Power equipment test data acquisition method and device based on multi-source data fusion

Abstract

The invention discloses a power equipment test data acquisition method and device based on multi-source data fusion. The method comprises the steps that each sensor acquires operation data of power equipment in real time; evaluating the operation data based on the abnormal condition and the equipment state to generate a multi-source data fusion parameter; generating a compression strategy parameter based on the multi-source data fusion parameter; based on multi-source data fusion parameters, fusing the operation data of the power equipment collected by the sensors in real time to obtain fused multi-source operation data; performing data compression on the fused multi-source operation data according to the compression strategy parameters; and transmitting the compressed fused multi-source operation data to a central processing unit. According to the invention, the real-time monitoring capability and the fault early warning level of the power equipment data acquisition system are improved, and the data transmission and storage cost is reduced.

Description

Technical Field

[0001] The present invention belongs to the field of computer technology, and in particular relates to a method and a device for collecting test data of electric power equipment based on multi-source data fusion. Background Art

[0002] With the continuous development of the power system and the improvement of its intelligence level, the demand for operating status monitoring and fault warning of power equipment is increasing. The traditional data collection method of power equipment mainly relies on a single sensor for data monitoring. This method is often difficult to fully and accurately reflect the actual operating status of the equipment when facing complex operating environments and changing equipment conditions. With the increase in the types and number of power equipment, the amount of data generated has shown explosive growth. The collection method of a single data source is difficult to meet the needs of real-time processing and analysis of large-scale, multi-dimensional data. In addition, the existing data acquisition system has a high degree of redundancy and transmission delay during data transmission, resulting in insufficient real-time performance and failure to detect and respond to equipment anomalies in a timely manner, affecting the overall operating efficiency and safety of the power system.

[0003] Therefore, there is an urgent need for an efficient and intelligent data acquisition method based on multi-source data fusion to improve the response speed, data processing capability and overall reliability of the power equipment monitoring system. Summary of the invention

[0004] In view of the defects existing in the above-mentioned prior art, the present invention provides a method for collecting test data of electric power equipment based on multi-source data fusion, the method comprising:

[0005] Step 1, each sensor collects the operating data of the power equipment in real time;

[0006] Step 2: Evaluate the operation data based on abnormal conditions and equipment status to generate multi-source data fusion parameters;

[0007] Step 3, generating compression strategy parameters based on multi-source data fusion parameters;

[0008] Step 4, based on the multi-source data fusion parameter, the operation data of the power equipment collected by the sensors in real time are fused to obtain fused multi-source operation data;

[0009] Step 5, compressing the fused multi-source operation data according to the compression strategy parameters;

[0010] Step 6: Transmit the compressed fused multi-source operation data to the central processing unit.

[0011] Furthermore, the method comprises:

[0012] Step 1: Each sensor collects the operating data of the power equipment in real time and transmits the data to the intelligent decision engine and the adaptive data fusion module;

[0013] Step 2: The intelligent decision engine receives the operation data of the power equipment collected in real time from various sensors, evaluates the operation data based on abnormal conditions and equipment status to generate multi-source data fusion parameters, and sends the multi-source data fusion parameters to the adaptive data fusion module;

[0014] Step 3: The intelligent decision engine generates compression strategy parameters based on multi-source data fusion parameters;

[0015] Step 4: The adaptive data fusion module fuses the multi-source operation data of the power equipment collected by the sensors in real time based on the multi-source data fusion parameters determined by the intelligent decision engine to obtain fused multi-source operation data, and sends the fused multi-source operation data to the intelligent compression module;

[0016] Step 5: The intelligent compression module compresses the fused multi-source operation data according to the compression strategy parameters determined by the intelligent decision engine;

[0017] Step 6: The compressed fused multi-source operation data is transmitted to the central processing unit in the cloud through the transmission module.

[0018] Among them, the execution time of step 4 is after step 2 and before step 5; the execution time of step 3 is after step 2 and before step 5; that is, step 2 and step 3 at least partially overlap in execution time.

[0019] Among them, in step 2, each sensor collects the power equipment operation data in real time, including voltage, current, temperature, vibration, and humidity parameters at key locations.

[0020] Among them, the operation data is evaluated based on abnormal conditions and equipment status to generate multi-source data fusion parameters, including:

[0021] Use the isolation forest algorithm to detect anomalies in the operating data of each type of sensor and obtain anomaly scores, which are as follows:

[0022] s(x): anomaly score of data point x;

[0023] E(h(x)): the average path length of data point x in the isolation forest;

[0024] c(n): adjustment parameter determined by the number of sensors n,

[0025] H(n): the nth harmonic number, That is, H(n-1) is the n-1th harmonic number;

[0026] A threshold is set according to the anomaly score, and data points with anomaly scores higher than the threshold are marked as abnormal data of this type of sensor, and the rest are normal data.

[0027] Among them, the operation data is evaluated based on abnormal conditions and equipment status to generate multi-source data fusion parameters, including:

[0028] When the proportion of abnormal data of a certain sensor exceeds a certain proportion of the total sampled data of the sensor, all operation data of the sensor are determined to be abnormal data, and all operation data collected by the sensor generating the abnormality are marked as abnormal data;

[0029] Furthermore, except for the sensor that has an abnormality, the operating data of other sensors are considered normal data.

[0030] Among them, the operation data is evaluated based on abnormal conditions and equipment status to generate multi-source data fusion parameters, which also includes:

[0031] Evaluate the equipment status of normal data of each type of sensor, identify potential equipment failure characteristics, and determine key data.

[0032] Among them, the normal data of each type of sensor is used to evaluate the equipment status, identify potential equipment failure characteristics, and determine key data, including:

[0033] Performing principal component analysis on the operating data of each type of sensor parameter, including constructing a data matrix of each type of sensor parameter, calculating a covariance matrix, and performing eigenvalue decomposition on the covariance matrix to obtain eigenvalues ​​and eigenvectors;

[0034] Select the first k principal components so that the cumulative contribution rate reaches the set threshold η;

[0035] Project the data into principal component space;

[0036] Calculate the reconstruction error for each sensor: in:

[0037] e i : The reconstruction error of the i-th sensor;

[0038] x i : the i-th raw sensor vector;

[0039] The reconstruction vector of the i-th sensor is: Z i is the data matrix after dimension reduction of the i-th sensor, W k is the matrix composed of the first k eigenvectors, T is the transpose of the matrix, is the mean vector of the data matrix;

[0040] Set the error threshold ε, when e i >ε, all corresponding sensor operation data are marked as key data in this type of sensor.

[0041] Among them, the data matrix satisfies the following relationship:

[0042]

[0043] in:

[0044] X: normal data matrix, dimension is n×p;

[0045] x ij : The jth eigenvalue of the i-th sensor of the corresponding sensor type, the sensor type is voltage, current, temperature, vibration, humidity, where the jth eigenvalue is the value collected in the j-th collection period;

[0046] n: number of sensors;

[0047] p: number of features (number of acquisition cycles);

[0048] Among them, in the step 3, compression strategy parameters are generated based on multi-source data fusion parameters, including an intelligent decision engine determining a compression ratio based on the multi-source data fusion parameters, the proportion of abnormal data and key data, and the available bandwidth.

[0049] wherein a compression algorithm is determined according to the determined compression ratio, and the received fused multi-source operation data is compressed;

[0050] When compression begins, the current available bandwidth is determined and a compression strategy is determined, including appending abnormal data segments and key data segments to the normal data packets if the available bandwidth permits; and, at the receiving end, comparing the abnormal data segments and key data segments appended to the normal data packets with the abnormal data packets and key data packets compressed losslessly to determine whether retransmission is required.

[0051] The present invention also discloses a power equipment test data acquisition system based on multi-source data fusion, which adopts the power equipment test data acquisition method based on multi-source data fusion. The system comprises:

[0052] A multi-source data acquisition module, which is used to collect the operating data of power equipment in real time;

[0053] An intelligent decision engine for evaluating operation data based on abnormal conditions and equipment status to generate multi-source data fusion parameters; and generating compression strategy parameters based on the multi-source data fusion parameters;

[0054] An adaptive data fusion module, which is used to fuse the operation data of the power equipment collected in real time by the sensors based on multi-source data fusion parameters to obtain fused multi-source operation data;

[0055] An intelligent compression module, which is used to compress the fused multi-source operation data according to compression strategy parameters;

[0056] The transmission module is used to transmit the compressed fused multi-source operation data to the central processing unit.

[0057] The present invention adopts a method combining an intelligent decision engine with an adaptive data fusion module, and the present invention realizes efficient fusion and dynamic compression of multi-source data, effectively reduces data redundancy, and improves the efficiency of data transmission and processing. The intelligent decision engine can perform anomaly detection and equipment status assessment based on real-time monitoring data, generate multi-source data fusion parameters and compression strategy parameters suitable for the current environment, and thus guide the adaptive data fusion module to optimize the data processing flow. The fusion results after data compression are efficiently transmitted to the cloud central processing unit through the transmission module, ensuring the real-time and integrity of the data. The overall technical solution not only improves the real-time monitoring capability and fault warning level of the power equipment data acquisition system, but also significantly reduces the cost of data transmission and storage, and improves the overall reliability and scalability of the system.

[0058] The method of the present invention not only improves the real-time monitoring capability and fault warning level of the power equipment data acquisition system, but also significantly reduces the cost of data transmission and storage, and improves the overall reliability and scalability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present disclosure will become readily understood. In the accompanying drawings, several embodiments of the present disclosure are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0060] Figure 1 The flowchart shows a method for collecting test data of electric power equipment based on multi-source data fusion according to an embodiment of the present invention.

[0061] Figure 2 It is a schematic diagram showing a power equipment test data acquisition system based on multi-source data fusion according to an embodiment of the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings, and "multiple" generally includes at least two.

[0064] It should be understood that although the terms first, second, third, etc. may be used to describe ... in the embodiments of the present invention, these ... should not be limited to these terms. These terms are only used to distinguish .... For example, without departing from the scope of the embodiments of the present invention, the first ... may also be referred to as the second ..., and similarly, the second ... may also be referred to as the first ....

[0065] It should be understood that the term "and / or" used in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0066] As used herein, the words "if" and "if" may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0067] It should also be noted that the term "includes", "comprising" or any other variation thereof is intended to cover non-exclusive inclusion, so that a commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprising a ..." do not exclude the existence of other identical elements in the commodity or device including the elements.

[0068] Traditional data collection methods often have difficulty in effectively integrating real-time data from multiple sensors, resulting in inefficient data processing and difficulty in timely detection of equipment anomalies or status assessments. In addition, with the dramatic increase in the amount of data, transmission and storage costs have risen significantly, further restricting the performance and reliability of the power equipment monitoring system. The prior art lacks adaptability in the fusion and compression of multi-source data, making it difficult to dynamically adjust data processing strategies to adapt to different operating environments and equipment status changes, thereby affecting the response speed and accuracy of the overall system. The present invention provides solutions to technical problems such as data redundancy, transmission delays, and lack of real-time performance that exist in the process of collecting and processing massive multi-source data generated by power equipment during operation.

[0069] The present invention aims to provide a method for collecting test data of electric power equipment based on multi-source data fusion, which realizes the coordinated optimization of adaptive fusion and compression of data through an intelligent decision engine, thereby improving the efficiency and quality of test data collection of electric power equipment.

[0070] like Figure 1 As shown, the present invention discloses a method for collecting test data of electric power equipment based on multi-source data fusion, the method comprising:

[0071] Step 1: Each sensor collects the operating data of the power equipment in real time and transmits the data to the intelligent decision engine and the adaptive data fusion module.

[0072] Step 2: The intelligent decision engine receives the operating data of the power equipment collected in real time from various sensors, evaluates the operating data based on abnormal conditions and equipment status to generate multi-source data fusion parameters, and sends the multi-source data fusion parameters to the adaptive data fusion module.

[0073] Step 3: The intelligent decision engine generates compression strategy parameters based on multi-source data fusion parameters.

[0074] Step 4: The adaptive data fusion module fuses the operating data of the power equipment collected by the sensors in real time to obtain fused multi-source operating data based on the multi-source data fusion parameters determined by the intelligent decision engine, and sends the fused multi-source operating data to the intelligent compression module.

[0075] Step 5: The intelligent compression module compresses the fused multi-source operation data according to the compression strategy parameters determined by the intelligent decision engine.

[0076] Step 6: The compressed fused multi-source operation data is transmitted to the central processing unit in the cloud through the transmission module.

[0077] The present invention realizes efficient fusion and dynamic compression of multi-source data by adopting a method combining an intelligent decision engine with an adaptive data fusion module, effectively reduces data redundancy, and improves the efficiency of data transmission and processing. The intelligent decision engine can perform anomaly detection and equipment status assessment based on real-time monitoring data, generate multi-source data fusion parameters and compression strategy parameters suitable for the current environment, and thus guide the adaptive data fusion module to optimize the data processing flow. The fusion results after data compression are efficiently transmitted to the cloud central processing unit through the transmission module, ensuring the real-time and integrity of the data.

[0078] The execution time of step 4 is after step 2 and before step 5; the execution time of step 3 is after step 2 and before step 5. That is, step 2 and step 3 at least partially overlap in execution time.

[0079] The execution time of step 2 and step 3 in the present invention is arranged before step 5 and at least partially overlaps, forming a parallel processing mechanism.

[0080] By partially overlapping step 2 (the intelligent decision engine generates multi-source data fusion parameters) and step 3 (the intelligent decision engine generates compression strategy parameters), the system can achieve multi-task parallel processing. This parallelism effectively shortens the overall data processing time cycle and improves data processing efficiency. Compared with sequential execution of each step, the overall processing time is longer, which affects real-time performance.

[0081] In the process of multi-source data fusion and data compression, the required computing resources may have certain independence or partial overlap. Through parallel processing, processor and memory resources can be used more efficiently, avoiding idleness and waste of resources, thereby improving the overall performance and processing capacity of the system. By optimizing the execution time schedule of the steps, the method of the present invention achieves higher parallelism and lower latency in the multi-source data processing process, significantly improving the efficiency of data transmission and processing. This not only reduces the overall operating cost of the system, but also enhances the scalability and adaptability of the system, and can better cope with the dynamic changes in the operating environment of power equipment and the rapid growth of data volume.

[0082] In one embodiment, the multi-source data acquisition module is composed of a plurality of sensor nodes, which are respectively installed at key positions of the power equipment.

[0083] Each sensor node transmits the collected data to the adaptive data fusion module through the wireless network.

[0084] In one embodiment, the system collects the operating data of the power equipment in real time through a variety of sensors to ensure comprehensive monitoring of the performance and status of the equipment, including:

[0085] First, there are a variety of sensors, including voltage sensors, current sensors, temperature sensors, vibration sensors, and humidity sensors. These sensors can be fixed at key locations on the equipment, or the detection equipment can measure specific points on the power equipment by plugging in wires. For example, current and voltage sensors may be installed at the inlet and outlet of the equipment, temperature sensors may be located near key heat sources of the equipment, and vibration sensors may be located near mechanical connection points or rotating parts.

[0086] Each sensor is equipped with a real-time data acquisition module. The operating data collected by the sensor is initially processed by the local processing unit, such as filtering, denoising and signal amplification, to improve data quality and reliability. The processed data is transmitted to the adaptive data fusion module through a wireless communication protocol, such as WiFi or LoRa.

[0087] In one embodiment, the power equipment test data collection based on the distributed sensor network is realized by multi-source data fusion. The sensor node is connected to the adaptive data fusion module through the wireless communication network, and the adaptive data fusion module is deployed on the local server or edge computing device. The adaptive data fusion module has high-performance data receiving and processing capabilities, and can simultaneously process a large amount of real-time data streams from multi-source sensors.

[0088] In one embodiment, the sensor node in step 1 is responsible for real-time collection of specific types of operating data and preliminary processing. The wireless communication network enables efficient data transmission between the sensor node and the adaptive data fusion module, ensuring low latency and high bandwidth. The adaptive data fusion module centrally receives, manages and pre-processes data from each sensor, providing high-quality input for subsequent data fusion and compression.

[0089] In one embodiment, the intelligent decision engine receives real-time operation data from each sensor through an efficient communication interface. These data include voltage, current, temperature, vibration and humidity parameters at key locations, and each sensor collects any type of data such as voltage, current, temperature, vibration and humidity. The intelligent decision engine adopts a high-throughput data receiving mechanism to support concurrent processing of multiple data streams. The received data first passes through a pre-processing module.

[0090] In a certain embodiment, the intelligent decision engine sends the generated optimization parameters (multi-source data fusion parameters) to the adaptive data fusion module and the intelligent compression module through a secure and reliable communication protocol.

[0091] In one embodiment, the system architecture of the intelligent decision engine includes:

[0092] The data receiving module is responsible for receiving the data stream from each sensor in real time;

[0093] Data preprocessing module, including data normalization function;

[0094] The data analysis and evaluation module uses machine learning models and statistical analysis methods to analyze the received data and evaluate the current data fusion and compression requirements;

[0095] A parameter determination module, used to generate optimal data fusion and compression parameters;

[0096] Parameter delivery module: responsible for delivering parameters to corresponding related modules.

[0097] In one embodiment, in step 2, the intelligent decision engine receives the power equipment operation data collected in real time from various sensors, including parameters such as voltage, current, temperature, vibration, humidity, etc. at key locations. Based on the evaluation of abnormal conditions and equipment status, the intelligent decision engine generates multi-source data fusion parameters and sends these parameters to the adaptive data fusion module.

[0098] In one embodiment, the intelligent decision engine receives the operation data collected by each sensor in real time through an efficient communication interface, and normalizes parameter data of different dimensions and units to a unified range.

[0099] For each type of sensor (voltage, current, temperature, vibration, humidity), a normalization operation is performed. The normalization formula is:

[0100]

[0101] in:

[0102] x i : The i-th value of the original data;

[0103] x i ′: the i-th value of the normalized data;

[0104] x min 、x max : The minimum and maximum values ​​in the data set.

[0105] Afterwards, the intelligent decision engine uses machine learning models and statistical analysis methods to perform anomaly detection on the preprocessed data, identify anomalies in the data, and classify the data.

[0106] Optionally, an isolation forest algorithm is used to perform anomaly detection on data of each type of sensor parameter, wherein each type of sensor data includes voltage, current, temperature, vibration, and humidity.

[0107] The anomaly scoring formula is:

[0108] in:

[0109] s(x): anomaly score of data point x;

[0110] E(h(x)): the average path length of data point x in the isolation forest;

[0111] c(n): adjustment parameter determined by the number of sensors n,

[0112] H(n): the nth harmonic number,

[0113] A threshold is set according to the anomaly score, and data points with anomaly scores higher than the threshold are marked as anomaly data of this type of sensor data, and the rest are normal data.

[0114] Among them, according to the anomaly detection results, the data is divided into:

[0115] Normal data: data that is not marked as abnormal.

[0116] Abnormal data: data marked as abnormal.

[0117] Optionally, when the proportion of abnormal data of a certain sensor exceeds a certain proportion of the total sampled data of the sensor, it is determined that all data of the sensor are abnormal data, and all data collected by the sensor generating the abnormality are marked as abnormal data.

[0118] And, except for the sensor that generates the abnormality, the data of other sensors are considered as normal data. Optionally, the normal data of each type of sensor parameter is evaluated for the equipment status, potential equipment failure characteristics are identified, and key data is determined, including the following processes:

[0119] The data of each type of sensor parameter was subjected to principal component analysis (PCA):

[0120] Construct a data matrix of sensor parameters for each type:

[0121] in:

[0122] X: preprocessed normal data matrix, dimension is n×p;

[0123] x ij : The jth characteristic value of the i-th sensor of the corresponding sensor type, the sensor type is voltage, current, temperature, vibration, humidity, wherein the jth characteristic value may be the value collected in the j-th collection period;

[0124] n: number of sensors;

[0125] p: number of features (number of acquisition cycles).

[0126] Compute the covariance matrix:

[0127] in:

[0128] C: covariance matrix, dimension is p×p;

[0129] Mean vector of the data matrix.

[0130] Perform eigenvalue decomposition on the covariance matrix to obtain eigenvalues ​​and eigenvectors: in:

[0131] Λ=diag(λ1λ2,…,λ p ): diagonal matrix containing eigenvalues ​​in descending order;

[0132] W=[w1,w2,…,w p ]: The corresponding eigenvector matrix.

[0133] Select the first k principal components so that the cumulative contribution rate reaches the set threshold η (for example, 90%):

[0134]

[0135] Project the data into the principal component space: Z = XW k ,

[0136] in:

[0137] Z: data matrix after dimensionality reduction, dimension is n×k;

[0138] W k : A matrix consisting of the first k eigenvectors with a dimension of p×k.

[0139] Among them, key data identification includes:

[0140] Calculate the reconstruction error for each sensor:

[0141] in:

[0142] e i : The reconstruction error of the i-th sensor;

[0143] x i : the i-th raw sensor vector;

[0144] The reconstruction vector of the ith sensor.

[0145] Set the error threshold ε, when e i >ε, all corresponding sensor data are marked as key data in this type of sensor.

[0146] In a certain embodiment, when the intelligent decision engine has determined abnormal data and key data, it is necessary to effectively transmit this information to the adaptive data fusion module and enable it to perform corresponding data fusion according to the instructions.

[0147] The intelligent decision engine labels the data and classifies it as “normal data”, “abnormal data” or “critical data”. This labeling process is based on the previous anomaly detection and equipment status assessment results.

[0148] In one embodiment, to ensure that the intelligent decision engine and the adaptive data fusion module have consistent understanding of the parameters, a predefined structured format is used:

[0149] All sensor properties are arranged in a fixed order, such as voltage, current, temperature, vibration, and humidity.

[0150] The state encoding uses 2 bits of binary to represent the state of each sensor:

[0151] Normal: 00

[0152] Abnormal: 01

[0153] Critical: 10

[0154] The state information and fusion weights and other parameters are packed into compact binary bytes in a predefined order: for example, when there are five temperature sensors, the states of the five temperature sensors are normal, abnormal, critical, normal, and abnormal, respectively, and their binary representation is 0001100001, which are merged into 0001100001.

[0155] In a certain embodiment, the intelligent decision engine in step 3 generates compression strategy parameters based on multi-source data fusion parameters, including the intelligent decision engine determining an appropriate compression ratio based on the multi-source data fusion parameters, taking into account the proportion of abnormal data and key data and the available bandwidth.

[0156] In one embodiment, the intelligent decision engine first calculates the proportion of abnormal data and key data in the total data to determine the compression ratio that needs to be achieved for normal data.

[0157] Specifically, calculate the proportion of abnormal data and key data:

[0158]

[0159] in:

[0160] P abnormal : The proportion of abnormal data to the total data;

[0161] V abnormal : The total amount of abnormal data (unit: bytes);

[0162] V total : The total amount of all data (in bytes).

[0163] in,

[0164] in:

[0165] P critical : The proportion of key data to total data;

[0166] V critical : The total amount of key data (unit: bytes).

[0167] The normal data ratio is: P normal =1-P abnormal -P critical ,

[0168] Determine the compression ratio required for normal data:

[0169] First, calculate the bandwidth required for abnormal data and key data without compression:

[0170] BW abnormal =R abnormal =V abnormal ×f sampling ,

[0171] BW critical =R critical =V critical ×f sampling ,

[0172] in:

[0173] BW abnormal , BW critical : Bandwidth required for abnormal data and critical data;

[0174] R abnomal , R critical : Data rate of abnormal data and key data;

[0175] f sampling : The sampling frequency of the sensor.

[0176] The remaining bandwidth available for normal data transmission is: BW normal =BWa vailable -BW abnormal -BWcritical，

[0177] in:

[0178] BW normal : The bandwidth available for normal data;

[0179] BW available : The total bandwidth available to the system.

[0180] Then, calculate the compression ratio that needs to be achieved for normal data: in:

[0181] CR normal : The compression ratio required for normal data;

[0182] V normal : The total amount of normal data.

[0183] To ensure data transfer complies with bandwidth limits, the following conditions need to be met:

[0184] Secondly, the intelligent decision engine introduces factors such as bit error rate, adjusts the calculation formula of compression ratio, and comprehensively considers the compression ratio ECR. normal The calculation formula is:

[0185] in:

[0186] E normal : The maximum bit error rate allowed for normal data, which usually cannot be higher than 0.15;

[0187] H: The maximum expansion ratio of normal data, H is a natural number greater than or equal to 2.

[0188] Other parameters are the same as described above.

[0189] Then determine the bandwidth EBW corresponding to the compression ratio of comprehensive factors normal For: EBW normal =BW normal ×H (1-Enormal) .

[0190] Optionally, the maximum expansion ratio H of normal data is related to the current sensor data reporting period, which is an integer multiple of a basic reporting period. The basic reporting period is preset, for example, the basic reporting period is equal to 4 times the sensor acquisition period.

[0191] The current sensor data reporting period is user-selectable.

[0192] The maximum expansion ratio H of normal data is selected as 2. When the current sensor data reporting period currently set is greater than the basic reporting period, the calculation formula of the maximum expansion ratio H of normal data is:

[0193]

[0194] Among them, floor function is the rounding function, period is the current collection period, and the basic collection period is period base .

[0195] In one embodiment, in step 5, the intelligent compression module compresses data according to the compression ratio determined by the intelligent decision engine, including:

[0196] The received fused multi-source operation data is compressed. When the intelligent compression module starts to compress, the currently available transmission bandwidth is determined again. If the available bandwidth allows, a compression strategy is formulated, which specifically includes:

[0197] If the current available bandwidth BW is determined <BW normal (W normal is the bandwidth available for normal data determined in step 4), then wait for the detection of available bandwidth BW ≥ BW normal Then transfer.

[0198] If the current available bandwidth BW is determined normal ≤BW<EBW normal The following strategies are adopted:

[0199] Use the calculated CR for normal data normal Compression (data packet 1);

[0200] Losslessly compress abnormal data (data packet 2) and key data (data packet 3);

[0201] Transmit packets 1, 2, 3.

[0202] If the current available bandwidth BW ≥ EBW normal , the following strategies are adopted:

[0203] Losslessly compress abnormal data (data packet 2) and key data (data packet 3);

[0204] The calculated CR is used after appending the abnormal data segment and the key data segment to the normal data packet. normal Compression (data packet 4);

[0205] Transmit data packets 4, 2, 3;

[0206] And, a comparison is performed at the receiving end, and based on the comparison of lossless data, it is determined whether retransmission is required, including:

[0207] At the receiving end, the received data packet 4 is decompressed and compared with the data of data packets 2 and 3. Calculation error: E = |D normal -D important |,

[0208] in:

[0209] E: error matrix;

[0210] D normal : Abnormal data segment and key data segment in decompressed data packet 4;

[0211] D important : Lossless data of abnormal data and key data.

[0212] If the error E exceeds the preset threshold E threshold , then request the sender to retransmit the corresponding data to ensure the accuracy of the data.

[0213] In the case of insufficient bandwidth, no abnormal and key data segments are attached to the normal data packet, and no comparison, calibration and retransmission are performed. Only the normal data packet and the abnormal data and key data of lossless compression are transmitted, so as to avoid further network congestion and aggravate the situation of data retention and uploading. The intelligent decision engine of the present invention, taking into account the proportion of abnormal data and key data, data statistical characteristics and available bandwidth conditions, gives priority to guaranteeing the lossless transmission of abnormal data and key data when the available bandwidth permits, and attaches abnormal data segments and key data segments to the normal data packet, that is, transmits a comprehensive data packet. At this time, by comparing the abnormal data segments and key data segments in the comprehensive data packet and the abnormal data and key data of lossless compression at the receiving end, it is determined whether retransmission is needed to ensure the reliability and accuracy of data transmission. If the bandwidth is insufficient, the abnormal data segments and key data segments are not attached to the normal data segment, only the compressed normal data is transmitted, and the compression ratio and sampling frequency are adjusted according to the specific situation to ensure the operation efficiency of the system and the effectiveness of data transmission.

[0214] In one embodiment, in step four, the adaptive data fusion module processes the power equipment operation data collected by each sensor in real time according to the multi-source data fusion parameters provided by the intelligent decision engine, including dividing the data into four different data packets according to the marking of the data by the intelligent decision engine: a normal data packet, an abnormal data packet, a key data packet, and a comprehensive data packet containing all the data.

[0215] In one embodiment, the adaptive data fusion module receives real-time data from each sensor and the tags of the intelligent decision engine. Based on the tag information, the adaptive data fusion module classifies it. For example, sensor data marked as "normal" and "critical" are classified into normal data packets; sensor data marked as "abnormal" are classified into abnormal data packets; sensor data marked as "critical" are classified into critical data packets. In addition, in order to quickly compress in certain situations when the intelligent compression module performs compression, the adaptive data fusion module also creates a comprehensive data packet containing all sensor data. After classification, the adaptive data fusion module adds necessary metadata to each data packet, such as a timestamp. After completing the data packetization, the adaptive data fusion module sends these data packets to the intelligent compression module for compression processing.

[0216] In one embodiment, in step 5, the intelligent compression module determines the compression ratio CR according to the intelligent decision engine. normal Select an appropriate compression algorithm and normal Determine specific compression parameters and compress the received fused multi-source operation data.

[0217] The intelligent compression module first receives four data packets from the adaptive data fusion module, and formulates corresponding compression strategies for the selected data packets based on the compression ratio provided by the intelligent decision engine and the current available transmission bandwidth.

[0218] For normal data packets, the intelligent compression module selects an appropriate lossy compression algorithm to compress normal data. Among the lossy compression algorithms, commonly used methods include predictive coding, quantization, and wavelet transform.

[0219] Taking quantization compression as an example, specific compression parameters (such as quantization step size) can be determined according to CR normal The basic principle of quantization is to discretize the continuous signal amplitude and reduce the amount of data by reducing the number of bits required to represent each sampling point.

[0220] First, determine the number of quantization bits n:

[0221] Assuming the original data is 16-bit integer data, if the compression ratio CR is required normal =4,

[0222] The number of quantization bits is:

[0223] Next, calculate the quantization step size Δ:

[0224] Among them, x max and x minare the maximum and minimum values ​​of the data set respectively. The quantization step size Δ determines the quantization accuracy. A larger Δ will result in a larger quantization error, but a higher compression ratio; a smaller Δ will have the opposite effect.

[0225] By adjusting the number of quantization bits n and the step size Δ, the intelligent compression module can achieve the required compression ratio CR normal During the compression process, the intelligent compression module quantizes the original data according to the quantization step size, maps continuous values ​​to a limited set of discrete values, and thus reduces the amount of data.

[0226] For abnormal data packets, the intelligent compression module uses a lossless compression algorithm for the abnormal data. Commonly used lossless compression algorithms include Huffman coding, arithmetic coding, etc.

[0227] For key data packets, the intelligent compression module can perform lossless compression on key data.

[0228] If you need to select a comprehensive data packet, do not select a normal data packet. The intelligent compression module selects the same compression algorithm and compression ratio as the normal data packet. You can also select a compression ratio of (ECR normal -TH), where TH is the overflow prevention value set by the user or a preset ratio (ECR normal A certain proportion of), need to meet (ECR normal -TH)>CR normal After completing compression and encapsulation, the intelligent compression module sends the compressed data packet to the transmission module. The transmission module is responsible for data strategy.

[0229] In one embodiment, in step six, the transmission module selects an appropriate transmission protocol (such as TCP, UDP, etc.) and adopts necessary transmission control mechanisms (such as flow control, congestion control, etc.) to reliably and efficiently transmit the compressed fused multi-source operation data to the central processing unit (CPU) of the cloud.

[0230] The present invention adopts a method combining an intelligent decision engine with an adaptive data fusion module, and the present invention realizes efficient fusion and dynamic compression of multi-source data, effectively reduces data redundancy, and improves the efficiency of data transmission and processing. The intelligent decision engine can perform anomaly detection and equipment status assessment based on real-time monitoring data, generate multi-source data fusion parameters and compression strategy parameters suitable for the current environment, and thus guide the adaptive data fusion module to optimize the data processing flow. The fusion results after data compression are efficiently transmitted to the cloud central processing unit through the transmission module, ensuring the real-time and integrity of the data. The overall technical solution not only improves the real-time monitoring capability and fault warning level of the power equipment data acquisition system, but also significantly reduces the cost of data transmission and storage, and improves the overall reliability and scalability of the system.

[0231] The method of the present invention not only improves the real-time monitoring capability and fault warning level of the power equipment data acquisition system, but also significantly reduces the cost of data transmission and storage, and improves the overall reliability and scalability of the system.

[0232] like Figure 2 As shown, the present invention also provides a power equipment test data acquisition system based on multi-source data fusion, using the above-mentioned power equipment test data acquisition method based on multi-source data fusion, the system comprises:

[0233] A multi-source data acquisition module, which is used to collect the operating data of power equipment in real time;

[0234] An intelligent decision engine for evaluating operation data based on abnormal conditions and equipment status to generate multi-source data fusion parameters; and generating compression strategy parameters based on the multi-source data fusion parameters;

[0235] An adaptive data fusion module, which is used to fuse the operation data of the power equipment collected by the sensors in real time based on multi-source data fusion parameters to obtain fused multi-source operation data; an intelligent compression module, which is used to compress the fused multi-source operation data according to compression strategy parameters;

[0236] The transmission module is used to transmit the compressed fused multi-source operation data to the central processing unit.

[0237] It should be noted that the computer-readable medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0238] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0239] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0240] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0241] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware, wherein the name of a unit does not, in some cases, limit the unit itself.

[0242] The above introduces the preferred embodiments of the present invention, which is intended to make the spirit of the present invention clearer and easier to understand, but is not intended to limit the present invention. All modifications, substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection outlined by the claims attached to the present invention.

Claims

1. A method for collecting test data of electric power equipment based on multi-source data fusion, the method comprising: Step 1, each sensor collects the operating data of the power equipment in real time; Step 2: Evaluate the operation data based on abnormal conditions and equipment status to generate multi-source data fusion parameters; Step 3, generating compression strategy parameters based on multi-source data fusion parameters; Step 4, based on the multi-source data fusion parameter, the operation data of the power equipment collected by the sensors in real time are fused to obtain fused multi-source operation data; Step 5, compressing the fused multi-source operation data according to the compression strategy parameters; Step 6: Transmit the compressed fused multi-source operation data to the central processing unit.

2. The method for collecting test data of electric power equipment based on multi-source data fusion according to claim 1, characterized in that: The execution time of step 4 is after step 2 and before step 5; the execution time of step 3 is after step 2 and before step 5; that is, step 2 and step 3 at least partially overlap in execution time.

3. The method for collecting test data of electric power equipment based on multi-source data fusion according to claim 1, characterized in that: In step 2, each sensor collects real-time operating data of the power equipment, including voltage, current, temperature, vibration, and humidity parameters at key locations.

4. The method for collecting test data of electric power equipment based on multi-source data fusion according to claim 3, characterized in that: Evaluate operating data based on abnormal conditions and equipment status to generate multi-source data fusion parameters, including: Use the isolation forest algorithm to detect anomalies in the operating data of each type of sensor; A threshold is set according to the anomaly score, and data points with anomaly scores higher than the threshold are marked as abnormal data of this type of sensor, and the rest are normal data.

5. The method for collecting test data of electric power equipment based on multi-source data fusion according to claim 4, characterized in that: Evaluate operating data based on abnormal conditions and equipment status to generate multi-source data fusion parameters, including: When the proportion of abnormal data of a certain sensor exceeds a certain proportion of the total sampled data of the sensor, all operation data of the sensor are determined to be abnormal data, and all operation data collected by the sensor generating the abnormality are marked as abnormal data; Furthermore, except for the sensor that has an abnormality, the operating data of other sensors are considered normal data.

6. The method for collecting test data of electric power equipment based on multi-source data fusion according to claim 4, characterized in that: Evaluate operating data based on abnormal conditions and equipment status to generate multi-source data fusion parameters, including: Evaluate the equipment status of normal data of each type of sensor, identify potential equipment failure characteristics, and determine key data.

7. The method for collecting test data of electric power equipment based on multi-source data fusion according to claim 6, characterized in that: Evaluate the equipment status of normal data of each type of sensor, identify potential equipment failure characteristics, and determine key data, including: Performing principal component analysis on the operating data of each type of sensor, including constructing a data matrix for each type of sensor, calculating a covariance matrix, and performing eigenvalue decomposition on the covariance matrix to obtain eigenvalues ​​and eigenvectors; Select the first k principal components so that the cumulative contribution rate reaches the set threshold η; Project the data into principal component space; Calculate the reconstruction error for each sensor: in: e i : The reconstruction error of the i-th sensor; x i : the i-th sensor vector; The reconstruction vector of the i-th sensor; Set the error threshold ε, when e i >ε, all the operating data of the corresponding sensors are marked as key data in this type of sensor.

8. The method for collecting test data of electric power equipment based on multi-source data fusion according to claim 1, characterized in that: In the step 3, based on the multi-source data fusion parameters, compression strategy parameters are generated, including determining the compression ratio based on the multi-source data fusion parameters, the proportion of abnormal data and key data, and the available bandwidth.

9. The method for collecting test data of electric power equipment based on multi-source data fusion according to claim 8, characterized in that: Determine a compression algorithm according to the determined compression ratio, and compress the received fused multi-source operation data; when starting to compress, determine the current available bandwidth and determine the compression strategy, including appending abnormal data segments and key data segments after normal data packets if the available bandwidth allows; Furthermore, the receiving end compares the abnormal data segment and the key data segment appended to the normal data packet with the losslessly compressed abnormal data packet and the key data packet to determine whether retransmission is required.

10. A power equipment test data acquisition system based on multi-source data fusion, characterized in that: According to any one of claims 1 to 9, the method for collecting test data of electric power equipment based on multi-source data fusion is adopted, wherein the system comprises: A multi-source data acquisition module, which is used to collect the operating data of power equipment in real time; An intelligent decision engine for evaluating operation data based on abnormal conditions and equipment status to generate multi-source data fusion parameters; and generating compression strategy parameters based on the multi-source data fusion parameters; An adaptive data fusion module, which is used to fuse the operation data of the power equipment collected in real time by the sensors based on multi-source data fusion parameters to obtain fused multi-source operation data; An intelligent compression module, which is used to compress the fused multi-source operation data according to compression strategy parameters; The transmission module is used to transmit the compressed fused multi-source operation data to the central processing unit.

Images