investigate the impact of EH circuit and system parameters on PTE. UIV ,y UIV , zUIV )=(0,1,15) flies to the destination (x) at a constant speed v=5m / s UIV ,y UIV , z UIV )=(0,70,15). The positions of ET and RIS are set to (x ET ,y ET , z ET )=(0,0,15) and (x RIS ,y RIS , z RIS )=(-9, 30, 15). In addition, the time interval of each time slot is set to ΔT=0.2s, which determines the number of time slots N of UIV at a fixed speed. Suppose UIV is in each time slot P th The power demand is equal to the propulsion power consumption of UIV. Unless otherwise specified, other system parameters are Q = 4900, R = 4900, L = 3, P max =1200w.

[0216] When N=1, Figure 8 The change of system PTE under different maximum allowed power settings is shown. Algorithm 3 (jointly optimizing PS ratio and transmit power p t ) Results and baseline scenario (p t Set to P max , the PS ratio is fixed at α l =1 / L, Figure 8 denoted as “Average PS” in the table above. Fig. 9 It can be seen that when L = 3, at low P max In this case, PTE is maximum when there is no PS. When a circuit reaches saturation, by adjusting p t To prevent the input power from further increasing, PTE is increased. max As the situation continues to increase, the three EH circuits come into operation, significantly enhancing the system PTE.

[0217] Finally, in Fig. 9 and Fig.10 By comparing the design algorithm of the present invention with other methods, it can be seen that the method used in the present invention can achieve a larger PTE and minimal energy consumption at ET.

[0218] By using the algorithm designed in the present invention, the joint optimization of the transmission power and beamforming at the ET, the reflected beamforming at the RIS and the power allocation ratio at the receiver can be effectively achieved, which significantly improves the overall PTE during the inspection task.

[0219] Finally, it should be noted that the above examples of the present invention are merely examples for illustrating the present invention, and are not intended to limit the embodiments of the present invention. Although the applicant has described the present invention in detail with reference to the preferred embodiments, for those of ordinary skill in the art, other different forms of changes and modifications can be made based on the above description. It is impossible to list all the embodiments here. Any obvious changes or modifications derived from the technical solution of the present invention are still within the scope of protection of the present invention.

Claims

1. A wireless charging system for unmanned inspection equipment of a power system, comprising an unmanned inspection equipment, wherein the unmanned inspection equipment has a receiving antenna, an energy collection circuit and a battery; characterized in that: It also includes a microwave energy transmitter and a reconfigurable smart surface; the power system is connected to the energy transmitter to provide a DC power supply to the energy transmitter, the radio frequency signal output by the energy transmitter is connected to a beamformer, a part of the radio frequency signal processed by the beamformer is directly received by the receiving antenna on the unmanned inspection equipment, and the other part of the radio frequency signal is received by the receiving antenna on the unmanned inspection equipment after being processed by the beamforming of the reconfigurable smart surface; the energy collection circuit is a nonlinear energy collection circuit and is a plurality of parallel-connected circuits, the radio frequency power output by the receiving antenna is connected to the input end of all the energy collection circuits connected in parallel through a power splitter, the radio frequency power is distributed to each energy collection circuit by the power splitter, the output end of all the energy collection circuits connected in parallel is connected to a battery, and the energy collection circuit rectifies the radio frequency power into DC power to charge the battery.

2. A wireless charging method for unmanned inspection equipment in a power system, characterized in that: The following steps are performed: 1) constructing the wireless charging system for unmanned inspection equipment of the power system as described in claim 1; and then constructing an energy transmission model and an energy collection model respectively; 1.1) Constructing energy transfer model The beamformer adopts an analog beamformer equipped with a finite discrete phase shifter to enhance the signal spatial transmission gain; The analog beamformer f has R low-resolution phase shifters, each with unit modulus and a phase shifter consisting of b ET Discrete phase of control, b ET is the phase resolution, and the set of possible phases for each phase shifter is defined as In this structure, the signal transmitted by the energy transmitter is Where, the total number of time slots is N, and any time slot p t [n] represents the transmission power at the energy transmitter in the nth time slot, f[n] is the analog beamforming vector of the nth time slot, s is a dedicated wireless power transmission signal, and is a 1×1 order deterministic signal satisfying |s|=1; The reconfigurable smart surface is used to compensate for the channel path loss experienced by the transmission signal; the phase shift matrix of the reconfigurable smart surface is defined as is a diagonal matrix representing the effective phase shift applied by the reflective elements of the reconfigurable smart surface; φ = [φ1, φ2, …, φ Q ] T Depend on Composition, of which b RIS bit resolution for representing the phase shift of the reconfigurable smart surface; The signal received by the unmanned inspection device combines the direct transmission from the energy transmitter and the signal reflected from the reconfigurable smart surface, which is expressed as where h EU [n], h RU [n] and H ER represents the equivalent complex baseband channel coefficients between the unmanned inspection device and the energy transmitter, between the unmanned inspection device and the reconfigurable smart surface, and between the reconfigurable smart surface and the energy transmitter in each time slot n. H represents the conjugate transpose. is additive Gaussian white noise; Therefore, the power carried by the RF signal received by the receiving antenna in the nth time slot is approximately P[n]=p t [n]||(h EU [n] H +h RU [n] H Φ[n]H ER )f[n]|| 2 (9) 1.2) Building an energy harvesting model Assume that the RF signal received by the receiving antenna is divided into L parts through a power splitter, and each part is sent to a different energy harvesting circuit; define α l [n] is the power allocation ratio of the lth energy harvesting circuit in the nth time slot, Then the output DC power of each energy harvesting circuit in each time slot is adjusted according to the respective input streams. The relationship between the output DC power and the input RF power is as follows: in M represents the saturation upper limit of the output DC power. Constants a and b represent the combined effects of resistance, capacitance and circuit sensitivity on the rectification process. α[n]=[α1[n],α2[n],…,α L [n]] T is the power allocation ratio vector of the nth time slot, satisfying the condition To ensure the total input power of all energy harvesting circuits is conserved; 2) Taking the maximization of the power transmission efficiency of the wireless charging system as the optimization goal, a joint optimization model of beamforming and power allocation is established; the model is expressed as In the formula, f r [n] represents the rth element of the transmit analog beamforming vector f[n] in the nth time slot, and Φ(q, q)[n] represents the Qth diagonal element in the diagonal matrix Φ[n]. Constraints (12a) and (12b) respectively ensure that the elements of the transmit analog beamformer and the passive reflection beamformer are selected in a finite resolution set. Constraints (12c) and (12d) manage the power distributed in the energy harvesting circuit of the unmanned inspection device to ensure that no circuit operates in a saturated state. Constraint (12e) ensures that the energy harvested by the unmanned inspection device in each time slot exceeds a predefined threshold value P. th ; Constraint (12f) limits the transmission power at the energy transmitter to prevent exceeding the maximum allowed power P max .

3. A wireless charging method for unmanned inspection equipment of a power system according to claim 2, characterized in that: The establishment of the joint optimization model of beamforming and power allocation in step 2) is decomposed into the following two steps: 2.1) Focusing on the maximum channel power gain in each time slot, assuming a fixed transmit power to maximize the channel gain, the channel power gain maximization problem is expressed as 2.2) Dynamically adjust the transmission power of the energy transmitter and the power splitter of the unmanned inspection equipment to maximize the overall power transmission efficiency η during the mission; expressed as:

4. A wireless charging method for unmanned inspection equipment of a power system according to claim 3, characterized in that: In step 2.1), the beamforming is optimized using an alternating optimization method, and the beamforming is converted into two problems: transmit beamforming and reflective beamforming. The transmit beamforming and the reflected beamforming are implemented by Algorithm 1. The steps of Algorithm 1 are as follows: 1a: Use the formula Generate exponential coefficient set c; 1b: Calculation The corner boundary is Definition of the real and imaginary parts of ; 1c: Sort the calculated corner boundaries in ascending order; 1d: Selecting angles from consecutive sorted boundaries 1e: Ensure Falling Within the effective range of , and calculate the beamforming vector f of all R antennas of the energy transmitter; 1f: Evaluate the objective function for each calculated beamforming vector; 1g: Select the beamforming vector f that maximizes the objective function; Where g = (h H EU +h H RU ΦH ER ) H represents the cascade channel between ET and UIV, θ represents the candidate angle boundary, Indicates auxiliary angle; definition * indicates conjugation, h d =f H h EU , t is an auxiliary variable introduced; the reflection beamforming optimization problem is expressed as Where Σ is the diagonal matrix containing the eigenvalues ​​after eigenvalue decomposition of R, and U is the matrix containing the eigenvectors after eigenvalue decomposition of R. Use Algorithm 1 to solve equation (27) and obtain the optimal solution x opt , then at this time x opt The first N terms of * opt , the last item is t opt , that is, x opt =[φ * opt t opt ] T , T means taking the transpose, then the objective function To ensure the consistency of the objective function, the modified optimal transmit beamforming vector is φ * opt t * opt ; Finally, the channel gain between the energy transmitter and the energy harvesting circuit is optimized in each time slot using the following iterative algorithm 2: 2a: Initialization: Set a feasible initial solution φ (0) , iteration index j = 0; 2b: Update the transmit beamforming vector: Given the value of φ at the jth iteration φ (j) , use Algorithm 1 to find the optimal f obtained in the j+1th iteration, that is, f (j+1) ; 2c: Update the reflection beamforming vector: Use the optimized f (j+1) , use Algorithm 1 to determine the current Then update φ (j+1) is φ * opt t * opt ; 2d: Iteration: If the convergence condition is met, go to step 2e, otherwise increase j and return to step 2b; 2e: Output: Returns the optimized f (j+1) and φ (j+1) .

5. A wireless charging method for unmanned inspection equipment of a power system according to claim 3, characterized in that: The power distribution ratio α l [n] and the transmission power p t [n] merged into a single variable y l [n] = α l [n]p t [n](h opt [n]) 2 ,h opt [n] is the optimal channel gain for each time slot; the power transmission efficiency in step 2.2) is expressed as Where P c is the other energy loss at ET; Using the Dinkelbach algorithm, equation (31) is transformed into the following equation (32): Where η opt It is an auxiliary parameter introduced in the Dinkelbach algorithm; Further introduce auxiliary variable τ l [n] non-convex Convert to convex form and use the SCA algorithm to transform equation (32) into the following equation (33): in Constraint (33b) is transformed using a first-order Taylor expansion into where τ l [n] (j) is the current value of the corresponding variable at the jth iteration; The power allocation problem established in equation (33) is solved by Algorithm 3. The detailed steps of Algorithm 3 are as follows: 3a: Initialization: Set a feasible initial solution {y l [n] (0) }, η opt , iteration index j = 0, stop threshold ∈ η and ∈ obj ; 3b: External loop: When η opt (j+1) -η opt (j) >∈ η , loop to perform the following operations; 3c: Inner loop: When the difference between the objective function of the j+1th and jth equation (33) is greater than ∈ obj , loop to perform the following operations; 3d: Pass Update τ l [n] (j) ; 3e: Solve the convex problem of equation (33) and express its optimal solution as τ l [n] * and l [n]*; 3f: Update y l [n] j+1 = y l [n] * ; 3g: When the difference between the objective function of the j+1th and jth equations (33) is less than ∈ obj , end the inner loop, otherwise return to step 3c; 3h: Update η opt ; 3i: when η opt (j+1) -η opt (j) <∈ η , end the outer loop, otherwise return to step 3b; 3j: Output: Returns the optimized y and calculates the optimal power allocation ratio and transmit power for each time slot.

Images