Soil moisture sensor distribution optimization method and device based on wind turbine environment

Abstract

A soil moisture sensor distribution optimization method and device based on a wind turbine environment, the method estimates and locates the positions of unknown position nodes in a soil moisture sensor network through an unconstrained optimization algorithm based on the position information of known position nodes in the soil moisture sensor network; constructs an objective function; obtains the gradient of the objective function relative to the position of the unknown position node through iterative calculation; obtains the Barzilai-Borwein step length by setting a strategy calculation; updates the position of the unknown position node according to the obtained Barzilai-Borwein step length; sets the stop condition of the iterative calculation of the objective function, and when the stop condition is met, stops the iterative calculation to obtain the position of the unknown position node. The present invention can determine a better sensor layout according to different detection targets and environmental conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of humidity sensor layout, and in particular to a soil humidity sensor distribution optimization method and device based on a wind turbine environment. Background Art

[0002] Uneven distribution of rainwater can cause uneven settlement of the tower, which can easily cause the wind turbine tower to tilt, creating safety hazards. It is necessary to use soil moisture sensors to set safety warnings. Wind turbine towers are generally built in open fields, mountainous areas or coastal areas. The soil moisture conditions in these areas may be affected by many factors, such as climate conditions, topography, vegetation cover, wind turbine rotation, etc. Therefore, in these environments, the distribution method of soil moisture sensors needs to take these factors into account to ensure accurate and efficient warnings.

[0003] The verticality detection of wind turbine towers is a very important task in the daily operation and maintenance of wind power plants. During the operation of wind turbine towers, due to long-term reciprocating cyclic loads and the influence of external forces, the foundation at the bottom of the tower may become loose, and the tower body may tilt and bend from top to bottom, which may eventually cause abnormal operation or tipping of the wind turbine. In order to prevent the occurrence of such safety accidents and reduce the loss of national property, it is necessary to conduct regular safety inspections on the verticality of wind turbine towers.

[0004] Wireless sensor networks (WSNs) are self-organizing networks composed of a large number of tiny sensors. Sensor nodes in WSNs can detect physical information in the monitoring area and process data, and transmit the processed data to the base station in the form of wireless communication. In recent years, wireless communication technology, nanotechnology and micro-electromechanical system technology have developed rapidly. These technologies have reduced the size, power consumption and cost of sensors, allowing sensors to be deployed on a large scale. WSNs have many applications, such as patient detection in medical applications, volcano detection in environmental applications, water use detection in household applications, etc. In the above wide applications, the detected information needs to be combined with the location of the sensor node to provide more effective data information. Therefore, the positioning method of sensor nodes in WSNs has received widespread attention.

[0005] At present, the main technical issues for wind turbine tower soil moisture monitoring include:

[0006] First, China's Beidou satellite navigation system module or global positioning system module is embedded in the sensor for positioning. However, these modules are costly and power-hungry, and are not suitable for large-scale wireless sensor networks.

[0007] Second, the distributed positioning method effectively reduces the communication cost and computational complexity, has good scalability, and can be used in large-scale wireless sensor networks. However, its disadvantage is that it uses less information and the positioning accuracy will be reduced.

[0008] Third, the commonly used genetic algorithms are prone to premature convergence, that is, they focus on a small area of ​​the solution space too early in the search process, thus missing the global optimal solution.

[0009] Fourth, compared with other traditional optimization methods, genetic algorithms are usually less efficient because they require multiple generations of iterations to gradually approach the optimal solution.

[0010] Therefore, how to invent a method to optimize sensor distribution and determine a better sensor layout according to different detection targets and environmental conditions has become an urgent problem to be solved. Summary of the invention

[0011] To this end, the present invention provides a soil moisture sensor distribution optimization method and device based on a wind turbine environment, which can determine a better sensor layout according to different detection targets and environmental conditions.

[0012] In order to achieve the above object, the present invention provides the following technical solution: a soil moisture sensor distribution optimization method based on a wind turbine environment, comprising:

[0013] According to the location information of the known location nodes in the soil moisture sensor network, the location of the unknown location nodes in the soil moisture sensor network is estimated and located through an unconstrained optimization algorithm;

[0014] According to the estimated position information of the unknown position node, construct an objective function; the objective function is defined as the position function of all the unknown position nodes in the soil moisture sensor network;

[0015] Obtaining the gradient of the objective function relative to the position of the unknown position node by iterative calculation;

[0016] The Barzilai-Borwein step size is obtained by setting the strategy calculation;

[0017] Update the position of the unknown position node according to the obtained Barzilai-Borwein step size;

[0018] A stop condition for the iterative calculation of the objective function is set, and when the stop condition is met, the iterative calculation of the objective function is stopped to obtain the position of the unknown position node.

[0019] As a preferred solution of the soil moisture sensor distribution optimization method based on the wind turbine environment, in the process of estimating and locating the position of the unknown position node in the soil moisture sensor network by the unconstrained optimization algorithm, the expression of the unconstrained optimization algorithm is:

[0020]

[0021] In the formula, x i 、x j are the i-th and j-th nodes in the unknown position nodes, i=1,2,3,…,n; j∈N i ; N i is the neighbor node set of node i in the unknown location node; m ij and m iλ All are weights; M i is the set of nodes with known locations; ij and q iλ It is a noisy distance measurement.

[0022] As a preferred solution of the soil moisture sensor distribution optimization method based on the wind turbine environment, the expression of the objective function is:

[0023]

[0024] Wherein, x is a vector containing the positions of all the unknown position nodes; m ij and m ia is the weight of the corresponding edge; q ij and q ia is the measured distance.

[0025] As a preferred solution of the soil moisture sensor distribution optimization method based on the wind turbine environment, the gradient calculation formula of the objective function relative to the position of the unknown position node is:

[0026]

[0027] In the formula, w ij and w ia All are environmental adaptability weights; d ij and d ia are the distances of unknown nodes obtained through measurement.

[0028] As a preferred solution of the soil moisture sensor distribution optimization method based on the wind turbine environment, in the process of calculating and obtaining the Barzilai-Borwein step size through the setting strategy, the setting strategy includes two calculation formulas BB1 and BB2, which are respectively:

[0029] BB1 step length calculation formula:

[0030]

[0031] BB2 step length calculation formula:

[0032]

[0033] in,

[0034] s k =x k -x k-1

[0035]

[0036] In the formula, is the vector difference between adjacent node positions.

[0037] As a preferred solution of the soil moisture sensor distribution optimization method based on the wind turbine environment, the expression for updating the position of the unknown position node according to the obtained Barzilai-Borwein step size is:

[0038]

[0039] In the formula, x i,k+1 is the updated position of the unknown position node; i,k is the position of the unknown position node before updating; k+1 is the Barzilai-Borwein step length; is the gradient.

[0040] As a preferred solution of the soil moisture sensor distribution optimization method based on a wind turbine environment, the stop conditions include: the modulus of the gradient is less than a set threshold value v, or the position change in continuous iterations is less than a set small amount δ, or a preset number of iterations K is reached.

[0041] As a preferred solution of the soil moisture sensor distribution optimization method based on the wind turbine environment, in the process of calculating the gradient of the objective function relative to the position of the unknown position node, the adjustment of the environmental adaptability weight is to dynamically adjust the environmental adaptability weight in combination with real-time soil moisture monitoring data to adapt to environmental changes; the calculation expression of the environmental adaptability weight is:

[0042]

[0043] Where f is an environmental impact function; env(x i ,x j ) is the environmental condition between sensors i and j.

[0044] As a preferred solution of the soil moisture sensor distribution optimization method based on the wind turbine environment, in the process of updating the position of the unknown position node according to the obtained Barzilai-Borwein step size, the Barzilai-Borwein step size is selected and adjusted according to the terrain where the wind turbine tower is located through corresponding terrain environmental parameters; the terrain environmental parameters include: terrain parameters, climate parameters and vegetation density.

[0045] The present invention also provides a soil moisture sensor distribution optimization device based on a wind turbine environment, based on the above soil moisture sensor distribution optimization method based on a wind turbine environment, comprising:

[0046] The node estimated position information acquisition module is used to estimate and locate the position of the unknown position node in the soil moisture sensor network through an unconstrained optimization algorithm according to the position information of the known position node in the soil moisture sensor network;

[0047] An objective function construction module is used to construct an objective function according to the estimated position information of the unknown position node; the objective function is defined as a position function of all the unknown position nodes in the soil moisture sensor network;

[0048] A gradient calculation module, used for obtaining the gradient of the objective function relative to the position of the unknown position node through iterative calculation;

[0049] The Barzilai-Borwein step length calculation module is used to obtain the Barzilai-Borwein step length by setting the strategy calculation;

[0050] A node position updating module, used for updating the position of the unknown position node according to the obtained Barzilai-Borwein step size;

[0051] The node position acquisition module is used to set the stopping condition of the iterative calculation of the objective function. When the stopping condition is met, the iterative calculation of the objective function is stopped to obtain the position of the unknown position node.

[0052] As a preferred solution of the soil moisture sensor distribution optimization device based on the wind turbine environment, in the node estimated position information acquisition module, in the process of estimating and locating the position of the unknown position node in the soil moisture sensor network by the unconstrained optimization algorithm, the expression of the unconstrained optimization algorithm is:

[0053]

[0054] In the formula, x i 、x jare the i-th and j-th nodes in the unknown position nodes, i=1,2,3,…,n; j∈N i ; N i is the neighbor node set of node i in the unknown location node; m ij and m iλ All are weights; M i is the set of nodes with known locations; ij and q iλ It is a noisy distance measurement.

[0055] As a preferred solution of the soil moisture sensor distribution optimization device based on the wind turbine environment, in the objective function construction module, the expression of the objective function is:

[0056]

[0057] Wherein, x is a vector containing the positions of all the unknown position nodes; m ij and m ia is the weight of the corresponding edge; q ij and q ia is the measured distance.

[0058] As a preferred solution of the soil moisture sensor distribution optimization device based on the wind turbine environment, in the gradient calculation module, the gradient calculation formula of the objective function relative to the position of the unknown position node is:

[0059]

[0060] In the formula, w ij and w ia All are environmental adaptability weights; d ij and d ia are the distances of unknown nodes obtained through measurement.

[0061] As a preferred solution of the soil moisture sensor distribution optimization device based on the wind turbine environment, in the Barzilai-Borwein step length calculation module, in the process of calculating and obtaining the Barzilai-Borwein step length through the setting strategy, the setting strategy includes two calculation formulas BB1 and BB2, which are respectively:

[0062] BB1 step length calculation formula:

[0063]

[0064] BB2 step length calculation formula:

[0065]

[0066] in,

[0067] s k =x k -x k-1

[0068]

[0069] In the formula, is the vector difference between adjacent node positions.

[0070] As a preferred solution of the soil moisture sensor distribution optimization device based on the wind turbine environment, in the node position update module, the expression for updating the position of the unknown position node according to the obtained Barzilai-Borwein step size is:

[0071]

[0072] In the formula, x i,k+1 is the updated position of the unknown position node; i,k is the position of the unknown position node before updating; k+1 is the Barzilai-Borwein step length; is the gradient.

[0073] As a preferred solution for the soil moisture sensor distribution optimization device based on a wind turbine environment, in the node position acquisition module, the stop conditions include: the modulus of the gradient is less than a set threshold ∈, or the position change in continuous iterations is less than a set small amount δ, or the preset number of iterations K is reached.

[0074] As a preferred solution of the soil moisture sensor distribution optimization device based on the wind turbine environment, in the gradient calculation module, in the process of calculating the gradient of the objective function relative to the position of the unknown position node, the adjustment of the environmental adaptability weight is to dynamically adjust the environmental adaptability weight in combination with real-time soil moisture monitoring data to adapt to environmental changes; the calculation expression of the environmental adaptability weight is:

[0075]

[0076] Where f is an environmental impact function; env(x i ,x j ) is the environmental condition between sensors i and j.

[0077] As a preferred solution of the soil moisture sensor distribution optimization device based on the wind turbine environment, in the node position update module, in the process of updating the position of the unknown position node according to the obtained Barzilai-Borwein step size, the Barzilai-Borwein step size is selected and adjusted according to the terrain where the wind turbine tower is located through corresponding terrain environment parameters; the terrain environment parameters include: terrain parameters, climate parameters and vegetation density.

[0078] The present invention has the following advantages: according to the position information of the known position nodes in the soil moisture sensor network, the position of the unknown position nodes in the soil moisture sensor network is estimated and located by an unconstrained optimization algorithm; according to the estimated position information of the unknown position nodes, an objective function is constructed; the objective function is defined as the position function of all the unknown position nodes in the soil moisture sensor network; the gradient of the objective function relative to the position of the unknown position nodes is obtained by iterative calculation; the Barzilai-Borwein step length is obtained by setting a strategy calculation; the position of the unknown position node is updated according to the obtained Barzilai-Borwein step length; the stop condition of the iterative calculation of the objective function is set, and when the stop condition is met, the iterative calculation of the objective function is stopped to obtain the position of the unknown position node. The present invention has high positioning accuracy and low computational complexity. According to the detection target and environmental conditions, a better sensor layout can be found through a mathematical model algorithm to achieve more efficient and accurate detection. Through a multi-camera network, double protection is provided for the safe and reliable operation of wind power towers. The tilt of the wind turbine tower can be observed intuitively in the form of images or videos, avoiding the waste of manpower and material resources caused by misjudgment of soil moisture sensors due to special climate reasons. China's wind farms are divided into wasteland wind farms, mountain wind farms, and seaside wind farms. In view of the different terrains of the three wind farms, the present invention proposes different detection thresholds and step thresholds to make the soil moisture test results more accurate and effective. The present invention fully considers the variable environmental conditions of the wind turbine tower, such as terrain, climate, and vegetation, and based on the Barzilai-Borwein gradient method and distributed positioning algorithm, it realizes efficient and accurate positioning of unknown node positions in the sensor network, and can provide double protection through camera interconnection to avoid data anomalies caused by extreme special weather. This method not only improves the accuracy and efficiency of soil moisture monitoring of wind turbine towers, but also significantly enhances the robustness and applicability of the algorithm through a dynamic adjustment mechanism that adapts to environmental changes. Through practical application and testing, this method has been proven to be able to provide reliable soil moisture monitoring in a variety of complex environments, providing strong technical support for the safety management and operation and maintenance of wind power plants. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0080] The structures, proportions, sizes, etc. illustrated in this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with the technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantial technical significance. Any structural modification, change in proportion or adjustment of size shall still fall within the scope of the technical contents disclosed in the present invention without affecting the effects and purposes that can be achieved by the present invention.

[0081] Figure 1 This is a schematic flow chart of a method for optimizing soil moisture sensor distribution based on a wind turbine environment provided in Embodiment 1 of the present invention;

[0082] Figure 2 A schematic diagram of sensor node distribution in a soil moisture sensor distribution optimization method based on a wind turbine environment provided in Embodiment 1 of the present invention;

[0083] Figure 3 It is a flow chart of iterative calculation of the objective function in the soil moisture sensor distribution optimization method based on the wind turbine environment provided in Example 1 of the present invention;

[0084] Figure 4 A schematic diagram of a high mountain wind field in the soil moisture sensor distribution optimization method based on a wind turbine environment provided in Example 1 of the present invention;

[0085] Figure 5 A schematic diagram of a plain wind field in the soil moisture sensor distribution optimization method based on a wind turbine environment provided in Example 1 of the present invention;

[0086] Figure 6 A schematic diagram of a coastal wind farm in the soil moisture sensor distribution optimization method based on a wind turbine environment provided in Example 1 of the present invention;

[0087] Figure 7 This is a schematic diagram of the distribution of sensor measuring points in a high mountain wind field in the soil moisture sensor distribution optimization method based on a wind turbine environment provided in Example 1 of the present invention;

[0088] Figure 8 This is a schematic diagram of the distribution of sensor measuring points in a plain wind field in the soil moisture sensor distribution optimization method based on a wind turbine environment provided in Example 1 of the present invention;

[0089] Fig. 9 This is a schematic diagram of the distribution of sensor measuring points in a coastal wind farm in the soil moisture sensor distribution optimization method based on a wind turbine environment provided in Example 1 of the present invention;

[0090] Fig.10 This is a schematic diagram of the architecture of the soil moisture sensor distribution optimization device based on the wind turbine environment provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0091] The following is a description of the implementation of the present invention by specific embodiments. People familiar with the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. Obviously, the described embodiments are 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.

[0092] Example 1

[0093] See also Figure 1 Embodiment 1 of the present invention provides a soil moisture sensor distribution optimization method based on a wind turbine environment, comprising the following steps:

[0094] S1. According to the location information of the known location nodes in the soil moisture sensor network, the location of the unknown location nodes in the soil moisture sensor network is estimated and located by an unconstrained optimization algorithm;

[0095] S2. constructing an objective function according to the estimated position information of the unknown position node; the objective function is defined as the position function of all the unknown position nodes in the soil moisture sensor network;

[0096] S3. Obtaining the gradient of the objective function relative to the position of the unknown position node through iterative calculation;

[0097] S4, obtain the Barzilai-Borwein step size by setting the strategy calculation;

[0098] S5. Update the position of the unknown position node according to the obtained Barzilai-Borwein step size;

[0099] S6. Setting a stop condition for the iterative calculation of the objective function. When the stop condition is met, stopping the iterative calculation of the objective function and obtaining the position of the unknown position node.

[0100] In this embodiment, in step S1, based on the position information of the known position nodes in the soil moisture sensor network, the positions of the unknown position nodes in the soil moisture sensor network are estimated and located by an unconstrained optimization algorithm;

[0101] Specifically, the soil moisture sensor network (wireless sensor network WSN) is a self-organizing network that can be described by an undirected graph G = (V, E). V represents the set of nodes and E represents the set of edges. V1 = {1, 2, 3, ..., n} represents the set of unknown position nodes LU in the soil moisture sensor network; V2 = {1, 2, 3, ..., m} represents the set of known position nodes LA in the soil moisture sensor network.

[0102] Where E = {e iλ , e ij}, (i=1,2,3,…,n; j=1,2,3,…,n; λ=1,2,3,…,m). Among them, e iλ Indicates that LU node i and LA node λ can communicate directly; e ij It means that LU node i and LU node j can communicate directly.

[0103] Due to the limitation of sensor node power, sensor nodes can only communicate directly with nodes within the communication radius R. In this way, the undirected graph of the wireless sensor network can be divided into partially overlapping subgraphs. The WSN is divided into partially overlapping subgraphs with the LU node as the center:

[0104]

[0105] G S =(V S ,E S )

[0106] Where V S =V S1 ∪V S2 , V S1 Represents subgraph G S The set of LU nodes in V S2 Represents subgraph G S LA node set; E s Represents the subgraph G S The set of edges between midpoints.

[0107] like Figure 2 As shown in the figure, the sensor nodes are distributed in a 1×1 unit area. The red dots represent LU nodes; the solid circles represent the partially overlapping subgraphs divided by the LU node as the center; and R is the radius.

[0108] A large number of sensor nodes are deployed in the monitoring area Rd (d ≥ 1) dimensional space, and these nodes constitute a wireless sensor network. There are N nodes in the wireless sensor network, including m LA nodes and n LU nodes. Consider that the sensor nodes are deployed in a two-dimensional Euclidean space of d = 2. The positioning problem can be described as: using the location information of m LA nodes, the node's neighbor information and the noisy distance information between nodes, estimate the location of n LU nodes x i ; (i=1,2,3,…,n) position. Therefore, the node positioning problem in the soil moisture sensor network can be reduced to an unconstrained optimization problem:

[0109]

[0110] In the formula, x i 、x j are the i-th and j-th nodes in the unknown position nodes, i=1,2,3,…,n; j∈N i ; N i is the neighbor node set of node i in the unknown location node; m ij and m iλ All are weights; M i is the set of nodes with known locations; ij and q iλ It is a noisy distance measurement. Therefore, a larger weight is set for the distance measurement with higher credibility; conversely, a smaller weight is set for the distance measurement with lower credibility.

[0111] In this embodiment, in step S2, an objective function is constructed based on the estimated position information of the unknown position node; the objective function is defined as the position function of all the unknown position nodes in the soil moisture sensor network, representing the total positioning error of the entire network.

[0112] Specifically, the expression of the objective function is:

[0113]

[0114] Wherein, x is a vector containing the positions of all the unknown position nodes; m ij and m ia is the weight of the corresponding edge; q ij and q ia is the measured distance.

[0115] In this embodiment, in step S3, the gradient of the objective function relative to the position of the unknown position node is obtained by iterative calculation;

[0116] Specifically, the gradient calculation formula of the objective function relative to the position of the unknown position node is:

[0117]

[0118] In the formula, w ij and w ia All are environmental adaptability weights; d ij and d ia are the distances of unknown nodes obtained through measurement.

[0119] In the process of calculating the gradient of the objective function relative to the position of the unknown position node, the adjustment of the environmental adaptability weight is to dynamically adjust the environmental adaptability weight in combination with the real-time soil moisture monitoring data to adapt to environmental changes; the calculation expression of the environmental adaptability weight is:

[0120]

[0121] Where f is an environmental impact function; env(x i ,x j ) is the environmental condition between sensors i and j.

[0122] In this embodiment, in step S4, the Barzilai-Borwein step size is obtained by setting a strategy calculation;

[0123] Specifically, the setting strategy includes two calculation formulas, BB1 and BB2, which are:

[0124] BB1 step length calculation formula:

[0125]

[0126] BB2 step length calculation formula:

[0127]

[0128] in,

[0129] s k =x k -x k-1

[0130]

[0131] In the formula, is the vector difference between adjacent node positions.

[0132] In this embodiment, in step S5, the position of the unknown position node is updated according to the obtained Barzilai-Borwein step size;

[0133] Specifically, the expression is:

[0134]

[0135] In the formula, x i,k+1 is the updated position of the unknown position node; i,k is the position of the unknown position node before updating; k+1 is the Barzilai-Borwein step length; is the gradient.

[0136] In this embodiment, a local sensitivity analysis is also performed on each LU node to determine a local strategy for step size adjustment:

[0137] The sensitivity S of node i i When it is larger, the step size needs to be reduced to avoid over-adjustment.

[0138]

[0139] In this embodiment, it also includes: selecting and adjusting the Barzilai-Borwein step size according to the terrain where the wind turbine tower is located through corresponding terrain environment parameters; the terrain environment parameters include: terrain parameters, climate parameters and vegetation density.

[0140] Specifically, the step size of the Barzilai-Borwein gradient method will be different according to the different terrains where the wind turbine tower is located, and the soil moisture measurement data can be more accurate. According to the specific environment around the wind turbine tower (such as terrain, vegetation, climate, etc.), the weight in the objective function is dynamically adjusted to reflect the credibility of the measurement distance in different environments.

[0141] Scene 1: Alpine wind farm

[0142] Step length adjustment of high mountain wind farm: Figure 4 As shown in the figure, in mountainous areas with large terrain undulations, the communication distance will be limited. The soil layer is thin and there are mostly rocks or gravel underneath. Sensors need to be distributed in the upper layer. The step length needs to be adjusted according to the hardness and stability of the geology to ensure that the sensor network can adapt to different geological conditions. Therefore, a terrain adaptation factor is introduced, which can adjust the step length based on the slope and geological hardness.

[0143] Terrain adaptation factor:

[0144] η=η slope η geology

[0145] Where η slope is the slope adaptation factor, which can be calculated based on the slope s. The calculation formula is:

[0146]

[0147] Where β is a coefficient determined according to the terrain.

[0148] η geology It is the geological adaptation factor, which can be calculated based on the geological hardness G. The calculation formula is:

[0149]

[0150] Where γ is the geological hardness influence coefficient.

[0151] Extreme weather such as strong winds and snow may occur in the mountains, which have a great impact on the data measured by the sensor. It is necessary to increase the step size to quickly adapt to environmental changes. You can introduce:

[0152]

[0153] Among them, h is the climate parameter influence function.

[0154] h(climate parameter)=1+γ*(k1+k2)

[0155] Among them, γ is the climate influence coefficient, k1 and k2 represent humidity and wind speed respectively.

[0156] Scene 2: Plain wind farm

[0157] Step size adjustment of plain wind farm: Figure 5 As shown in the figure, the terrain of plain wind farms is relatively flat, and the soil layer is usually thick. The soil is mostly loam, clay or sand, with good bearing capacity. At the same time, the geological conditions are relatively simple, and there are usually not many rocks exposed.

[0158] In plain areas, the initial step size can be set relatively large because the terrain has little impact on the sensor network. Considering the density and moisture of the soil, the geological influence factor η is introduced g To adjust the step size:

[0159]

[0160] Where η g It can be adjusted according to the density and moisture of the soil.

[0161] The gradient threshold can also be set relatively small to ensure that the algorithm converges quickly to the exact solution:

[0162] ε=ε0η g

[0163] The vegetation in plain wind farms is usually dense, and vegetation coverage will also affect the measurement of sensor data. The step size needs to be adjusted according to the vegetation density. The formula can be introduced:

[0164]

[0165] Where v is the vegetation density influence function.

[0166]

[0167] Where k3 is the vegetation coverage rate.

[0168] Scenario 3: Coastal wind farm

[0169] Coastal wind farm step adjustment: Figure 6 As shown, coastal wind farms may include sand dunes, mudflats, or rocky coasts. The soil is mostly sandy soil with a lot of salt. The geology mostly includes rocks, sediments, and possible coral reef structures. In coastal wind farms, especially in areas facing the sea, soil moisture becomes more humid due to salt spray and higher humidity brought by the sea breeze. This environmental factor has a significant impact on the distribution and positioning algorithm of soil moisture sensors, so the adjustment of the step size needs to be considered from the following aspects.

[0170] First, the weight adjustment of environmental adaptability:

[0171]

[0172] In the formula, is the initial weight; f is an environmental impact function that takes into account humidity and wind speed.

[0173] The second is the step length adjustment strategy:

[0174]

[0175] In the formula, a k is the current step size, β is an adjustment coefficient, is the currently measured humidity, is the desired soil moisture level.

[0176] The third is dynamic threshold adjustment:

[0177]

[0178] Where ε0 is the initial gradient threshold, k a is the range over which humidity may vary.

[0179] Fourth, local sensitivity analysis:

[0180]

[0181] When the sensitivity S of node i i When it is larger, the step size needs to be reduced to avoid over-adjustment, which can bias the measured data, especially in areas with high humidity.

[0182] In addition to adjusting the step length according to different wind fields, the step length of environmental perception should also be limited. To prevent the step length from being too large or too small, the upper and lower limits of the step length need to be set according to the corresponding environmental parameters.

[0183] a min ≤a k+1 ≤a max

[0184] In the formula, a min and a max are the lower and upper limits of the step size determined according to environmental parameters.

[0185] a min =a base ·exp(-λ min env)

[0186] a max =a base ·exp(-λ max env)

[0187] For a changing environment, a comprehensive evaluation model needs to be established to consider all environmental factors, such as terrain, climate, and vegetation cover. The model combines the impact of these factors on the performance of the sensor network through a weighted sum.

[0188] k4 (comprehensive influencing factor) = w1g (topographic parameters) + w2h (climate parameters) + w3v (vegetation density)

[0189] Where w1, w2, and w3 are the weights of different environmental factors, which can be adjusted according to the degree of influence of environmental factors on sensor performance.

[0190] g (terrain parameter) is defined as the terrain influence function, which can adjust the step size of the sensor according to the complexity of the terrain.

[0191] h (climate parameter) is defined as the climate impact function, which can adjust the sensor step size according to the changes in climate conditions.

[0192] Define v (vegetation density) as the vegetation influence function, and adjust the sensor step size according to the density of vegetation coverage

[0193] In this embodiment, in step S6, a stop condition for the iterative calculation of the objective function is set. When the stop condition is met, the iterative calculation of the objective function is stopped to obtain the position of the unknown position node.

[0194] Specifically, the expression of the stop condition is:

[0195]

[0196] The meaning of the expression of the stop condition is: the modulus of the gradient is less than the set threshold ∈, or the position change in continuous iterations is less than the set small amount δ, or the preset number of iterations K is reached.

[0197] like Figure 3 As shown, the position of the unknown node LU is obtained by iterative calculation of the objective function.

[0198] The wind farms in the three scenarios embodied in step S5 also need to adjust the stop conditions according to environmental factors and algorithm performance to ensure that the algorithm can achieve optimal convergence performance in different environments.

[0199] Specifically, the detailed stopping conditions of high mountain wind farms are as follows:

[0200] Gradient threshold ε hl The adjustment formula is:

[0201] ε hl =ε0·exp(-θ·g)

[0202] Where θ is the slope inclination angle;

[0203] Considering the extreme climate and complex terrain in mountainous areas, the gradient threshold needs to be set lower to ensure that the algorithm can capture small errors caused by terrain changes.

[0204] Position change threshold δ hl The formula is:

[0205] δ hl =δ0·(1+h)

[0206] Due to the influence of strong wind and snow accumulation, the sensor position may change, and the position change threshold needs to be adjusted according to the terrain stability.

[0207] Specifically, the detailed stopping conditions for coastal wind farms are as follows:

[0208] Gradient threshold ε co The adjustment formula is:

[0209]

[0210] The high humidity and salt content in coastal areas will greatly affect the sensor readings. The gradient threshold needs to take these environmental factors into consideration and make appropriate adjustments.

[0211] Position change threshold δ co The formula is:

[0212] δ co =δ0·min(1,ν 侵蚀速率 )

[0213] Erosion and sedimentation in coastal areas may cause sensor positions to shift, and smaller position change thresholds need to be set to accommodate such changes.

[0214] Specifically, the stop conditions for the plain wind farm are as follows:

[0215] Gradient threshold ε pl The adjustment formula is:

[0216] ε pl =ε0·(1+K pl ν)

[0217] In the formula, K pl It is an adjustment factor used to balance the influence of basic threshold and terrain factors.

[0218] The terrain in the plain area is flat and there are fewer gravels, so the gradient threshold can be set relatively high.

[0219] Position change threshold δ pl The formula is:

[0220] δ pl =δ0·max(1,β 土壤密实度 )

[0221] The soil density and moisture in plain areas are relatively uniform, and the position change threshold can be set relatively large to reduce the number of iterations.

[0222] Specific, complex terrain refinement stopping conditions:

[0223] Gradient threshold: ε = ε0*k4

[0224] Position change threshold: δ = δ0*k4

[0225] Number of iterations: k = k0 + z0 (environment complexity coefficient)

[0226] Where ε0 and δ0 are the initial gradient threshold and position change threshold, which can be set according to the initial performance of the algorithm. z0 is the environmental complexity coefficient, which can be used to adjust the number of iterations according to the complexity of the environment to ensure that the algorithm can be fully iterated in a complex environment.

[0227] After iterative calculation by the present invention, the sensor measuring point distribution optimization diagrams of the three terrains can be obtained.

[0228] Distribution map of high mountain wind field measurement points: Through algorithm optimization, soil moisture sensors will be more concentrated in key areas such as ridges, mountain tops, and leeward areas of mountains, which are easily affected by climate and terrain. At the same time, their regional layout will avoid exposed rocks and cliffs, such as Figure 7 shown.

[0229] Distribution of wind field measurement points in plains: Through algorithm optimization, a more regular grid layout will be adopted in plain areas to facilitate management and data collection. At the same time, the density of sensors will be increased in areas with dense vegetation, while the number of sensors will be reduced in open areas. The distribution is as follows: Figure 8 shown.

[0230] Distribution map of wind farm measurement points along the coast: Through algorithm optimization, the distribution density of sensors will be adjusted according to the different salinity of the soil. Areas with high salinity will require a denser sensor network. At the same time, sensors are set up behind areas that may be eroded by sea breeze and salt to serve as a monitoring buffer for erosion and sedimentation. Fig. 9 As shown in Figure 2, soil moisture sensors are more densely distributed near the coastline.

[0231] In summary, the present invention estimates and locates the positions of unknown position nodes in the soil moisture sensor network through an unconstrained optimization algorithm based on the position information of known position nodes in the soil moisture sensor network; constructs an objective function based on the estimated position information of the unknown position nodes; the objective function is defined as the position function of all the unknown position nodes in the soil moisture sensor network; obtains the gradient of the objective function relative to the position of the unknown position node through iterative calculation; obtains the Barzilai-Borwein step length by setting a strategy calculation; updates the position of the unknown position node according to the obtained Barzilai-Borwein step length; sets the stop condition of the iterative calculation of the objective function, and when the stop condition is met, stops the iterative calculation of the objective function to obtain the position of the unknown position node. The present invention has high positioning accuracy and low computational complexity. According to the detection target and environmental conditions, a better sensor layout can be found through a mathematical model algorithm to achieve more efficient and accurate detection. Through a multi-camera network, double protection is provided for the safe and reliable operation of wind power towers. The tilt of the wind turbine tower can be observed intuitively in the form of images or videos, avoiding the waste of manpower and material resources caused by misjudgment of soil moisture sensors due to special climate reasons. China's wind farms are divided into wasteland wind farms, mountain wind farms, and seaside wind farms. In view of the different terrains of the three wind farms, the present invention proposes different detection thresholds and step thresholds to make the soil moisture test results more accurate and effective. The present invention fully considers the variable environmental conditions of the wind turbine tower, such as terrain, climate, and vegetation, and based on the Barzilai-Borwein gradient method and distributed positioning algorithm, it realizes efficient and accurate positioning of unknown node positions in the sensor network, and can provide double protection through camera interconnection to avoid data anomalies caused by extreme special weather. This method not only improves the accuracy and efficiency of soil moisture monitoring of wind turbine towers, but also significantly enhances the robustness and applicability of the algorithm through a dynamic adjustment mechanism that adapts to environmental changes. Through practical application and testing, this method has been proven to be able to provide reliable soil moisture monitoring in a variety of complex environments, providing strong technical support for the safety management and operation and maintenance of wind power plants.

[0232] It should be noted that the method of the embodiment of the present disclosure can be performed by a single device, such as a computer or a server. The method of the present embodiment can also be applied in a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present disclosure, and the multiple devices will interact with each other to complete the described method.

[0233] It should be noted that the above describes some embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0234] Example 2

[0235] See also Fig.10 Embodiment 2 of the present invention further provides a soil moisture sensor distribution optimization device based on a wind turbine environment, including:

[0236] The node estimated position information acquisition module 001 is used to estimate and locate the position of the unknown position node in the soil moisture sensor network through an unconstrained optimization algorithm according to the position information of the known position node in the soil moisture sensor network;

[0237] The objective function construction module 002 is used to construct an objective function according to the estimated position information of the unknown position node; the objective function is defined as the position function of all the unknown position nodes in the soil moisture sensor network;

[0238] A gradient calculation module 003, used for obtaining the gradient of the objective function relative to the position of the unknown position node through iterative calculation;

[0239] The Barzilai-Borwein step length calculation module 004 is used to obtain the Barzilai-Borwein step length by setting a strategy calculation;

[0240] A node position updating module 005 is used to update the position of the unknown position node according to the obtained Barzilai-Borwein step size;

[0241] The node position acquisition module 006 is used to set the stopping condition of the iterative calculation of the objective function. When the stopping condition is met, the iterative calculation of the objective function is stopped to obtain the position of the unknown position node.

[0242] In this embodiment, in the node estimated position information acquisition module 001, in the process of estimating and locating the position of the unknown position node in the soil moisture sensor network by using the unconstrained optimization algorithm, the expression of the unconstrained optimization algorithm is:

[0243]

[0244] In the formula, xi 、x j are the i-th and j-th nodes in the unknown position nodes, i=1,2,3,…,n; j∈N i ;

[0245] N i is the neighbor node set of node i in the unknown location node; m ij and m iλ All are weights; M i is the set of nodes with known locations; ij and q iλ It is a noisy distance measurement.

[0246] In this embodiment, in the objective function construction module 002, the expression of the objective function is:

[0247]

[0248] Wherein, x is a vector containing the positions of all the unknown position nodes; m ij and m ia is the weight of the corresponding edge; q ij and q ia is the measured distance.

[0249] In this embodiment, in the gradient calculation module 003, the gradient calculation formula of the objective function relative to the position of the unknown position node is:

[0250]

[0251] In the formula, w ij and w ia All are environmental adaptability weights; d ij and d ia are the distances of unknown nodes obtained through measurement.

[0252] In this embodiment, in the Barzilai-Borwein step length calculation module 004, in the process of calculating and obtaining the Barzilai-Borwein step length through the setting strategy, the setting strategy includes two calculation formulas BB1 and BB2, which are respectively:

[0253] BB1 step length calculation formula:

[0254]

[0255] BB2 step length calculation formula:

[0256]

[0257] in,

[0258] s k =x k -x k-1

[0259]

[0260] In the formula, is the vector difference between adjacent node positions.

[0261] In this embodiment, in the node position updating module 005, the expression for updating the position of the unknown position node according to the obtained Barzilai-Borwein step size is:

[0262]

[0263] In the formula, x i,k+1 is the updated position of the unknown position node; i,k is the position of the unknown position node before updating; k+1 is the Barzilai-Borwein step length; is the gradient.

[0264] In this embodiment, in the node position acquisition module 006, the stop condition includes: the modulus of the gradient is less than a set threshold ∈, or the position change in continuous iterations is less than a set small amount δ, or a preset number of iterations K is reached.

[0265] In this embodiment, in the gradient calculation module 003, in the process of calculating the gradient of the objective function relative to the position of the unknown position node, the adjustment of the environmental adaptability weight is to dynamically adjust the environmental adaptability weight in combination with real-time soil moisture monitoring data to adapt to environmental changes; the calculation expression of the environmental adaptability weight is:

[0266]

[0267] Where f is an environmental impact function; env(x i ,x j ) is the environmental condition between sensors i and j.

[0268] In this embodiment, in the node position updating module 006, in the process of updating the position of the unknown position node according to the obtained Barzilai-Borwein step size, the Barzilai-Borwein step size is selected and adjusted according to the terrain where the wind turbine tower is located and the corresponding terrain environment parameters; the terrain environment parameters include: terrain parameters, climate parameters and vegetation density.

[0269] It should be noted that the information interaction, execution process and other contents between the modules of the above-mentioned device are based on the same concept as the method embodiment in Example 1 of the present application, and the technical effects they bring are the same as those of the method embodiment of the present application. For specific contents, please refer to the description in the method embodiment shown in the previous part of the present application, and will not be repeated here.

[0270] Example 3

[0271] Embodiment 3 of the present invention provides a non-transitory computer-readable storage medium, in which a program code for a soil moisture sensor distribution optimization method based on a wind turbine environment is stored, and the program code includes instructions for executing embodiment 1 or any possible implementation method of the soil moisture sensor distribution optimization method based on a wind turbine environment.

[0272] The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state drive (SSD)).

[0273] Example 4

[0274] Embodiment 4 of the present invention provides an electronic device, including: a memory and a processor;

[0275] The processor and the memory communicate with each other via a bus; the memory stores program instructions that can be executed by the processor, and the processor calls the program instructions to execute the soil moisture sensor distribution optimization method based on a wind turbine environment of Example 1 or any possible implementation thereof.

[0276] Specifically, the processor can be implemented by hardware or by software. When implemented by hardware, the processor can be a logic circuit, an integrated circuit, etc.; when implemented by software, the processor can be a general-purpose processor implemented by reading software codes stored in a memory. The memory can be integrated in the processor or can be located outside the processor and exist independently.

[0277] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable systems. The computer instructions can be stored in a computer-readable storage medium, or transmitted from a computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center.

[0278] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing system, they can be concentrated on a single computing system, or distributed on a network composed of multiple computing systems, and optionally, they can be implemented by a program code executable by a computing system, so that they can be stored in a storage system and executed by the computing system, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0279] Although the present invention has been described in detail above by general description and specific embodiments, it is obvious to those skilled in the art that some modifications or improvements can be made to the present invention. Therefore, these modifications or improvements made without departing from the spirit of the present invention all belong to the scope of protection claimed by the present invention.

Claims

1. A soil moisture sensor distribution optimization method based on a wind turbine environment, characterized in that: include: According to the location information of the known location nodes in the soil moisture sensor network, the location of the unknown location nodes in the soil moisture sensor network is estimated and located through an unconstrained optimization algorithm; According to the estimated position information of the unknown position node, construct an objective function; the objective function is defined as the position function of all the unknown position nodes in the soil moisture sensor network; Obtaining the gradient of the objective function relative to the position of the unknown position node by iterative calculation; The Barzilai-Borwein step size is obtained by setting the strategy calculation; Update the position of the unknown position node according to the obtained Barzilai-Borwein step size; Setting a stop condition for the iterative calculation of the objective function, and when the stop condition is met, stopping the iterative calculation of the objective function, and obtaining the position of the unknown position node; The gradient calculation formula of the objective function relative to the position of the unknown position node is: In the formula, x i 、x j are the i-th and j-th nodes in the unknown position nodes, i=1,2,3,…,n; j∈N i ; N i is a set of neighbor nodes of node i in the unknown location node; w ij and w ia All are environmental adaptability weights; d ij and d ia are the distances of unknown nodes obtained through measurement; In the process of calculating the gradient of the objective function relative to the position of the unknown position node, the adjustment of the environmental adaptability weight is to dynamically adjust the environmental adaptability weight in combination with the real-time soil moisture monitoring data to adapt to environmental changes; the calculation expression of the environmental adaptability weight is: Where f is an environmental impact function; env(x i ,x j ) is the environmental condition between sensors i and j.

2. The soil moisture sensor distribution optimization method based on wind turbine environment according to claim 1 is characterized in that: In the process of estimating and locating the position of the unknown position node in the soil moisture sensor network by using the unconstrained optimization algorithm, the expression of the unconstrained optimization algorithm is: In the formula, m ij and m iλ All are weights; M i is the set of nodes with known locations; ij and q iλ It is a noisy distance measurement.

3. The soil moisture sensor distribution optimization method based on wind turbine environment according to claim 2 is characterized in that: The expression of the objective function is: Wherein, x is a vector containing the positions of all the unknown position nodes; m ij and m ia is the weight of the corresponding edge; q ij and q ia is the measured distance.

4. The soil moisture sensor distribution optimization method based on wind turbine environment according to claim 1 is characterized in that: In the process of calculating and obtaining the Barzilai-Borwein step size through the setting strategy, the setting strategy includes two calculation formulas, BB1 and BB2, which are respectively: BB1 step length calculation formula: BB2 step length calculation formula: in, s k =x k -x k-1 In the formula, is the vector difference between adjacent node positions.

5. The soil moisture sensor distribution optimization method based on wind turbine environment according to claim 4 is characterized in that: The expression for updating the position of the unknown position node according to the obtained Barzilai-Borwein step size is: In the formula, x i,k+1 is the updated position of the unknown position node; i,k is the position of the unknown position node before updating; k+1 is the Barzilai-Borwein step length; is the gradient.

6. The soil moisture sensor distribution optimization method based on wind turbine environment according to claim 5 is characterized in that: The stopping conditions include: the modulus of the gradient is less than a set threshold ∈, or the position change in continuous iterations is less than a set small amount δ, or a preset number of iterations K is reached.

7. The soil moisture sensor distribution optimization method based on wind turbine environment according to claim 1 is characterized in that: In the process of updating the position of the unknown position node according to the obtained Barzilai-Borwein step size, the Barzilai-Borwein step size is selected and adjusted according to the terrain where the wind turbine tower is located through corresponding terrain environmental parameters; the terrain environmental parameters include: terrain parameters, climate parameters and vegetation density.

8. A soil moisture sensor distribution optimization device based on a wind turbine environment, adopting the soil moisture sensor distribution optimization method based on a wind turbine environment according to any one of claims 1 to 7, characterized in that: include: The node estimated position information acquisition module is used to estimate and locate the position of the unknown position node in the soil moisture sensor network through an unconstrained optimization algorithm according to the position information of the known position node in the soil moisture sensor network; An objective function construction module is used to construct an objective function according to the estimated position information of the unknown position node; the objective function is defined as a position function of all the unknown position nodes in the soil moisture sensor network; A gradient calculation module, used for obtaining the gradient of the objective function relative to the position of the unknown position node through iterative calculation; The Barzilai-Borwein step length calculation module is used to obtain the Barzilai-Borwein step length by setting the strategy calculation; A node position updating module, used for updating the position of the unknown position node according to the obtained Barzilai-Borwein step size; The node position acquisition module is used to set the stopping condition of the iterative calculation of the objective function. When the stopping condition is met, the iterative calculation of the objective function is stopped to obtain the position of the unknown position node.

Images