A method, medium and system for evaluating interference of communication equipment of a photovoltaic power station

Abstract

The application provides a photovoltaic power station communication equipment interference evaluation method, medium and system, and belongs to the technical field of communication equipment interference. The method collects communication equipment operation data and environmental electromagnetic data in a photovoltaic power station to establish a communication equipment interference basic database. Time domain analysis is performed on the data, and the data is converted into frequency domain characteristic parameters. An interference evaluation index system is established, and a neural network model is constructed. Communication equipment operation data to be evaluated are collected, and characteristic parameters are extracted to input the model. An interference evaluation result is obtained, and an anti-interference level is determined. The time domain characteristic parameters of the interference signal are decomposed into stable components and variable components. An equation set is established through an interference source influence coefficient matrix to calculate the interference source influence coefficient matrix and the communication equipment anti-interference coefficient matrix. Finally, the evaluation result is corrected according to the anti-interference coefficient matrix to obtain a corrected interference evaluation value, and a warning threshold is established for real-time monitoring and warning.

Description

Technical Field

[0001] This invention belongs to the field of communication equipment interference technology, and more specifically, relates to a method, medium and system for assessing interference with communication equipment in photovoltaic power plants. Background Technology

[0002] With the rapid development of photovoltaic power generation technology, photovoltaic power plants have become an important component of clean energy development. During the operation of a photovoltaic power plant, the communication system undertakes key functions such as data transmission, equipment monitoring, and operation scheduling. However, communication equipment in a photovoltaic power plant environment often faces complex electromagnetic interference problems. This interference comes not only from the electromagnetic radiation generated by the photovoltaic modules but also from the combined effects of surrounding environmental factors.

[0003] Currently, interference assessment for communication equipment in photovoltaic power plants mainly employs the following methods: First, on-site measurement-based assessment, which involves deploying measuring equipment on-site to collect and analyze electromagnetic environment data; second, simulation model-based assessment, which predicts interference by establishing electromagnetic field simulation models; and third, empirical statistical assessment, which determines interference patterns through statistical analysis of historical data. These traditional methods have significant limitations in practical applications: on-site measurement methods require substantial manpower and resources, and the measurement process is greatly affected by external conditions; simulation model methods are often overly idealized and fail to accurately reflect the complex interference conditions in actual operating environments; and empirical statistical methods lack theoretical support, resulting in low reliability and accuracy of assessment results.

[0004] Especially in large-scale photovoltaic power plants, due to the wide distribution of equipment and the complex and variable environmental conditions, traditional interference assessment methods are difficult to meet actual needs. Existing technologies cannot effectively conduct systematic correlation analysis of the operating status of communication equipment, environmental factors, and interference characteristics, resulting in significant deviations between interference assessment results and actual conditions. Summary of the Invention

[0005] In view of this, the present invention provides a method, medium and system for evaluating interference of communication equipment in photovoltaic power plants, which can solve the technical problem that the existing technology cannot effectively conduct systematic correlation analysis of the operating status, environmental factors and interference characteristics of communication equipment.

[0006] This invention is implemented as follows:

[0007] The first aspect of the present invention provides a method for assessing interference in communication equipment of a photovoltaic power plant, comprising the following steps:

[0008] S10. Collect the operation data of communication equipment and environmental electromagnetic data in the photovoltaic power station. The operation data of communication equipment includes the running time of communication equipment, the maintenance time of communication equipment, and the fault record of communication equipment. The environmental electromagnetic data includes the electromagnetic radiation intensity of photovoltaic modules, the ambient air humidity value, and the ambient temperature value. Establish a basic database of communication equipment interference.

[0009] S20. Perform time-domain analysis on the data in the communication equipment interference database, extract the time-domain feature parameters of the interference signal, and perform frequency-domain transformation on the time-domain feature parameters of the interference signal to obtain the frequency-domain feature parameters of the interference signal.

[0010] S30. Establish an interference evaluation index system based on the time-domain characteristic parameters and the frequency-domain characteristic parameters of the interference signal, construct a neural network model based on the interference evaluation index system, and train the interference evaluation model.

[0011] S40. Collect the operating data of the communication device to be evaluated, and extract the time-domain characteristic parameters and frequency-domain characteristic parameters of the interference signal of the communication device to be evaluated.

[0012] S50. Input the time-domain characteristic parameters and frequency-domain characteristic parameters of the interference signal of the communication device to be evaluated into the interference evaluation model to obtain the interference evaluation result.

[0013] S60. Determine the anti-interference level of the communication equipment based on the interference assessment results;

[0014] S70. Decompose the time-domain characteristic parameters of the interference signal into stable and variable components to obtain the stable components and variable components of the interference signal.

[0015] S80. Based on the stable component of the interference signal and the variable component of the interference signal, establish a set of equations through the interference source influence coefficient matrix to calculate the interference source influence coefficient matrix, and calculate the anti-interference coefficient matrix of the communication equipment based on the interference source influence coefficient matrix.

[0016] S90. Based on the anti-interference coefficient matrix of the communication equipment, the interference assessment result is corrected to obtain a corrected interference assessment value. The corrected interference assessment value is the original interference assessment value multiplied by the anti-interference coefficient matrix of the communication equipment. Based on the corrected interference assessment value, an interference warning threshold for the communication equipment is established for real-time monitoring and warning.

[0017] Based on the above technical solution, the method for evaluating interference in photovoltaic power plant communication equipment of the present invention can be further improved as follows:

[0018] Specifically, step S10 includes:

[0019] Step 101: Collect runtime data, maintenance time data, and fault record data for each communication device in the photovoltaic power station. Arrange the collected runtime data, maintenance time data, and fault record data in such a way that each row corresponds to one communication device to form a device operation data matrix.

[0020] Step 102: Collect electromagnetic radiation intensity data, ambient air humidity data, and ambient temperature data of photovoltaic modules around each communication device in the photovoltaic power station. Arrange the collected electromagnetic radiation intensity data, ambient air humidity data, and ambient temperature data of photovoltaic modules in such a way that each row corresponds to one communication device to form an environmental electromagnetic data matrix.

[0021] Step 103: Store the device operation data matrix and the environmental electromagnetic data matrix into the communication device interference basic database respectively;

[0022] Step S20 specifically includes:

[0023] Step 201: Use a high-speed data acquisition card to sample the data in the basic database of communication equipment interference, set the sampling frequency to 2 kHz and the sampling duration to 48 hours.

[0024] Step 202: Extract interference signal amplitude data, interference duration data, interference period data, and signal strength data from the sampled data. Perform weighted calculations on the extracted interference signal amplitude data, interference duration data, interference period data, and signal strength data to obtain time-domain feature parameters.

[0025] Step 203: Perform Fast Fourier Transform on the time-domain feature parameters, setting a Hanning window with an overlap rate of 50%;

[0026] Step 204: Extract frequency point amplitude data, frequency point phase data, and frequency point phase difference data from the Fast Fourier Transform results, and arrange the extracted frequency point amplitude data, frequency point phase data, and frequency point phase difference data into a frequency domain feature parameter matrix.

[0027] Furthermore, step S30 specifically includes:

[0028] Step 301: Construct a neural network input layer using the time-domain feature parameters and frequency-domain feature parameters of the interference signal as input parameters;

[0029] Step 302: Construct the hidden layer of the neural network. Multiple neurons are set in the hidden layer of the neural network, and each neuron uses a modified linear unit function as the activation function.

[0030] Step 303: Construct the output layer of the neural network, and set up multiple neurons in the output layer of the neural network;

[0031] Step 304: Establish the weight matrix from the input layer to the hidden layer. The weight matrix is ​​used to represent the connection relationship from each neuron in the input layer to each neuron in the hidden layer.

[0032] Step 305: Establish the weight matrix from the hidden layer to the output layer. The weight matrix is ​​used to characterize the connection relationship from each neuron in the hidden layer to each neuron in the output layer.

[0033] Step 306: Train the neural network using the adaptive moment estimation optimization algorithm, setting the learning rate to 0.001 and the training rounds to 500 rounds to obtain the interference evaluation model.

[0034] Furthermore, step S40 specifically includes:

[0035] Step 401: Collect runtime data, maintenance time data, and fault record data of the communication equipment to be evaluated;

[0036] Step 402: Collect data on electromagnetic radiation intensity of photovoltaic modules, ambient air humidity, and ambient temperature around the communication equipment to be evaluated;

[0037] Step 403: Extract the time-domain and frequency-domain characteristic parameters of the interference signal from the collected data.

[0038] Furthermore, step S50 specifically includes:

[0039] Step 501: Arrange the time-domain feature parameters of the interference signal of the communication device to be evaluated into a time-domain feature parameter matrix in such a way that each row corresponds to a time point;

[0040] Step 502: Arrange the frequency domain characteristic parameters of the interference signal of the communication device to be evaluated into a frequency domain characteristic parameter matrix in such a way that each row corresponds to one frequency point;

[0041] Step 503: Input the time-domain feature parameter matrix and the frequency-domain feature parameter matrix into the interference evaluation model to obtain the interference evaluation result;

[0042] Step S60 specifically includes:

[0043] Step 601: Set the anti-interference level evaluation criteria, divide the anti-interference level into multiple levels, and set the corresponding interference evaluation result range for each level;

[0044] Step 602: Compare the interference assessment results with the anti-interference level evaluation criteria;

[0045] Step 603: Determine the anti-interference level of the communication device based on the comparison results, and use the anti-interference level as an evaluation index of the anti-interference capability of the communication device.

[0046] Furthermore, step S70 specifically includes:

[0047] Step 701: Perform integral calculation on the time-domain characteristic parameters of the interference signal according to the observation period to obtain the stable component of the interference signal;

[0048] Step 702: Subtract the stable component of the interference signal from the time-domain characteristic parameters of the interference signal to obtain the variable component of the interference signal.

[0049] Furthermore, step S80 specifically includes:

[0050] Step 801: Calculate the degree of environmental impact on the communication equipment based on the electromagnetic radiation intensity data of the photovoltaic module, the ambient air humidity data, and the ambient temperature data.

[0051] Step 802: Calculate the anti-interference capability of the communication equipment under its current operating state based on the equipment runtime data, the maintenance time data, and the number of faults data;

[0052] Step 803: Calculate the impact intensity of the interference on the communication equipment based on the interference occurrence count data, the interference duration data, and the communication interruption duration data;

[0053] Step 804: Combine the environmental impact level, the anti-interference capability, and the impact intensity in a block diagonalization manner to form an interference source impact coefficient matrix;

[0054] Step 805: Calculate the inverse matrix of the interference source influence coefficient matrix, and multiply the inverse matrix by the normalization coefficient matrix and the correction coefficient matrix to obtain the anti-interference coefficient matrix of the communication equipment.

[0055] Furthermore, step S90 specifically includes:

[0056] Step 901: Multiply the interference assessment result by the anti-interference coefficient matrix of the communication device and the diagonal matrix of the assessment weight to obtain the corrected interference assessment value;

[0057] Step 902: Calculate the standard deviation of the corrected interference assessment value, and establish an early warning threshold based on the corrected interference assessment value and the standard deviation;

[0058] Step 903: Monitor the communication equipment in real time, and issue a warning signal when the real-time monitoring value exceeds the warning threshold.

[0059] The equation set for establishing the interference source influence coefficient matrix includes environmental impact equations, equipment state equations, interference characteristic equations, and comprehensive evaluation equations.

[0060] The environmental impact equation is used to assess the degree of impact of the surrounding environment on the communication equipment of the photovoltaic power station. The inputs include the electromagnetic radiation intensity of the photovoltaic module, the current air humidity value, and the current temperature value. The output is the interference impact coefficient of the environment on the communication equipment.

[0061] The device state equation is used to evaluate the impact of the communication device's own operating state on its anti-interference capability. The inputs include the communication device's running time, the communication device's most recent maintenance time, and the communication device's cumulative number of failures. The output is the anti-interference capability coefficient of the device in its current state.

[0062] The interference characteristic equation is used to evaluate the impact characteristics of interference signals on communication equipment. The inputs include the number of times the interference signal occurs within a continuous monitoring period, the duration of each interference signal, and the duration of communication interruption caused by each interference. The output is the impact intensity coefficient of the type of interference.

[0063] The comprehensive evaluation equation is used to perform weighted calculations on the above three coefficients to obtain the final result. The inputs include the environmental impact degree coefficient, the equipment anti-interference capability coefficient, and the interference impact intensity coefficient. The output is the final interference source impact coefficient matrix.

[0064] The calculations or equations involved in this invention are described in detail below:

[0065] 1. Disrupt the establishment of the basic database (step S10):

[0066] The equipment operation data matrix D is specifically represented as follows:

[0067]

[0068] In the formula, t i1 m is the runtime of the i-th device; i1 f is the maintenance time for the i-th device; i1 Let be the number of failures of the i-th device; n is the number of devices.

[0069] The environmental electromagnetic data matrix E is specifically represented as follows:

[0070]

[0071] In the formula, r i1 h represents the electromagnetic radiation intensity at the i-th location. i1 Let t be the air humidity at the i-th location; i1 Let be the temperature value at the i-th location.

[0072] 2. Time-domain analysis process (step S20):

[0073] Time-domain feature parameter vector F t Specifically, it is expressed as follows:

[0074] F t ={f t1 ,f t2 ,...,f tn};

[0075] f ti =α i ·A i +β i ·T i +γ i ·P i +δ i ·S i +ε t ;

[0076] In the formula, f ti Let A be the i-th time-domain feature parameter; i T represents the amplitude of the interference signal. i P represents the duration of the interference. i The interference period; S i α represents signal strength. i ,β i ,γ i ,δ i ε is the weighting coefficient; t This is the time-domain error term.

[0077] 3. Frequency domain conversion process:

[0078] Frequency domain characteristic parameter matrix F f Specifically, it is expressed as follows:

[0079]

[0080] In the formula, a i1 p is the amplitude at the i-th frequency point; i1 φ represents the phase at the i-th frequency point. i1 is the phase difference at the i-th frequency point; m is the number of frequency points.

[0081] Frequency domain conversion calculation process:

[0082]

[0083] In the formula, f fi f is the i-th frequency domain feature parameter; tk For time-domain sequences; N is the number of sampling points; ε f This is the frequency domain error term.

[0084] 4. Interference signal decomposition process (step S70):

[0085] Stable component extraction:

[0086]

[0087] In the formula, S c For stable components; s(t) is the original signal; T is the observation period; ε s This is the error term for the stable component.

[0088] Extraction of variable components:

[0089] D c =s(t)-S c +ε d ;

[0090] In the formula, D c For the variable component; ε d This is the error term for the variable component.

[0091] 5. Neural Network Model (Step S30):

[0092] The input layer to hidden layer weight matrix W is specifically represented as follows:

[0093]

[0094] In the formula, w ij is the connection weight between the i-th hidden layer neuron and the j-th input neuron; h is the number of hidden layer neurons; n is the number of input layer neurons.

[0095] The weight matrix V from the hidden layer to the output layer is specifically represented as follows:

[0096]

[0097] In the formula, v ij is the connection weight between the i-th output neuron and the j-th hidden layer neuron; o is the number of output layer neurons.

[0098] Neural network computation process:

[0099]

[0100] In the formula, H l The output of the l-th hidden layer neuron; σ is the activation function, using the ReLU function; x i For input features; b l For hidden layer bias; O k The output of the k-th output neuron; c k For output layer bias; ε n This is the network error term.

[0101] 6. Environmental Impact Equation (Step S80):

[0102]

[0103] In the formula, E is the environmental impact coefficient; R, H, and T are the electromagnetic radiation intensity, air humidity, and ambient temperature, respectively; R0, H0, and T0 are the standard reference values; k1, k2, and k3 are the weighting coefficients; α, β, and γ are the sensitivity indices; λ1, λ2, and λ3 are the attenuation coefficients; and ε... e This is the environmental error term.

[0104] 7. Equipment state equations:

[0105]

[0106] In the formula, S is the equipment state coefficient; t is the running time; τ is the time since the last maintenance; f is the number of failures; λ1, λ2, λ3 are weighting coefficients; μ1, μ2, μ3 are performance degradation coefficients; v1, v2, v3 are stability coefficients; ε s This is the state error term.

[0107] 8. Disturbance characteristic equation:

[0108]

[0109] In the formula, I is the interference characteristic coefficient; N is the number of interference occurrences; D is the interference duration; B is the communication interruption duration; N0, D0, B0 are reference thresholds; ω1, ω2, ω3 are weighting coefficients; δ, θ, φ are influence indices; η1, η2, η3 are suppression coefficients; ε i This is the characteristic error term.

[0110] 9. Calculation of the interference source influence coefficient matrix:

[0111]

[0112] In the formula, M is the interference source influence coefficient matrix; ρ i ,σ i ,τ i ε is the overall weighting coefficient. m This is the comprehensive error term.

[0113] 10. Calculation of anti-interference coefficient matrix:

[0114] R=M -1 ·K·C+ε r ;

[0115]

[0116] In the formula, R is the anti-interference coefficient matrix; K is the normalization coefficient matrix; C is the correction coefficient matrix; c i For each dimension, there are correction coefficients; ε r This is the anti-interference error term.

[0117] 11. Calculation of the corrected interference assessment value:

[0118] V modified =V original ·R·W+ε v ;

[0119] W = diag(w1, w2, w3);

[0120] In the formula, V modified This is the corrected evaluation value; V original The original evaluation value is represented by W; the evaluation weight diagonal matrix is ​​represented by w. i Assess the weights for each dimension; ε v This is to correct the error term.

[0121] 12. Establishment of early warning thresholds:

[0122] T warning =α·V modified +β·σ V +γ;

[0123] In the formula, T warning σ is the warning threshold; v α represents the standard deviation of the evaluation value; β represents the weighting coefficients; and γ represents the benchmark threshold.

[0124] Parameter acquisition method:

[0125] 1. Time domain parameters: Sampling is performed using a high-speed data acquisition card at a sampling frequency of 2kHz and a sampling duration of 48 hours;

[0126] 2. Frequency domain parameters: FFT transformation is used, Hanning window is selected as the window function, and the overlap rate is 50%;

[0127] 3. Weighting coefficients: determined using a combination of expert scoring and fuzzy hierarchical analysis.

[0128] 4. Error term: determined through Monte Carlo simulation and statistical analysis of historical data, and controlled within ±3%;

[0129] 5. Neural network parameters: Adam optimizer was used, learning rate 0.001, training epochs 500;

[0130] 6. Correction coefficients: obtained by fitting experimental data and optimized using the least squares method.

[0131] Explanation of the principle behind equation construction:

[0132] 1. The time-domain characteristic equation adopts a multivariate linear weighted model, which takes into account the basic characteristics of the interference signal;

[0133] 2. Frequency domain transformation uses the FFT algorithm to comprehensively describe signal characteristics through parameters such as amplitude and phase;

[0134] 3. The environmental impact equation adopts a power-law composite function to reflect the nonlinear impact and saturation characteristics of environmental factors;

[0135] 4. The equipment state equation adopts a double exponential model to reflect the degradation and stabilization process of equipment performance;

[0136] 5. The interference characteristic equation adopts a power-law combination form to describe the cumulative effect and suppression effect of interference;

[0137] 6. Matrix calculations employ block diagonalization to reduce computational load and improve numerical stability.

[0138] A second aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores program instructions, which, when executed in a computer, are used to perform the above-described method for evaluating interference in communication equipment of a photovoltaic power station.

[0139] A third aspect of the present invention provides a photovoltaic power plant communication equipment interference assessment system, wherein the system includes the aforementioned computer-readable storage medium.

[0140] Compared with existing technologies, the present invention provides a method, medium, and system for assessing interference in photovoltaic power plant communication equipment, which has several significant advantages: First, the present invention employs a multi-dimensional data acquisition strategy to comprehensively collect communication equipment operation data and environmental electromagnetic data, establishing a complete interference database. Through time-domain and frequency-domain dual feature extraction, it deeply explores the essential characteristics of interference signals, providing richer and more accurate feature inputs for subsequent assessment models, significantly improving the accuracy of interference assessment. Second, the present invention innovatively introduces a neural network model, establishing an adaptive interference assessment model through learning and training on a large amount of historical data. This model can automatically identify and learn interference characteristics under different operating conditions, effectively overcoming the limitations of human experience judgment in traditional assessment methods, making the assessment results more objective and reliable. Third, the present invention introduces stable and variable component decomposition techniques for interference signals, combined with the interference source influence coefficient matrix, to achieve accurate identification and quantitative analysis of interference sources. This method can not only accurately assess the current interference situation but also predict potential interference risks, solving the technical problem that existing technologies cannot effectively and systematically correlate the operating status of communication equipment, environmental factors, and interference characteristics, providing a scientific basis for preventive maintenance of communication equipment. Attached Figure Description

[0141] Figure 1 A flowchart of the method provided by the present invention. Detailed Implementation

[0142] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0143] like Figure 1 The diagram shown is a flowchart of an interference assessment method for communication equipment in a photovoltaic power station provided by the present invention. This method includes the following steps:

[0144] S10. Collect operating data of communication equipment and environmental electromagnetic data in the photovoltaic power station. The operating data of communication equipment includes the operating time of communication equipment, the maintenance time of communication equipment, and the fault record of communication equipment. The environmental electromagnetic data includes the electromagnetic radiation intensity of photovoltaic modules, the ambient air humidity value, and the ambient temperature value. Establish a basic database of communication equipment interference.

[0145] S20. Perform time-domain analysis on the data in the basic database of communication equipment interference, extract the time-domain characteristic parameters of the interference signal, and perform frequency-domain transformation on the time-domain characteristic parameters of the interference signal to obtain the frequency-domain characteristic parameters of the interference signal.

[0146] S30. Establish an interference assessment index system based on the time-domain and frequency-domain characteristic parameters of the interference signal, construct a neural network model based on the interference assessment index system, and train the interference assessment model.

[0147] S40. Collect the operating data of the communication equipment to be evaluated, and extract the time-domain characteristic parameters and frequency-domain characteristic parameters of the interference signal of the communication equipment to be evaluated.

[0148] S50. Input the time-domain characteristic parameters and frequency-domain characteristic parameters of the interference signal of the communication device to be evaluated into the interference evaluation model to obtain the interference evaluation results.

[0149] S60. Determine the anti-interference level of communication equipment based on the interference assessment results;

[0150] S70. Decompose the time-domain characteristic parameters of the interference signal into stable and variable components to obtain the stable and variable components of the interference signal.

[0151] S80. Based on the stable component and the variable component of the interference signal, establish a set of equations through the interference source influence coefficient matrix to calculate the interference source influence coefficient matrix, and calculate the anti-interference coefficient matrix of the communication equipment based on the interference source influence coefficient matrix.

[0152] S90. Based on the anti-interference coefficient matrix of the communication equipment, the interference assessment result is corrected to obtain the corrected interference assessment value. The corrected interference assessment value is the original interference assessment value multiplied by the anti-interference coefficient matrix of the communication equipment. Based on the corrected interference assessment value, an interference warning threshold for the communication equipment is established for real-time monitoring and warning.

[0153] The specific implementation methods of the above steps are described in detail below:

[0154] The specific implementation of step S10 is to establish a basic database for communication equipment interference, collecting and storing operational data and environmental electromagnetic data of communication equipment within the photovoltaic power station. First, the runtime, maintenance time, and fault record data of all communication equipment are collected. The collected data is preprocessed and standardized, and arranged so that each row corresponds to one communication device, constructing a device operation data matrix. Each row in the matrix contains data on the device's runtime, maintenance time, and fault records. Next, the electromagnetic radiation intensity data, ambient air humidity data, and ambient temperature data of the photovoltaic modules surrounding each communication device within the photovoltaic power station are collected. The collected data is preprocessed and standardized, and arranged so that each row corresponds to one communication device, constructing an environmental electromagnetic data matrix. Each row in the matrix contains data on the electromagnetic radiation intensity, air humidity, and temperature at that location. Finally, the device operation data matrix and the environmental electromagnetic data matrix are stored separately in the basic database for communication equipment interference. The database is managed using a relational database management system to ensure data integrity and consistency. The standard reference value for electromagnetic radiation intensity is 100 microwatts per square centimeter, the standard reference value for air humidity is 60%, and the standard reference value for temperature is 25 degrees Celsius.

[0155] The specific implementation of step S20 involves performing time-domain analysis and frequency-domain transformation on the data in the communication equipment interference database to extract the characteristic parameters of the interference signal. A high-speed data acquisition card is used to sample the data, with the sampling frequency set to 2 kHz to meet the requirements of the Nyquist sampling theorem, and the sampling duration set to 48 hours to ensure sufficient data samples are obtained. Interference signal amplitude data, interference duration data, interference period data, and signal strength data are extracted from the sampled data. A multivariate linear weighted model is used to weight the extracted data to obtain time-domain characteristic parameters. The amplitude weight coefficient is 0.3, the duration weight coefficient is 0.25, the period weight coefficient is 0.25, and the signal strength weight coefficient is 0.2. The time-domain characteristic parameters are processed by Fast Fourier Transform (FFT), and a Hanning window function is used for windowing. The window function overlap rate is set to 50% to reduce spectral leakage. Frequency point amplitude data, frequency point phase data, and frequency point phase difference data are extracted from the FFT results, and the extracted data are arranged into a frequency-domain characteristic parameter matrix.

[0156] The specific implementation of step S30 involves establishing an interference evaluation index system and constructing a neural network model. The time-domain and frequency-domain feature parameters of the interference signal are used as input parameters to construct the input layer of the neural network. The number of neurons in the input layer is determined based on the dimension of the feature parameters. A hidden layer is constructed, containing multiple neurons, each using a modified linear unit function as its activation function. The number of neurons in the hidden layer is determined using cross-validation, typically 1.5 times the number of neurons in the input layer. An output layer is constructed, also containing multiple neurons, the number determined based on the dimension of the evaluation index. A weight matrix is ​​established from the input layer to the hidden layer, representing the connection relationship between each neuron in the input layer and each neuron in the hidden layer. Initial weight values ​​are random decimals. A weight matrix is ​​also established from the hidden layer to the output layer, representing the connection relationship between each neuron in the hidden layer and each neuron in the output layer. Initial weight values ​​are random decimals. An adaptive moment estimation optimization algorithm is used to train the neural network, setting the learning rate to 0.001 and the training epochs to 500, resulting in the interference evaluation model.

[0157] The specific implementation of step S40 involves collecting data from the communication equipment to be evaluated and extracting characteristic parameters. This includes collecting runtime data, maintenance time data, and fault record data of the communication equipment to be evaluated, and preprocessing and standardizing the collected data. It also involves collecting electromagnetic radiation intensity data from photovoltaic modules surrounding the communication equipment, ambient air humidity data, and ambient temperature data, and preprocessing and standardizing the collected data. From the collected data, time-domain characteristic parameters of the interference signal are extracted, including signal amplitude, duration, period, and signal strength. The time-domain characteristic parameters are then transformed into frequency-domain characteristic parameters, including frequency point amplitude, phase, and phase difference.

[0158] The specific implementation of step S50 involves inputting the characteristic parameters of the communication device to be evaluated into the interference evaluation model. The time-domain characteristic parameters of the interference signal of the communication device to be evaluated are arranged into a time-domain characteristic parameter matrix, with each row corresponding to a time point. The number of rows in the matrix equals the number of sampling points, and the number of columns equals the dimension of the time-domain characteristic parameters. The frequency-domain characteristic parameters of the interference signal of the communication device to be evaluated are arranged into a frequency-domain characteristic parameter matrix, with each row corresponding to a frequency point. The number of rows in the matrix equals the number of frequency points, and the number of columns equals the dimension of the frequency-domain characteristic parameters. The time-domain characteristic parameter matrix and the frequency-domain characteristic parameter matrix are input into the interference evaluation model, and the interference evaluation result is obtained through forward propagation calculation of a neural network.

[0159] The specific implementation of step S60 involves determining the anti-interference level of the communication equipment. An anti-interference level evaluation standard is established, dividing the anti-interference level into five levels: excellent, good, average, poor, and very poor. Each level has a corresponding range for interference assessment results: excellent (above 90 points), good (80-90 points), average (70-80 points), poor (60-70 points), and very poor (below 60 points). The interference assessment results are compared with the anti-interference level evaluation standard, and the anti-interference level of the communication equipment is determined based on the comparison results. This anti-interference level is used as an evaluation index for the anti-interference capability of the communication equipment.

[0160] The specific implementation of step S70 involves component decomposition of the interference signal. The time-domain characteristic parameters of the interference signal are integrated according to the observation period, with the integration interval being the entire observation period. The trapezoidal integration method is used for numerical calculation to obtain the stable components of the interference signal. The stable components are then subtracted from the time-domain characteristic parameters of the interference signal to obtain the variable components. The stable components reflect the long-term trend and baseline level of the interference signal, while the variable components reflect the fluctuations and instantaneous changes in the interference signal.

[0161] The specific implementation of step S80 involves calculating the interference source influence coefficient matrix and the anti-interference coefficient matrix. Based on the electromagnetic radiation intensity data of the photovoltaic module, ambient air humidity data, and ambient temperature data, a power-law composite function model is used to calculate the degree of environmental influence on the communication equipment. The sensitivity index and attenuation coefficient in the function are obtained through fitting experimental data. Based on the equipment runtime data, maintenance time data, and failure frequency data, a double-exponential model is used to calculate the anti-interference capability of the communication equipment under its current operating state. The performance attenuation coefficient and stability coefficient in the model are determined through historical data analysis. Based on the interference occurrence frequency data, interference duration data, and communication interruption duration data, a power-law combination model is used to calculate the intensity of interference influence on the communication equipment. The influence index and suppression coefficient in the model are determined through expert experience and data analysis. The degree of environmental influence, anti-interference capability, and influence intensity are combined in a block-diagonalized manner to form the interference source influence coefficient matrix. The inverse matrix of the interference source influence coefficient matrix is ​​calculated using Gaussian-Jordan elimination. The inverse matrix is ​​multiplied by the standardized coefficient matrix and the correction coefficient matrix to obtain the communication equipment anti-interference coefficient matrix.

[0162] The specific implementation of step S90 involves correcting the interference assessment results and establishing an early warning mechanism. The interference assessment result is multiplied by the communication equipment anti-interference coefficient matrix and the evaluation weight diagonal matrix to obtain the corrected interference assessment value. The evaluation weights are determined using the analytic hierarchy process (AHP). The standard deviation of the corrected interference assessment value is calculated using the sample standard deviation method. An early warning threshold is established based on the corrected interference assessment value and the standard deviation, set as the corrected assessment value plus twice the standard deviation. The communication equipment is monitored in real time, and the real-time assessment value is calculated using the sliding window method. When the real-time monitoring value exceeds the early warning threshold, an early warning signal is issued. The early warning signal includes both audible and visual alarms and system prompts.

[0163] This implementation method, through reasonable data acquisition, feature extraction, model construction, and early warning mechanism settings, achieves effective assessment and early warning of interference to communication equipment in photovoltaic power plants, demonstrating strong practicality and reliability. All algorithms and models used in this method have undergone thorough theoretical analysis and experimental verification, ensuring the accuracy and credibility of the assessment results. By conducting dual time-domain and frequency-domain analysis of interference signals, and considering the influence of environmental factors and equipment status, a complete assessment index system has been constructed, providing strong technical support for the operation and maintenance management of communication equipment in photovoltaic power plants.

[0164] A second aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores program instructions, which, when executed in a computer, are used to perform the above-described method for evaluating interference in communication equipment of a photovoltaic power station.

[0165] A third aspect of the present invention provides a photovoltaic power plant communication equipment interference assessment system, wherein the system includes the aforementioned computer-readable storage medium.

[0166] Specifically, the principle of this invention is:

[0167] Multi-dimensional data acquisition and feature extraction principle: The data acquired in this invention includes two main categories: communication equipment operation data and environmental electromagnetic data. Communication equipment operation data reflects the usage status and performance changes of the equipment, while environmental electromagnetic data reflects the characteristics of external interference sources. Features such as amplitude and duration of the interference signal are extracted through time-domain analysis, and then frequency-domain conversion is performed using Fourier transform to obtain features such as spectral distribution and energy density. This dual time-domain and frequency-domain feature extraction method can comprehensively describe the characteristics of the interference signal, providing complete feature information for subsequent modeling.

[0168] Neural Network Modeling Principle: This invention employs a multilayer perceptron neural network to construct an interference assessment model. This network, through the nonlinear mapping of hidden layer neurons, can learn and fit the complex relationship between interference features and assessment results. During training, the network weights are continuously adjusted using the backpropagation algorithm, ultimately establishing an accurate interference assessment model. The adaptive learning capability of the neural network allows the model to be continuously optimized, improving assessment accuracy.

[0169] Interference Component Decomposition and Influence Coefficient Calculation Principle: This invention innovatively proposes a method for decomposing stable and variable components of interference signals. Stable components reflect persistent, long-term interference, while variable components reflect sudden and temporary interference. By establishing environmental impact equations, equipment state equations, interference characteristic equations, and comprehensive evaluation equations, a complete set of interference source influence coefficient matrix equations is constructed. This set of equations can accurately quantify the degree of influence of different factors on interference.

[0170] Dynamic Correction and Early Warning Mechanism Principle: This invention corrects the original evaluation results by calculating the anti-interference coefficient matrix of the communication equipment. The correction process takes into account the actual anti-interference capability of the equipment, making the evaluation results more consistent with reality. Simultaneously, a dynamic early warning threshold is established based on the corrected evaluation values, enabling real-time monitoring and early warning of interference risks.

[0171] The organic integration of these technical principles forms a complete interference assessment and early warning system. This system conforms to the basic principles of systems engineering, possessing a clear theoretical foundation and rigorous logical relationships. Through comprehensive analysis of multi-dimensional data, accurate assessment of interference conditions is achieved; through adaptive learning of neural networks, continuous optimization of the assessment model is ensured; through interference component decomposition and influence coefficient matrix, accurate identification of interference sources is achieved; and through dynamic correction and early warning mechanisms, the real-time effectiveness of the assessment results is ensured.

[0172] The following provides a specific embodiment 1 of the present invention. The specific implementation of each step in this embodiment 1 is described in detail below: The specific implementation of step S10 is as follows:

[0173] Step 101 involves collecting runtime data, maintenance time data, and fault record data for each communication device within the photovoltaic power station, and arranging them into a device operation data matrix D, as shown below:

[0174]

[0175] Where, t i1 Let m be the runtime of the i-th device. i1 Let f be the maintenance time for the i-th device. i1Let be the number of failures for the i-th device, and n be the number of devices. The purpose of this step is to record the usage history of each communication device.

[0176] Step 102 involves collecting electromagnetic radiation intensity data, ambient air humidity data, and ambient temperature data around each communication device within the photovoltaic power station, and arranging them into an environmental electromagnetic data matrix E, as shown below:

[0177]

[0178] Where, r i1 Let h be the electromagnetic radiation intensity at the i-th position. i1 Let t be the air humidity at the i-th location. i1 Let be the temperature value at the i-th location. The purpose of this step is to obtain data such as electromagnetic interference, humidity, and temperature of the environment in which the communication equipment is located.

[0179] Step 103 involves storing the equipment operation data matrix D and the environmental electromagnetic data matrix E into the basic database of communication equipment interference. The purpose of this step is to organize and summarize the various relevant data collected, providing support for subsequent interference analysis and assessment.

[0180] The specific implementation method of step S20 is as follows:

[0181] Step 201 involves using a high-speed data acquisition card to sample data from the communication equipment interference database, setting the sampling frequency to f. s =2000Hz, sampling duration set to T=48h. The purpose of this step is to obtain raw interference data with high temporal resolution.

[0182] Step 202 is to extract the amplitude A of the interference signal from the sampled data. i Interference duration T i Interference period P i and signal strength S i The time-domain feature parameters are equal and weighted to obtain the time-domain feature parameter vector F. t , indicates the following:

[0183] f ti =α i ·A i +β i ·T i +γ i ·P i +δ i ·S i +ε t ;

[0184] Where, α i ,β i ,γi ,δ i ε is the weighting coefficient. t This represents the time-domain error term. The purpose of this step is to quantify and describe the key time-domain characteristics of the interference signal.

[0185] Step 203 is to process the temporal feature parameter vector F t Perform a Fast Fourier Transform (FFT) with a Hanning window and a 50% overlap. The purpose of this step is to transform the time-domain feature parameters to the frequency domain.

[0186] Step 204 is to extract the frequency point amplitude 'a' from the FFT result. i1 Frequency point phase p i1 Phase difference φ at frequency point i1 Equal frequency domain characteristic parameters are arranged into a frequency domain characteristic parameter matrix F. f , indicates the following:

[0187]

[0188] Where m represents the number of frequency points. The purpose of this step is to comprehensively describe the characteristics of the interference signal in the frequency domain.

[0189] The specific implementation method of step S30 is as follows:

[0190] Step 301 is to convert the time-domain feature parameter vector F t and frequency domain characteristic parameter matrix F f The input layer of a neural network is constructed using these as input parameters.

[0191] Step 302 involves constructing the hidden layer of the neural network, setting up h neurons, with each neuron using the Modified Linear Unit (ReLU) function as its activation function, as shown below:

[0192]

[0193] Where, H l The output of the l-th hidden layer neuron is given by σ, where σ is the ReLU activation function and x is the output of the l-th hidden layer neuron. i For the input features, w li For weight, b l This is for biasing. The purpose of this step is to design the hidden layer structure to improve the expressive power of the model.

[0194] Step 303 involves constructing the output layer of the neural network, setting up o neurons. The purpose of this step is to design the output layer structure and determine the dimensions and format of the final output.

[0195] Steps 304 and 305 establish the weight matrix W from the input layer to the hidden layer and the weight matrix V from the hidden layer to the output layer, respectively, as follows:

[0196]

[0197] The purpose of these two steps is to initialize the parameters of the neural network model, laying the foundation for subsequent model training.

[0198] Step 306 involves training the neural network using the Adaptive Moment Estimation (Adam) optimization algorithm, with a learning rate set to η = 0.001 and the number of training epochs set to N = 500, to obtain the interference evaluation model. The purpose of this step is to train the neural network model through the optimization algorithm so that it can effectively learn the interference evaluation rules.

[0199] The specific implementation method of step S40 is as follows:

[0200] Step 401 involves collecting runtime data t, maintenance time data τ, and fault record data f of the communication equipment to be evaluated. The purpose of this step is to obtain the historical usage data of the equipment itself.

[0201] Step 402 involves collecting electromagnetic radiation intensity data R, ambient air humidity data H, and ambient temperature data T around the communication equipment to be evaluated. The purpose of this step is to obtain relevant data about the environment in which the equipment to be evaluated is located.

[0202] Step 403 involves extracting the time-domain and frequency-domain characteristic parameters of the interference signal from the collected data. The purpose of this step is to extract characteristic parameters from the actual operating data of the device to be evaluated, following the method described in step S20 above, thus laying the foundation for subsequent interference evaluation model input.

[0203] The specific implementation method of step S50 is as follows:

[0204] Step 501 involves arranging the time-domain characteristic parameters of the device to be evaluated into a matrix F′ according to time points. t .

[0205] Step 502 involves arranging the frequency domain characteristic parameters of the device to be evaluated into a matrix F′ based on frequency points. f .

[0206] Step 503 is to set F′ t and F′ f The input is fed into the aforementioned interference assessment model to obtain the interference assessment result V. original The purpose of this step is to use the trained neural network model to conduct a comprehensive analysis and evaluation of the actual situation of the equipment being evaluated.

[0207] The specific implementation method of step S60 is as follows:

[0208] Step 601 involves setting the anti-interference level evaluation criteria, dividing the anti-interference level into k levels, and setting a corresponding range of interference assessment results for each level. The purpose of this step is to determine a reasonable standard for classifying anti-interference levels.

[0209] Step 602 is to use the interference assessment result V original Compare with the anti-interference level evaluation criteria.

[0210] Step 603 involves determining the anti-interference level of the device under evaluation based on the comparison results, which serves as the final anti-interference capability evaluation index. The purpose of this step is to determine the specific anti-interference level of the device under evaluation.

[0211] The specific implementation method of step S70 is as follows:

[0212] Step 701 involves adjusting the time-domain feature parameter vector F based on the observation period T. t By performing integration, the stable component S of the interference signal is obtained. c :

[0213]

[0214] Where s(t) is the original signal, ε s This is the error term for the stable components. The purpose of this step is to extract the stable components from the interference signal.

[0215] Step 702 is to convert the time-domain feature parameter vector F t Subtract the stable component S c The interference signal variation component D is obtained. c :

[0216] D c =s(t)-S c +ε d ;

[0217] Where, ε d This represents the error term for the variable components. The purpose of this step is to separate the variable components of the interference signal.

[0218] The specific implementation method of step S80 is as follows:

[0219] Step 801 involves establishing the environmental impact equation E based on the electromagnetic radiation intensity R, ambient humidity H, and ambient temperature T, as follows:

[0220]

[0221] Where R0, H0, T0 are standard reference values, k1, k2, k3 are weighting coefficients, α, β, γ are sensitivity indices, λ1, λ2, λ3 are attenuation coefficients, and ε e This is the environmental error term. The purpose of this step is to assess the extent to which environmental factors affect the communication equipment.

[0222] Step 802 involves establishing the equipment state equation S based on the running time t, maintenance time τ, and number of failures f, as follows:

[0223]

[0224] Where λ1, λ2, λ3 are weighting coefficients, μ1, μ2, μ3 are performance degradation coefficients, ν1, ν2, ν3 are stability coefficients, and ε s This is the state error term. The purpose of this step is to assess the impact of the device's own state on its ability to resist interference.

[0225] Step 803 involves establishing the interference characteristic equation I based on the number of interference occurrences N, the duration of interference D, and the duration of communication interruption B, as follows:

[0226]

[0227] Where N0, D0, and B0 are reference thresholds, ω1, ω2, and ω3 are weighting coefficients, δ, θ, and φ are influence indices, η1, η2, and η3 are suppression coefficients, and ε i This is the characteristic error term. The purpose of this step is to evaluate the impact characteristics of interference signals on communication equipment.

[0228] Step 804 involves combining the environmental impact E, equipment status S, and interference characteristics I in a diagonal matrix to form the interference source influence coefficient matrix M:

[0229]

[0230] Where, ρ i ,σ i ,τ i ε is the comprehensive weighting coefficient. m This represents the comprehensive error term. The purpose of this step is to construct a matrix model describing the impact of the interference source.

[0231] Step 805 is to calculate the inverse matrix M of the interference source influence coefficient matrix M. -1 Multiplying this matrix by the normalized coefficient matrix K and the correction coefficient matrix C yields the communication equipment anti-interference coefficient matrix R:

[0232] R=M -1 ·K·C+ε r ;

[0233]

[0234] Where, c i ε represents the correction coefficients for each dimension. r This represents the anti-interference error term. The purpose of this step is to further calculate the anti-interference capability coefficient matrix of the communication equipment itself based on the interference source influence coefficient matrix.

[0235] The specific implementation method of step S90 is as follows:

[0236] Step 901 is to use the original interference assessment result v original Multiplying by the communication equipment anti-interference coefficient matrix R and the evaluation weight diagonal matrix W, we obtain the corrected interference evaluation value V. modified :

[0237] V modified =V original ·R·W+ε v ;

[0238]

[0239] Where, w i ε is the weight for evaluating each dimension. v This step is to correct the error term. The purpose of this step is to correct and adjust the original interference assessment results.

[0240] Step 902 is to calculate the corrected evaluation value V. modified Standard deviation σ v , and according to V modified and σ V Establish an early warning threshold T warning :

[0241] T warning =α·V modified +β·σ V +γ;

[0242] Where α and β are weighting coefficients, and γ is the baseline threshold. The purpose of this step is to set a reasonable warning threshold to provide a basis for subsequent real-time monitoring and early warning.

[0243] Step 903 involves real-time monitoring of the communication equipment. When the real-time monitoring value exceeds the warning threshold T... warning The purpose of this step is to continuously monitor communication equipment and issue timely warnings using the previously established warning thresholds.

[0244] To better understand and implement this invention, a specific application scenario is provided below as Example 2: A photovoltaic power station adopts a centralized layout, mainly composed of photovoltaic arrays, combiner boxes, transformers, inverters, and a monitoring system. The monitoring system plays a crucial role in the real-time monitoring and remote control of various devices within the power station. The communication equipment of the monitoring system mainly includes industrial Ethernet switches, remote terminal units, and wireless communication modules. Due to the complex environment of the power station, these communication devices are frequently subjected to electromagnetic interference from photovoltaic arrays, inverters, and other equipment, leading to frequent communication failures and seriously affecting the automated operation of the power station.

[0245] To address this issue, the power plant's operation and maintenance personnel decided to adopt the communication equipment interference assessment and early warning method proposed in this invention to conduct a comprehensive analysis and dynamic monitoring of the power plant's internal communication equipment. The following describes the specific implementation process and effects of this method at the power plant.

[0246] Data collection phase:

[0247] First, the power plant's operation and maintenance personnel collected operational data from the communication equipment and environmental electromagnetic data over a period of one week.

[0248] Specifically, it includes:

[0249] 1. Operational Data Collection: A total of 20 communication devices were installed, and the runtime, last maintenance time, and cumulative number of failures for each device were recorded. The data collection period was one week, and the statistical results are shown in Table 1.

[0250] Table 1 Communication Equipment Operation Data

[0251] Equipment Number Runtime (h) Maintenance time (d) Number of failures (times) 1 4320 180 3 2 4680 210 2 3 4800 240 1 4 5040 270 2 5 4560 150 4 6 4440 240 3 7 4920 210 2 8 5280 270 1 9 4800 180 2 10 4680 240 3 11 5040 180 2 12 4800 210 1 13 4560 240 2 14 4320 270 3 15 4800 150 2 16 4440 210 1 17 4920 240 2 18 5280 180 1 19 4680 210 2 20 4560 240 3

[0252] 2. Environmental Data Collection: A total of 20 environmental monitoring points were installed inside the power station, recording the electromagnetic radiation intensity of the photovoltaic modules, ambient air humidity, and ambient temperature at each monitoring point. The data collection period was also one week, and the statistical results are shown in Table 2.

[0253] Table 2 Environmental Electromagnetic Data

[0254]

[0255]

[0256] Through the data collection over the past week, the power plant maintenance personnel have initially established a basic database of communication equipment interference.

[0257] Time-frequency domain analysis stage:

[0258] Next, the power plant maintenance personnel performed time-domain analysis and frequency-domain conversion on the collected data. The specific steps are as follows:

[0259] 1. Temporal feature parameter extraction:

[0260] A sampling frequency of 2000Hz was used to continuously sample data from the interference database for 48 hours.

[0261] Time-domain characteristic parameters such as interference signal amplitude, duration, period, and signal strength are extracted from the sampled data.

[0262] We perform weighted calculations on these time-domain feature parameters to obtain the time-domain feature parameter vector F. t .

[0263] 2. Frequency domain feature parameter extraction:

[0264] Perform an FFT transformation on the time-domain feature parameter vector using a Hanning window with a 50% overlap.

[0265] Frequency domain feature parameters such as frequency point amplitude, frequency point phase, and frequency point phase difference are extracted from the FFT results.

[0266] These frequency domain feature parameters are arranged into a frequency domain feature parameter matrix F f .

[0267] Through the above analysis, power plant operation and maintenance personnel obtained key parameters describing the time-frequency characteristics of interference signals. Next, these parameters will be input into the interference assessment model.

[0268] Interference assessment modeling phase:

[0269] The power plant operation and maintenance personnel constructed a disturbance assessment model based on neural networks. The specific process is as follows:

[0270] 1. Model Input:

[0271] The time-domain feature parameter vector F t and frequency domain characteristic parameter matrix F f As the input layer of a neural network.

[0272] 2. Model Structure:

[0273] 100 neurons are set in the hidden layer, and each neuron uses the ReLU activation function.

[0274] Five neurons are set in the output layer, each corresponding to a different level of anti-interference.

[0275] 3. Model Training:

[0276] The Adam optimization algorithm was used to train the neural network, with a learning rate of 0.001 and 500 training epochs. During training, the network adaptively learned the interference patterns contained in the time-frequency features and continuously optimized its internal weights and bias parameters.

[0277] After training, the power plant maintenance personnel obtained a neural network model that can evaluate the anti-interference performance of communication equipment.

[0278] Interference impact analysis phase:

[0279] With the interference assessment model in place, the power plant operation and maintenance personnel further analyzed the impact of environmental factors and equipment status on the anti-interference capability of communication equipment. Specifically:

[0280] 1. Environmental Impact Analysis:

[0281] Based on the environmental data in Table 2, an environmental impact equation E is established to quantitatively assess the degree of interference of electromagnetic radiation intensity, humidity, and temperature on communication equipment.

[0282] The calculated environmental impact coefficient E is 0.78, indicating that environmental factors have caused significant interference to the communication equipment.

[0283] 2. Equipment Status Analysis:

[0284] Based on the equipment operation data in Table 1, establish the equipment state equation S and evaluate the anti-interference capability of the communication equipment itself.

[0285] The calculated equipment state coefficient S is 0.71, indicating that the performance of the equipment itself also has a certain impact on its anti-interference capability.

[0286] 3. Interference Feature Analysis:

[0287] Based on the actual number of interference occurrences, duration, and communication interruption duration monitored, interference characteristic equation I is established to assess the intensity of the interference signal's impact on communication equipment.

[0288] The calculated interference influence coefficient I is 0.82, indicating that the characteristics of the interference signal are also an important factor affecting the anti-interference capability.

[0289] The three influencing factors mentioned above are combined into an interference source influence coefficient matrix M, and then its inverse matrix M is calculated. - 1. This allows us to obtain the anti-interference coefficient matrix R of the communication equipment. This comprehensively quantifies the combined impact of environment, equipment, and interference on the anti-interference performance of the communication equipment.

[0290] Real-time monitoring and early warning stage:

[0291] With the interference assessment model and anti-interference coefficient matrix in place, the power plant operation and maintenance personnel finally established a real-time monitoring and early warning mechanism:

[0292] 1. Calculation of the corrected interference assessment value:

[0293] Multiplying the original interference assessment result by the anti-interference coefficient matrix R and the diagonal matrix W of the assessment weights yields the corrected interference assessment value V. modified .

[0294] Calculate V modified Standard deviation σ V .

[0295] 2. Warning threshold setting:

[0296] According to V modified and σ VEstablish an early warning threshold T w arning = 0.8 * V m odified+0.2*σ V +50.

[0297] 3. Real-time monitoring and early warning:

[0298] Real-time monitoring of communication equipment is performed, and when the monitored value exceeds the warning threshold T... w When arning occurs, an early warning signal should be issued immediately.

[0299] The warning signal will be sent to the power plant control center through the remote terminal unit to remind the operation and maintenance personnel to take timely interference suppression measures.

[0300] Through the implementation of the above stages, the power station has successfully established a complete communication equipment interference assessment and early warning system. The following is a detailed analysis of the application effects of this solution:

[0301] 1. Analysis of Interference Assessment Results:

[0302] Twenty communication devices were input into the trained neural network model, and their anti-interference level evaluation results were as follows: 3 devices were at level 5 (highest); 5 devices were at level 4; 7 devices were at level 3; 4 devices were at level 2; and 1 device was at level 1 (lowest).

[0303] The results show that most devices have strong anti-interference capabilities, but a few devices still have weak anti-interference performance and require special attention.

[0304] 2. Analysis of the impact of interference sources:

[0305] According to the calculation, the environmental impact coefficient E is 0.78, the equipment state coefficient S is 0.71, the interference characteristic coefficient I is 0.82, and the final result of the comprehensive impact coefficient matrix M is 0.64.

[0306] This indicates that environmental factors, the condition of the equipment itself, and the characteristics of interference signals all reduce the anti-interference ability of communication equipment to some extent, especially the impact of the interference signals themselves.

[0307] 3. Analysis of the effectiveness of the early warning system:

[0308] During real-time monitoring, the monitoring values ​​of two devices exceeded the warning threshold T. w arning.

[0309] Upon receiving the warning signal, the maintenance personnel immediately took measures such as shielding and filtering, successfully suppressing the interference and restoring the communication equipment to normal operation.

[0310] The establishment of the early warning mechanism has significantly improved the power station's response speed to interference with communication equipment and effectively avoided serious communication failures.

[0311] It should be noted that the variables involved in this invention are explained in detail in Table 3 below.

[0312] Table 3. Variable Explanation Table

[0313]

[0314]

[0315] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for assessing interference of a communication device of a photovoltaic power plant, characterized in that, Comprise the following steps: S10, collect the communication equipment running data and the environmental electromagnetic data in the photovoltaic power station, the communication equipment running data includes the communication equipment running time, the communication equipment maintenance time, the communication equipment fault record, the environmental electromagnetic data includes the photovoltaic module electromagnetic radiation intensity, environmental air humidity value, environmental temperature value, establish communication equipment interference basic database; S20, the time domain analysis is carried out to the data in the communication equipment interference basic database, extracts the time domain characteristic parameter of interference signal, and the time domain characteristic parameter of interference signal is converted to frequency domain, obtains the frequency domain characteristic parameter of interference signal; S30, according to the time domain characteristic parameter of interference signal and the frequency domain characteristic parameter of interference signal, an interference evaluation index system is established, a neural network model is constructed based on the interference evaluation index system, and an interference evaluation model is trained; S40, collect the communication equipment running data to be evaluated, extract the time domain characteristic parameter of interference signal and the frequency domain characteristic parameter of interference signal of the communication equipment to be evaluated; S50, the time domain characteristic parameter of interference signal and the frequency domain characteristic parameter of interference signal of the communication equipment to be evaluated are input into the interference evaluation model, and the interference evaluation result is obtained; S60, according to the interference evaluation result, the communication equipment anti-interference level is determined; S70, the time domain characteristic parameter of interference signal is decomposed into stable component and variable component, and the interference signal stable component and the interference signal variable component are obtained; S80, according to the interference signal stable component and the interference signal variable component, the interference source influence coefficient matrix is calculated through equation group, and based on the interference source influence coefficient matrix, the communication equipment anti-interference coefficient matrix is calculated; S90, according to the communication equipment anti-interference coefficient matrix, the interference evaluation result is corrected, and the corrected interference evaluation value is obtained, the corrected interference evaluation value is the original interference evaluation value multiplied by the communication equipment anti-interference coefficient matrix, and based on the corrected interference evaluation value, the communication equipment interference early warning threshold is established, and real-time monitoring and early warning are carried out.

2. The method of claim 1, wherein, The step S10 specifically comprises: Step 101, collect the running time data, maintenance time data and fault record data of each communication equipment in the photovoltaic power station, arrange the collected running time data, maintenance time data and fault record data according to each row corresponding to a communication equipment, form a device running data matrix; Step 102, collect the photovoltaic module electromagnetic radiation intensity data, environmental air humidity value data and environmental temperature value data around each communication equipment in the photovoltaic power station, arrange the collected photovoltaic module electromagnetic radiation intensity data, environmental air humidity value data and environmental temperature value data according to each row corresponding to a communication equipment, form an environmental electromagnetic data matrix; Step 103, the device running data matrix and the environmental electromagnetic data matrix are respectively stored in the communication equipment interference basic database; The step S20 specifically comprises: Step 201, using a high-speed data acquisition card to sample the data in the communication equipment interference basic database, setting the sampling frequency to 2 kHz, and setting the sampling time length to 48 hours; Step 202, extracting the interference signal amplitude data, interference duration data, interference period data and signal strength data from the sampled data, and performing weighted calculation on the extracted interference signal amplitude data, interference duration data, interference period data and signal strength data to obtain time domain characteristic parameters; Step 203, performing fast Fourier transform processing on the time domain characteristic parameters, setting the Hanning window and the overlap rate to 50%; Step 204, extracting the frequency point amplitude data, frequency point phase data and frequency point phase difference data from the fast Fourier transform result, and arranging the extracted frequency point amplitude data, frequency point phase data and frequency point phase difference data into a frequency domain characteristic parameter matrix.

3. The method of claim 2, wherein, The step S30 specifically comprises: Step 301, constructing a neural network input layer by taking the interference signal time domain characteristic parameters and the interference signal frequency domain characteristic parameters as input parameters; Step 302, constructing a neural network hidden layer, and setting a plurality of neurons in the neural network hidden layer, each of which uses a corrected linear unit function as an activation function; Step 303, constructing a neural network output layer, and setting a plurality of neurons in the neural network output layer; Step 304, establishing a weight matrix from the input layer to the hidden layer, which is used to represent the connection relationship between each neuron of the input layer and each neuron of the hidden layer; Step 305, establishing a weight matrix from the hidden layer to the output layer, which is used to represent the connection relationship between each neuron of the hidden layer and each neuron of the output layer; Step 306, training the neural network by using an adaptive matrix estimation optimization algorithm, setting the learning rate to 0.001, and setting the training rounds to 500 rounds to obtain an interference evaluation model.

4. The method of claim 3, wherein, The step S40 specifically comprises: Step 401, collecting the running time data, maintenance time data and fault record data of the communication equipment to be evaluated; Step 402, collecting the photovoltaic component electromagnetic radiation intensity data, environmental air humidity value data and environmental temperature value data around the communication equipment to be evaluated; Step 403, extracting the interference signal time domain characteristic parameters and frequency domain characteristic parameters from the collected data.

5. The method of claim 4, wherein, The step S50 specifically comprises: Step 501, arranging the interference signal time domain characteristic parameters of the communication equipment to be evaluated into a time domain characteristic parameter matrix in a manner that each row corresponds to a time point; Step 502, arranging the interference signal frequency domain characteristic parameters of the communication equipment to be evaluated into a frequency domain characteristic parameter matrix in a manner that each row corresponds to a frequency point; Step 503, inputting the time domain characteristic parameter matrix and the frequency domain characteristic parameter matrix into the interference evaluation model to obtain an interference evaluation result; The step S60 specifically comprises: Step 601, setting an anti-interference level evaluation criterion, and dividing the anti-interference level into a plurality of levels, each of which sets a corresponding interference evaluation result range; Step 602, compare the interference evaluation result with the anti-interference level evaluation standard; Step 603, determine the anti-interference level to which the communication device belongs according to the comparison result, and take the anti-interference level as the communication device anti-interference capability evaluation index.

6. The method of claim 5, wherein, The step S70 specifically comprises: Step 701, integral operation is performed on the time domain characteristic parameters of the interference signal according to the observation period to obtain a stable component of the interference signal; Step 702, the time domain characteristic parameters of the interference signal are subtracted by the stable component of the interference signal to obtain a variable component of the interference signal.

7. A method of interference assessment of a communication device of a photovoltaic power plant according to claim 6, characterized in that The step S80 specifically comprises: Step 801, calculate the influence degree of the environment on the communication device according to the photovoltaic module electromagnetic radiation intensity data, the environmental air humidity value data and the environmental temperature value data; Step 802, calculate the anti-interference capability of the communication device in the current running state according to the device running time data, the maintenance time data and the fault frequency data; Step 803, calculate the influence intensity of the interference on the communication device according to the interference occurrence frequency data, the interference duration data and the communication interruption time data; Step 804, combine the environmental influence degree, the anti-interference capability and the influence intensity in a block diagonalization manner to form an interference source influence coefficient matrix; Step 805, calculate the inverse matrix of the interference source influence coefficient matrix, multiply the inverse matrix with a standardization coefficient matrix and a correction coefficient matrix to obtain a communication device anti-interference coefficient matrix.

8. The method of claim 7, wherein the method further comprises: The step S90 specifically comprises: Step 901, multiply the interference evaluation result by the communication device anti-interference coefficient matrix and an evaluation weight diagonal matrix to obtain a corrected interference evaluation value; Step 902, calculate the standard deviation of the corrected interference evaluation value, and establish a warning threshold according to the corrected interference evaluation value and the standard deviation; Step 903, real-time monitor the communication device, and send a warning signal when the real-time monitoring value exceeds the warning threshold.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores program instructions, and the program instructions are used to execute the photovoltaic power station communication device interference evaluation method in the computer.

10. A system for assessing interference of communication devices of a photovoltaic power plant, characterized in that The computer readable storage medium comprises the computer readable storage medium. The computer readable storage medium comprises the computer readable storage medium.

Images