A self-powered detector array and its arrangement optimization method

Abstract

This invention discloses a self-powered detector array and its optimized arrangement method. It proposes an improved detector array structure and optimizes the array's arrangement, resulting in a self-powered detector array with higher neutron sensitivity and neutron flux measurement accuracy. The self-powered detectors used in this array are novel multi-layered detectors. The multi-layered design reduces the self-shielding effect of the detector emitter, thereby increasing detector sensitivity. The axial arrangement of the self-powered detector array is optimized based on a particle swarm optimization algorithm, improving measurement accuracy by reducing interference between detectors within the array. This self-powered detector array features high measurement accuracy and sensitivity due to the use of multi-layered self-powered detectors, strong method versatility, simple operation, wide applicability, and the ability to conform to the axial distribution of neutron flux at different locations within the reactor.

Description

Technical Field

[0001] This invention relates to the field of nuclear reactor neutron detection, specifically a self-powered detector array and its optimized arrangement method. Background Technology

[0002] The neutron flux distribution in a nuclear reactor directly affects the reactor core power distribution, serving as a crucial reflection of the core's operational status and a vital parameter for determining core operating limits and safety margins. Accurate measurement of the neutron flux distribution within the core can improve core operation safety, effectively reduce conservative margins in operating procedures, and ultimately enhance the economic efficiency of the nuclear power plant.

[0003] Currently, self-powered detectors are widely used in monitoring neutron flux in third-generation nuclear power plants such as AP1000 and Hualong One. Their ability to operate without an external power source and their simple, compact structure make them easy to integrate into the reactor core instrumentation tube. Their resistance to high temperatures, high pressures, and radiation makes them particularly suitable for in-reactor monitoring in reactor environments. Because self-powered detectors employ nuclear activation, the emitter material reacts with neutrons to emit beta or gamma rays, forming a detectable DC current. Therefore, self-powered detectors also have the advantages of requiring no bias voltage and being able to stably measure neutron flux online.

[0004] The main technical specifications of a self-sufficient detector are sensitivity, response time, and burnup lifetime. Sensitivity reflects the detector's ability to generate a current signal under a unit thermal neutron flux. Response time reflects the time it takes for the detector to generate a stable current after reacting with a neutron. Burnup lifetime reflects the estimated value that the detector can withstand under continuous irradiation by neutrons of a certain energy. After the burnup lifetime is exceeded, the detector's deviation will exceed the allowable range.

[0005] Currently, the mainstream detector arrays used in third-generation nuclear reactors are rhodium or vanadium self-powered detectors, and the mainstream array arrangement methods are axial equal-length arrangement or ring array unequal-length arrangement. The main challenges are low signal response current, neutron fluence measurement signals being affected by core gamma rays, and mutual interference between detectors. Further research is needed on how to effectively improve neutron sensitivity to increase response current, how to reduce the influence of gamma response signals, how to reduce interference between detectors within the array, and how to reduce emitter consumables. Summary of the Invention

[0006] To overcome the problems existing in the prior art and improve the accuracy of neutron flux rate measurement in the reactor core, this invention provides a self-powered detector array and its arrangement optimization method. It uses a high-sensitivity multilayer self-powered detector and provides an optimization method for the detector array arrangement based on the particle swarm optimization algorithm.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] A self-powered detector group consists of 7 self-powered detectors with a multi-layer structure. The distance from the top of the sensitive part of each detector to the bottom of the cable sleeve and the length of the sensitive part of each detector are different. The distance from the top of the sensitive part to the bottom of the cable sleeve is hereinafter referred to as the height, and the length of the sensitive part of each detector is hereinafter referred to as the length.

[0009] The height and length of each detector in the detector group are integer multiples of the unit detector length, and the unit detector length is the ratio of the total height of the test area in the core to the number of test points.

[0010] In the detector group, the detectors are arranged in a ring array in the order of number 1-7, with number 1 located at the center of the ring and numbers 2-7 evenly distributed on the ring array in sequence.

[0011] The optimized height and length of each detector in the detector group are as follows:

[0012] serial number 1 2 3 4 5 6 7 Height / unit length 5 1 0 2 4 3 0 Length / Unit Length 2 2 3 4 2 2 4

[0013] The detector group is placed inside the outer casing. The outer casing is of similar length to the instrument tube of the pressurized water reactor assembly and is arranged inside the instrument tube. The space between the detector group and the outer casing is filled with nitrogen. Each detector in the detector group is connected to the top of the casing by a cable.

[0014] The self-powered detector group consists of 7 self-powered detectors with a multi-layer structure. The main body of each detector is coaxial and has five layers from the inside out: inner collector 1, inner insulator 2, emitter 3, outer insulator 4, and outer collector 5.

[0015] The bottom ends of the inner collector 1, inner insulator 2, emitter 3, and outer collector 5 are closed, and the top ends of the inner collector 1, emitter 3, and outer collector 5 are connected to the cable.

[0016] The outer collecting body 5 is a hollow cylindrical tube with a round bottom;

[0017] The emitter 3 is separated from the inner collector 1 and the outer collector 5 by the inner insulator 2 and the outer insulator 4, respectively. The bottom of the outer collector 5 is filled with a filling material 6 to separate it from the outer insulator 4.

[0018] The number of test points, n, is 7.

[0019] The emitter 3 of the detector is made of high-purity rhodium.

[0020] The detector's inner insulator 2, outer insulator 4, and cable filling material are aluminum oxide or magnesium oxide.

[0021] The detector's collection body 3 is made of Inconel or stainless steel.

[0022] The method for optimizing the arrangement of a self-powered detector group includes the following steps:

[0023] Step 1: Obtain the reference solution vector of the core axial flux distribution at the location where the detector is to be deployed.

[0024] Step 2: Abstract the actual detector array arrangement as an n-dimensional position vector x p and an n-dimensional length vector x l For the n-dimensional position vector x of each detector in a detector group p and an n-dimensional length vector x l Perform sampling;

[0025] Step 3: The x values ​​of each detector within a detector group... p and x l Form a layout matrix X, and perform an observability test on each matrix X. If the matrix X is not invertible, it means that the layout matrix equation is not valid. If it is unsolvable, we will return to step 2 to resample. If the matrix X is invertible, it means that the matrix equation of the arrangement method is solvable, and then we will proceed to step 4.

[0026] Step 4: Model and simulate the detector group using the Monte Carlo method to obtain the total neutron flux density on the surface of each detector under the corresponding arrangement.

[0027] Step 5: Solve the matrix equation of the arrangement method Obtain the axial neutron flux density distribution at the measurement point location under the corresponding arrangement.

[0028] Step 6: According to Reference solution at the corresponding height Obtain the root mean square error of the measurement flux for the corresponding arrangement. Record the mean squared error during each iteration, with the minimum value denoted as the historical mean squared error σ. min ;

[0029] Step 7: Use formula (1) and formula (2) respectively to sample the velocity vector v of the position change. p and the velocity vector of length change v l And update the position vector in the next iteration. and length vector Returning to step 3, begin a new round of iterations. Note that after sampling, the position change velocity vector v needs to be processed. p and the velocity vector of length change v lThe sampling results are rounded; the specific processing method is shown in formula (1) and formula (2):

[0030]

[0031]

[0032] in These are the velocity vectors of position change and length change, respectively, in the k-th iteration. Let c1 and c2 be the position change velocity vector and length change velocity vector, respectively, for the (k-1)th iteration; ω, c1, and c2 be the inertia weight, self-learning factor, and population learning factor, respectively; and Rand be a random number in the range (0, 1). These represent the optimal position and optimal length of an individual's history during the iteration process, respectively. These represent the best historical position and the best historical length of the population during the iteration process, respectively. These are the position vector and length vector for the (k-1)th iteration, respectively, where k is the iteration algebra notation; after sampling, the position vector and length vector for the kth iteration are obtained. The boundary conditions are judged and corrected, and the observability of matrix X is determined.

[0033] Step 8: When the historical mean squared error σ min The loop terminates when the value is less than the set value or the number of iterations reaches a certain value. The optimized position and length of each detector in the detector group are obtained. The optimal detector position and length are used to uniquely determine the arrangement with the highest measurement accuracy.

[0034] Based on the principle of particle swarm optimization, the position and length parameters of each detector in the detector group are optimized. The actual detector group model is abstracted into a mathematical model using the arrangement matrix. The height parameter of each detector in the group is abstracted into a position vector, and the length parameter of each detector in the group is abstracted into a length vector. Each arrangement matrix is ​​uniquely determined by the position and length vectors corresponding to the detector group, and can be set as two populations for iterative optimization to obtain the optimal arrangement solution.

[0035] Compared with existing methods, the present invention has the following advantages:

[0036] 1. The main body of the novel self-powered detector used in this invention adopts a five-layer structure design. By increasing the volume of the emitter in the original design, the deposition probability of effective electrons by the emitter material is reduced, thereby reducing the self-shielding effect of the self-powered detector and increasing the electron escape probability. Therefore, it has higher sensitivity than the original design.

[0037] 2. Compared with the original design detector of the same size, the novel multilayer self-powered detector proposed in this invention not only improves sensitivity, but also significantly reduces material consumption of the transmitter part, and can achieve higher working performance through the optimization of the structural dimensions of each layer.

[0038] 3. The multi-layer detector structure design proposed in this invention improves neutron sensitivity and neutron partial response current while reducing photon sensitivity and photon partial response current through the self-compensation capability of the inner and outer dual collector electrodes. Furthermore, it uses the background current core wire to compensate for the core background noise signal, thereby achieving a higher n / γ signal ratio.

[0039] 4. The detector arrangement method of this invention can be applied to any test location in the reactor core. Using the axial power distribution at the test point obtained by measurement or simulation as a standard, the corresponding detector axial arrangement method can be obtained using this algorithm. Compared with the existing ring-shaped unequal length array arrangement, the arrangement obtained by the method of this invention has less interference between detectors and a larger output current. Attached Figure Description

[0040] Figure 1a This is a schematic diagram of a ring array arrangement of self-powered detectors. Figure 1b This is a schematic diagram showing the height and length distribution of a self-powered detector group. Figure 1c This is a schematic diagram of the main body of each detector.

[0041] Figure 2 This is a structural diagram of a multi-layer self-powered detector.

[0042] Figure 3 This is a schematic diagram of the matrix abstraction method for arrangement.

[0043] Figure 4 This is a flowchart for optimizing the detector array deployment method.

[0044] Figure 5 This is a comparison diagram of the detector group arrangement method before and after optimization. Detailed Implementation

[0045] The structure of the present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0046] This invention discloses a self-powered detector array, which consists of seven self-powered detectors with a multi-layered structure, arranged as follows: Figure 1a , Figure 1b and Figure 1c As shown, the distance from the top of the sensitive part of each detector to the bottom of the cable sleeve (hereinafter referred to as height) and the length of the sensitive part of each detector (hereinafter referred to as length) are different.

[0047] The height and length of each detector in the detector group are integer multiples of the unit detector length. The unit detector length is the ratio of the total height of the test area in the core to the number of test points. Generally, the number of test points n is 7.

[0048] In the detector group, the detectors are arranged in a ring array in the order of number 1-7, with number 1 located at the center of the ring and numbers 2-7 evenly distributed on the ring array in sequence.

[0049] The optimized height and length of each detector in the detector group are as follows:

[0050] serial number 1 2 3 4 5 6 7 Height / unit length 5 1 0 2 4 3 0 Length / Unit Length 2 2 3 4 2 2 4

[0051] The specific arrangement of the self-powered detector group is obtained through an arrangement optimization method, such as... Figure 4 As shown, the specific optimization steps are as follows:

[0052] Step 1: Obtain the reference solution vector of the core axial flux distribution at the location where the detector is to be deployed through actual measurement or numerical methods. The number of elements contained should be the same as the number of detectors n in the detector group. If a reference solution cannot be obtained by the above method, it can also be approximated by arranging n unit length detectors equiaxially.

[0053] Step 2: Through Figure 3 The method shown abstracts the actual detector array arrangement as an n-dimensional position vector x. p and an n-dimensional length vector x l For the n-dimensional position vector x of each detector in a detector group p and an n-dimensional length vector x l Sampling is performed, with each detector group corresponding to a pair of position and length vectors. Where x... p The ratio of the height of each detector to the length of a detector per unit length, x l The ratio of the length of each detector to the length of a detector per unit length, x p (i) takes values ​​in the range [0, n-1], x l (i) takes values ​​in the range [1, nx] p (i)], (i=1,2,3,...,n);

[0054] Step 3: From x p and x l Form an arrangement matrix X, and perform an observability test on matrix X, that is, determine whether the arrangement matrix is ​​invertible. If the matrix X is not invertible, it means that the arrangement matrix equation... If the solution is unsolvable, return to step 2 for resampling. If the matrix X is invertible, it means the matrix equation for the arrangement is solvable, then proceed to step 4. Here, X is an n-order square matrix composed of 1s and 0s, and each row represents the specific arrangement of a detector.

[0055] Step 4: Model and simulate the detector group using the Monte Carlo method to obtain the total neutron flux density on the surface of each detector under the corresponding arrangement. At this time, the result was obtained. This is the sum of the neutron flux density on each detector;

[0056] Step 5: Solve the matrix equation of the arrangement method Obtain the axial neutron flux density distribution at the measurement point location under the corresponding arrangement. At this time, the result was obtained. Neutron flux density at n axial measurement points corresponding to the axis;

[0057] Step 6: According to Reference solution at the corresponding height Obtain the root mean square error of the measurement flux for the corresponding arrangement. Record the mean squared error during each iteration, with the minimum value denoted as the historical mean squared error σ. min ;

[0058] Step 7: Use formula (1) and formula (2) respectively to sample the velocity vector v of the position change. p and the velocity vector of length change v l And update the position vector in the next iteration. and length vector Returning to step 3, begin a new round of iterations. Note that after sampling, the position change velocity vector v needs to be processed. p and the velocity vector of length change v l The sampling results are rounded. The specific processing method is shown in formulas (1) and (2):

[0059]

[0060]

[0061] in These are the velocity vectors of position change and length change, respectively, in the k-th iteration. Let c1 and c2 be the position change velocity vector and length change velocity vector, respectively, for the (k-1)th iteration; ω, c1, and c2 be the inertia weight, self-learning factor, and population learning factor, respectively; and Rand be a random number in the range (0, 1). These represent the optimal position and optimal length of an individual's history during the iteration process, respectively. These represent the best historical position and the best historical length of the population during the iteration process, respectively. These are the position vector and length vector for the (k-1)th iteration, respectively, where k is the iteration algebra notation; after sampling, the position vector and length vector for the kth iteration are obtained. The boundary conditions are judged and corrected, and the observability of matrix X is determined.

[0062] Step 8: When the historical mean squared error σ min The loop terminates when the value is less than the set value or the number of iterations reaches a certain value. The optimized position and length of each detector in the detector group are obtained. The optimal detector position and length can be used to uniquely determine the arrangement with the highest measurement accuracy.

[0063] The arrangement optimization method is characterized by: optimizing the position and length parameters of each detector in the detector group based on the principle of particle swarm optimization algorithm; using the arrangement matrix to abstract the actual detector group model into a mathematical model; abstracting the height parameter of each detector in the group into a position vector; abstracting the length parameter of each detector in the group into a length vector; each arrangement matrix can be uniquely determined by the position and length vectors corresponding to the detector group; and can be set as two populations for iterative optimization to obtain the optimal arrangement solution.

[0064] The self-powered detector group consists of seven self-powered detectors with a multi-layered structure, such as... Figure 2 As shown, each detector body has a coaxial structure with five layers from the inside out: inner collector 1, inner insulator 2, emitter 3, outer insulator 4, and outer collector 5.

[0065] The bottom ends of the inner collector 1, inner insulator 2, emitter 3, and outer collector 5 are closed, and the top ends of the inner collector 1, emitter 3, and outer collector 5 are connected to the cable.

[0066] The external collector is a hollow cylindrical tube with a round bottom.

[0067] The emitter 3 is separated from the inner collector 1 and the outer collector 5 by the inner insulator 2 and the outer insulator 4 respectively. The bottom of the outer collector 5 is filled with a filling material 6 to separate it from the outer insulator 4.

[0068] Preferably, the material of the detector's collector 3 is high-purity rhodium or high-purity vanadium.

[0069] Preferably, the inner insulator 2, outer insulator 4, and cable insulation layer 8 of the detector are made of aluminum oxide or magnesium oxide.

[0070] Preferably, the material of the detector's collection body 3 is Inconel.

[0071] The principle behind the arrangement optimization method used in this invention is as follows: Since self-powered detectors require nuclear reactions with neutrons for measurement, multiple detectors arranged in the same instrument tube will be interfered with by nearby detectors, preventing the acquisition of accurate neutron flux data. Furthermore, because the detector output current is positively correlated with the detector volume, a longer detector results in a larger output current, given a limited overall outer diameter. This optimization method, based on the particle swarm optimization algorithm, uses the root mean square error between the actual and theoretical values ​​of the self-powered detector as the objective function. It can obtain an arrangement matrix with the smallest possible error while maintaining a minimum detector length.

[0072] The arrangement of detectors within different detector groups can uniquely determine an arrangement matrix. This is achieved by abstracting the height parameters of each detector in an actual detector group as an n-dimensional position vector x. p The length parameters of each detector in the swarm are abstracted as an n-dimensional length vector x. l Let the detectors in the detector group be numbered i (i = 1, 2, 3, ..., n), then the i-th row of the arrangement matrix can be obtained from x p With x l The i-th component determines the arrangement. By abstracting the actual detector swarm model into a mathematical model, the arrangement matrix can be set as two populations for iterative optimization to obtain the optimal arrangement solution, such as... Figure 4 As shown.

[0073] A comparison of the neutron flux density measured by the detector group with the original unequal length arrangement of the ring array and the optimized arrangement is shown below. Figure 5 As shown, the optimized measurements are clearly more accurate, with a maximum relative error of 7%.

[0074] The novel self-powered detector with a multi-layer structure used in this invention operates on the following principle: When the multi-layer detector is irradiated by neutrons from the reactor core, the neutrons pass through the outer collector and outer insulator and undergo a nuclear reaction with the emitter. The atomic nuclei of the emitter material capture the neutrons and produce electrons. Due to the obstruction formed by the inner and outer insulators, the electrons produced in the emitter move towards the inner and outer collectors. During this process, some electrons are deposited in the insulator due to the obstruction of the insulator, forming an electrostatic field that further prevents the reverse flow of electrons.

[0075] Due to the self-shielding effect, the thicker the emitter, the greater the initial energy required for electrons to leave the emitter. The initial energy of electrons is given by the probability distribution of the β decay energy spectrum B(E) of the emitter material. Therefore, by reducing the thickness of the emitter, the distance required for electrons to move in the emitter is reduced, thereby increasing the probability of electrons leaving the emitter, passing through the insulator, and reaching the collector, ultimately achieving the effect of increasing sensitivity and output current.

[0076] The probability of an electron leaving the emitter can be calculated using the following formula:

[0077]

[0078]

[0079] In the formula:

[0080] E – Energy of an electron when it leaves the emitter

[0081] E represents the energy of the electron when it leaves the emitter. ′ Let E represent the energy an electron possesses upon decay, R(E) represent the distance an electron with energy E can travel, N(l)dl represent the probability that the electron travels a distance within the interval dl near l, and B(E)dE represent the probability that the electron's energy falls within the interval dE near E upon decay. The inner and outer integration limits in the integral terms correspond one-to-one; changing the lower integration limit allows us to obtain the escape probability of electrons in emitters of different thicknesses.

[0082] On the other hand, the structure of adding an inner collector inside the emitter can also effectively compensate for part of the response current generated by external photons, reduce external photon noise interference, and thus achieve the goal of improving the n / γ signal ratio.

[0083] Considering the material consumption of the emitter during long-term operation, since the burnup of the emitter is mainly related to the neutron reaction cross section of its material, the structure of reducing the thickness of the emitter in this invention will not reduce the service life of the self-powered detector. However, after the sensitivity of the self-powered detector drops to below 60% of the initial value due to material burnup, a new self-powered detector still needs to be replaced.

Claims

1. A self-sustained detector swarm, characterized by: The self-powered detector group is composed of 7 self-powered detectors with multi-layer structure, the distance from the top of the sensitive part of each detector to the bottom of the cable sleeve and the length of the sensitive part of each detector are different, and the distance from the top of the sensitive part to the bottom of the cable sleeve is referred to as height, and the length of the sensitive part is referred to as length; The height and length of each detector in the detector group are integer multiples of the unit detector length, and the unit detector length is the ratio of the total height of the measured area in the core to the number of measured points; The detectors in the detector group are arranged in a ring array in the order of 1-7, with No. 1 at the center of the ring, and Nos. 2-7 evenly distributed on the ring array; The optimized height and length of each detector in the detector group are as follows: 。 2. A self-sustaining detector group according to claim 1, characterized in that: The detector group is placed in an outer sleeve, the outer sleeve has a length equivalent to the instrument tube of the pressurized water reactor assembly and is arranged in the instrument tube, the detector group and the outer sleeve are filled with nitrogen, and each detector in the detector group is connected to the top of the sleeve by a cable; Each detector in the self-powered detector group is composed of 7 self-powered detectors with multi-layer structure, the main part of each detector is coaxial structure, and there are five layers from inside to outside, which are inner collector (1), inner insulator (2), emitter (3), outer insulator (4) and outer collector (5); The bottom of the inner collector (1), the inner insulator (2), the emitter (3) and the outer collector (5) is closed, and the top of the inner collector (1), the emitter (3) and the outer collector (5) is connected with the cable; The outer collector (5) is a hollow cylindrical bottom pipe; The emitter (3) is separated from the inner collector (1) and the outer collector (5) by the inner insulator (2) and the outer insulator (4) respectively, and the bottom of the outer collector (5) is provided with a filling material (6) for separating from the outer insulator (4).

3. A self-sustaining detector group according to claim 1, characterized in that: The number of measured points n is 7.

4. A self-sustaining detector group according to claim 1, characterized in that: The material of the emitter (3) of the detector is high-purity rhodium.

5. A self-sustaining detector group according to claim 1, wherein: The inner insulator (2), the outer insulator (4) and the cable filling material of the detector are alumina or magnesium oxide.

6. A self-sustaining detector group according to claim 1, characterized in that: The collector (3) of the detector is inco nickel or stainless steel.

7. A method of optimizing the arrangement of a self-sufficient detector group according to any one of claims 1 to 6, characterized in that: The method comprises the following steps: Step 1 : Obtain a reference solution vector of the axial flux distribution of the core at the location where the detector is to be disposed Step 2: Abstractly represent the actual detector group arrangement as an n-dimensional position vector x p and an n-dimensional length vector x l , for each detector in a detector group p and an n-dimensional length vector x l , sample Step 3: Form a matrix X of the arrangement pattern from the x p and x l of each probe in a group of probes. Determine the observability of each matrix X. If the matrix X is not invertible, it means that the arrangement matrix equation is not solvable, and go back to Step 2 to resample. If the matrix X is invertible, it means that the arrangement matrix equation is solvable, and go to Step 4. Step 4: Solve the arrangement matrix equation to obtain the arrangement pattern. Step 4: Modeling and simulation of the detector group using the Monte Carlo method to obtain the total neutron flux density on the surface of each detector under the corresponding arrangement Step 5: Disassembling the matrix equation Obtaining the axial neutron flux density distribution at the measured positions under the corresponding arrangement Step 6: According to the reference solution at the corresponding height the measurement flux error mean square variance of the corresponding arrangement is obtained The mean square variance in each iteration process is recorded, and the minimum value is recorded as the historical mean square variance σ min ; Step 7: Sample the position change velocity vector v p and the length change velocity vector v l and update the position vector and the length vector in the next iteration, go back to Step 3 and start a new iteration, note that the sampling results of the position change velocity vector v p and the length change velocity vector v l need to be rounded; see Formula (1) and Formula (2) for specific processing methods: wherein are the position change velocity vector and length change velocity vector of the kth iteration, respectively, are the position change velocity vector and length change velocity vector of the (k-1)th iteration, respectively, ω, c1, c2 are the inertia weight, self-learning factor and population learning factor, respectively, and Rand is a random number in the range of (0, 1), are the individual historical best position and individual historical best length in the iteration process, respectively, are the population historical best position and population historical best length in the iteration process, respectively, are the position vector and length vector of the (k-1)th iteration, respectively, and k is the iteration index; after sampling, the position vector and length vector of the kth iteration are obtained judging and correcting the boundary conditions and judging the observability of the matrix X; Step 8: When the history mean square error σ min The cycle is terminated when the history mean square error σ is less than a set value or the iteration number reaches a certain value, and the optimized results of the positions and lengths of the detectors in the detector group are obtained. The optimized detector positions and lengths uniquely determine a measurement arrangement with the highest measurement accuracy.

8. A method of arranging optimization of a self-sufficient detector group according to claim 7, characterized in that: Based on the principle of particle swarm optimization algorithm, the position and length parameters of each detector in the detector group are optimized, the actual detector group model is abstracted into a mathematical model using an arrangement mode matrix, the height parameters of each detector in the group are abstracted into a position vector, the length parameters of each detector in the group are abstracted into a length vector, each arrangement mode matrix is uniquely determined by the corresponding position and length vectors of the detector group, and can be set to two populations for iterative optimization to obtain the best arrangement mode solution.

Images